Part I - Concepts, Tools, Processes

Section IA Concepts Introduction

Sorbie Annie

This volume commences with chapters that address key concepts that are embedded in health research regulation (HRR). Many of the concepts considered – autonomy, proportionality, the public interest – may feel familiar and could even be described as ‘regulatory buzzwords’. However, although they may be often used, they can also be elusive and ill defined. Here, the contributing authors explore these contours and provide the reader with the lie of the land in these areas. However, the chapters in this section go beyond this. The authors also challenge the operation of these concepts and provide their own views on how they may be deployed in HRR. This is not a ‘definitions’ section: the concepts explored may be used in different ways by authors in the course of the volume. However, the value of exposing these to scrutiny at the outset is twofold. First, this orientates the reader (whether they be a researcher, regulator, or engaged citizen) in the context of contemporary HRR and sets the scene for these familiar concepts to be approached with a critical eye. Second, this reveals the range of perspectives that may be brought to bear on these shared concepts in HRR, thus promoting productive interdisciplinary engagement and discourse in this and subsequent sections of this volume.

The section opens with Rogers’ chapter on Vulnerability. Rogers positions vulnerability as a widely accepted and relevant concept in HRR, but one that is also challenging to conceptualise and define. She engages with the difficulties that have arisen in conceptualising vulnerability, reconciling universal and special notions of vulnerability, and identifying the distinct duties and obligations that are triggered. In particular, she points to newer analytic approaches that conceptualise vulnerability as relational and dynamic, and identifies multiple potential sources of vulnerability, thus offering a more nuanced way of thinking about protections against the risks of research-related harm and wrongs. Stoljar’s chapter on Autonomy also addresses a pervasive concept in HRR. Stoljar challenges the individualistic conception of autonomy implicit in the current health research ethics guidelines and outlines a contrasting ‘relational’ approach to autonomy. Stoljar posits that the social conditions that a person inhabits, including their available options and interpersonal contexts, all affect the ability to make autonomous decisions. She identifies some of the implications of this for health research ethics, including the relationship between autonomy and vulnerability, and the claim that informed consent is sufficient for accordingly due respect to autonomy.

A similar interconnectedness can be seen between the next two chapters in this section. Schaefer’s chapter considers Proportionality in the context of HRR. He approaches this in terms of a justificatory relationship and underlines that, while a necessary part of responsible HRR, assessing proportionality is no easy task. In particular, it involves the weighing of different values that are non-commensurate and often non-quantifiable. Moving forward, Schaefer suggests several procedural approaches that can help improve the reliability and legitimacy of those assessments. Coleman further addresses risk–benefit analysis later in this volume (Chapter 13). Schaefer’s consideration of how to integrate the social value of research into proportionality is also complemented by van Delden and van der Graaf’s chapter that takes Social Value as its focus. The authors define the social value of an intervention as the value that it could eventually have on the well-being of groups of patients and/or society. However, they note, too, that to state a requirement for social value is one thing; to actually evaluate the social value of a research project is another. Overall, they find that social value has matured from an attractive but elusive idea into something that has to be assessed, evaluated and optimised, and can be used to address some of the justice issues in healthcare.

Three chapters follow that speak to the role of the individual and the collective in contemporary HRR. These begin with Kieslich and Prainsack’s chapter on Solidarity. This builds on existing and emerging research to explore the analytical and normative roles solidarity can play when designing HRR regimes. Kieslich and Prainsack illustrate their argument by reference to the European Union regulatory regime for research on rare diseases and orphan drugs. Through this discussion, the authors show how the concept of solidarity can be used to reframe the regulation of research from a market failure problem, which needs to be addressed through financial incentives, to a societal challenge in which the nature of barriers is not just financial. Next, Sorbie considers the Public Interest in HRR. This chapter provides an introduction to this elusive concept and considers two key ways that the public interest is constructed in HRR, namely as a legal device and through empirical evidence of the views of publics. Both conceptualisations are analysed with reference to the key challenges and opportunities that they present before a holistic concept of the public interest in HRR is proposed and consideration given to how this may be operationalised in practice. Townend’s examination on Privacy completes this trio of chapters. Here, he examines privacy as a well-established concept that has become a mainstay of good practice in research, yet also one that remains nebulous in character. Townend offers an explanation of why privacy is a difficult concept to express, how the law approaches the concept, and how it might be explored as a broader normative concept that can be operationalised by researchers. In the course of their analyses, Sorbie and Townend both consider matters at the intersection of governance and public involvement. Readers who wish to know more about the latter in HRR are directed to Aitken and Cunningham-Burley’s chapter, which specifically examines Forms of Engagement (Chapter 11) and to Burgess’ consideration of Mobilizing Public Expertise in Health Research Regulation (Chapter 25) later in this volume.

The final two chapters of the section consider key concepts specifically in their institutional and political contexts. Kerasidou’s chapter on Trustworthy Institutions in Global Health Research Collaborations delineates the differences between trust and trustworthiness, and argues that institutions committed to advancing the aims of global health should aim to promote fair and trusting collaborations. More specifically, she proposes that being trustworthy requires more than just the observation of rules or the incorporation of moral principles in policies and structures; it also demands attention to the relational aspect of trust. Finally, this section concludes as it started: with consideration of vulnerability. In Brassington’s chapter on Vulnerabilities and Power: The Political Side of Health Research, he argues that there is a political dimension to research, and that accounts of health research regulation that ignore political relations between stakeholders are therefore incomplete. He concludes that research promises us a way to address human vulnerabilities, but it may exacerbate others in the process and that the relationship between researcher and participant can only really be understood when its own inherent political dynamic is acknowledged too.

Each of the chapters in this section illustrate that HRR provides a dynamic area of study where even well-established concepts may be in various ways disputed and unsettled. In examining these dynamics many of the authors also address the relationship between the individual and the collective in HRR. As these and subsequent chapters show, this tension at the heart of HRR is accentuated by the drive towards data-driven and population-level biomedical research. Finally, many authors call for further work to deepen both how these concepts are understood in context and how they are operationalised in the health research endeavour. A response to this can be found in the pages that follow.

1 Vulnerability

Wendy A. Rogers

1.1 Introduction

Vulnerability is widely accepted as a relevant concept in human research regulation. Reflecting this, influential international research ethics guidelines require identification of, and protections for, participants who are deemed vulnerable.¹ Nonetheless, vulnerability is challenging to conceptualise and define, with ongoing disputes about the nature and extent of moral obligations to the vulnerable. This chapter maps the history of vulnerability in human research ethics guidelines and explores current debates regarding the role of vulnerability in guiding ethical deliberations about research participation.

1.2 Vulnerability in Research Ethics Guidelines

Concerns about vulnerability are implied rather than explicitly mentioned in some of the first formal research ethics guidelines such as the Guidelines for Human Experimentation (the Guidelines) issued by the German government in 1931, and the Nuremberg Code (the Code).² These early documents were concerned about experimentation on non-consenting individuals, especially those susceptible to exploitation due to various hardships. Both emphasised the importance of informed consent. The Code required the decision of the potential participant to be fully informed and ‘without the intervention of any element of force, fraud, deceit, duress, over-reaching, or other ulterior form of constraint or coercion’.³ Similarly, the Guidelines prohibited exploiting social hardships to secure research participants, as to do so would be ‘incompatible with the principles of medical ethics’.⁴ Without explicit use of the term vulnerability, these documents pinpointed concerns about exploitation and whether voluntary informed consent could protect participants suffering hardships. In subsequent guidelines, these concerns are conceptualised as indicators of vulnerability.

Vulnerability is first explicitly identified as a characteristic of individuals and groups who thereby require special protections in the 1979 Belmont Report (the Report). The Report intended to provide a comprehensive framework for resolving ethical problems arising from human research.⁵ Its three principles – respect for persons, beneficence and justice – offer protection to all research participants without exception. In addition to these universal protections, the Report identified three areas where participants may be especially vulnerable. The first, echoing the 1931 Guidelines, concerned the voluntariness of consent in situations where ordinarily acceptable inducements may become undue if the subject is especially vulnerable.⁶ The second required increased scrutiny of risks and benefits for research involving vulnerable populations, arguing that their involvement is more or less appropriate depending upon the nature and magnitude of risks, the anticipated benefits and the condition of the population involved.⁷ The third concerned the potential injustice of recruiting participants ‘solely for administrative convenience, or because they are easy to manipulate as a result of their illness or socioeconomic condition’.⁸ Here the Report referred to ‘racial minorities, the economically disadvantaged, the very sick, and the institutionalized’ whose ready availability may lead to their exploitation.

Thus the Report characterised the vulnerable as individuals and groups with potentially limited capacity to give consent and/or those liable to exploitation for various reasons. It required greater justifications for the inclusion of vulnerable participants, and identified exclusion altogether from research as one way of protecting the vulnerable.

Subsequent research ethics guidelines follow the Report in linking vulnerability to consent, exploitation and special protections. For example, the Declaration on Bioethics and Human Rights stipulates the following:

In applying and advancing scientific knowledge, medical practice and associated technologies, human vulnerability should be taken into account. Individuals and groups of special vulnerability should be protected and the personal integrity of such individuals respected.⁹

The Declaration of Helsinki requires ‘specifically considered protection’ for all vulnerable individuals and groups.¹⁰ Likewise, the current Council for International Organizations of Medical Sciences (CIOMS) guidelines invoke ‘specific protections’:

When vulnerable individuals and groups are considered for recruitment in research, researchers and research ethics committees must ensure that specific protections are in place to safeguard the rights and welfare of these individuals and groups in the conduct of the research.¹¹

This brief survey of research ethics guidance demonstrates that for nearly ninety years there have been concerns that some research participants are more vulnerable than others, and that vulnerable participants require special considerations. Despite the agreement that vulnerable participants require something more than routine ethical consideration, there is little consensus as to what characteristics make some participants more vulnerable than others and whether ‘special consideration’ is the appropriate moral response to vulnerability.

1.3 Vulnerability: An Ambiguous Concept in HRR

There is an ambiguity about vulnerability running through the heart of research ethics. Prima facie, research ethics guidance provides protections for all participants who are potentially exposed to research-related harms such as deception, coercion, injury, misuse of their data and other harms. Requirements such as informed consent and balancing benefits and risks aim to mitigate this vulnerability. Yet this universal vulnerability to the potential harms of research is not explicitly named. Instead, the category of ‘special’ vulnerability attributes vulnerability to groups or individuals such as those identified in the Belmont Report: ‘racial minorities, the economically disadvantaged, the very sick, and the institutionalized’.¹² This ‘special’ vulnerability approach reflects ambivalence about overtly acknowledging universal vulnerability and simultaneous recognition that some research participants do bear greater risk of harms than others. Ambivalence about the concept of vulnerability can be traced back to competing philosophical accounts.

The universal account takes vulnerability to be a type of fragility or susceptibility to suffering, linked to human embodiment. According to Fineman, vulnerability is an ontological necessity of our humanity, an ‘inevitable, enduring aspect of the human condition’.¹³ Understanding vulnerability as a universal feature reflects the shared human capacity for experiencing pain, frailty and other harms of existence and the inevitability of death for all humans. The notion of universal vulnerability underpins ethical concern for all research participants. The requirement for informed consent arises because all participants are potentially vulnerable to deception as they lack relevant skills to distinguish experimentation from accepted treatment. Similarly, requirements for pain relief in pertinent protocols reflect universal vulnerability to suffering pain. But universal vulnerability is not explicitly identified in research ethics guidelines. Instead, there is an assumed ‘normal’ research participant for whom standard ethical protections are adequate. This baseline normal research participant is characterised by Luna as ‘mature, moderately well-educated, clear thinking, literate, [and] self-supporting’.¹⁴ Vulnerable participants are identified against this implicit norm.

In contrast to universal vulnerability, accounts of special vulnerability claim that vulnerability is essentially specific and relational: individuals are vulnerable to particular agents regarding particular threats to their interests.¹⁵ Although everyone has interests that may be threatened, some individuals or groups have little or no capacity to protect themselves. On this account, vulnerable persons have a reduced capacity to safeguard their interests relative to others. Whereas the universal account stresses our common embodied humanity and equal susceptibility to suffering, the special vulnerability account emphasises the ways in which various inequalities make some individuals (‘the vulnerable’) especially susceptible to harm or exploitation by others.¹⁶

Within research ethics, the notion of special vulnerability is dominant. However, despite widespread requirements for special protections for the vulnerable, vulnerability is rarely defined in research ethics guidance. Bracken-Roche and others found that only three of eleven national and international guidelines contained definitions of vulnerability.¹⁷ Both CIOMS and the Canadian Tri-Council Policy Statement define vulnerability in terms of decreased ability to protect one’s own interests, secondary to intra-personal factors (e.g. reduced capacity to give informed consent) or contextual factors (e.g. limited access to social goods including rights, opportunities and power). The third guideline, the International Conference on Harmonisation – Good Clinical Practice guideline, does not define vulnerability per se, but defines vulnerable subjects in a glossary entry as those who may be unduly influenced either by expectations of benefit, or due to their subordinate place in a relevant hierarchy.¹⁸

The common approach in these and other guidelines is to rely upon lists of individuals or groups with characteristics that are taken to be indicators of vulnerability. The eleven guidelines analysed by Bracken-Roche and others list thirty-two characteristics signifying vulnerability. Individual indicators include being a prisoner, homeless person, woman, economically disadvantaged person, person lacking in political or social power, refugee, neonate and so forth. Identified vulnerable groups include very sick persons, children, minors or young persons, pregnant or breastfeeding women, the elderly, persons with mental illnesses, persons with limited capacity to consent and others. This dominant approach of labelling vulnerable participants has been subject to various critiques.

1.4 Problems with the Labelling Approach to Vulnerability in Research

The aim of labelling individuals or groups as vulnerable is to trigger special protections over and above those offered to all research participants, to prevent or decrease the risk of harms triggered by the vulnerability in question. Nevertheless, this approach is problematic as it leads to stereotyping, discrimination and unwarranted exclusion from research. Critics have argued that the labelling approach is both too narrow and too broad.¹⁹

One effect of the labelling approach is to focus narrowly on questions about capacity, and whether or not vulnerable individuals are able to give competent, informed consent for research participation.²⁰ This results in ethical review that attempts to identify all possible factors that might render consent less than fully valid, such as cognitive impairment or coercive circumstances. For individuals who lack capacity to provide valid consent, the proposed remedies for this vulnerability are proxy consent or exclusion from research. However, conceptualising vulnerability primarily in terms of incapacity to provide informed consent is ethically inadequate.

First, this approach fails to address the full range of moral issues raised by vulnerability, such as susceptibility to exploitation.²¹ Some research participants who are capable of providing informed consent may, nonetheless, be vulnerable to exploitation, due to poverty, power imbalances in the researcher–participant relationship or other circumstances. For example, pregnant women may be vulnerable to exploitation regarding participation in research aimed at benefitting the fetus. In this situation, pro-natalist pressures may unduly exacerbate women’s understandable concerns for foetal well-being, thereby creating opportunities for exploitation.²² Attention to informed consent processes will not necessarily resolve this kind of exploitation. Nor will it protect those who are vulnerable due to a lack of basic human rights.²³ For example, better information about a research protocol fails to mitigate participants’ vulnerability to coercion from local power brokers who stand to profit from the research.

Second, the narrow approach ‘can divert attention from features of the research itself, the institutional environment, or the social and economic context that can put participants in harm’s way’.²⁴ Informed consent does not provide protections against dangerous protocols, researchers with conflicts of interest or dysfunctional institutions, all of which make participants vulnerable by increasing their risk of harm. External factors such as these contributed to the deaths of Ellen Roche, a healthy volunteer who died in an asthma research trial,²⁵ and Dan Markingham, who died in an anti-psychotic medication trial.²⁶ These participants were made vulnerable by deficiencies in the manufacturing standards of the inhalant, poor standards in the research protocol review process by the institutional review board, inadequate oversight, and significant conflicts of interest at researcher and institutional levels. These factors affect the safety of all participants rather than reflecting specific vulnerabilities of these particular participants.

A third problem with a narrow focus arising from the labelling approach is that concentrating on informed consent offers few options for mitigating vulnerability. For individuals who lack capacity – such as unconscious persons, babies, young children or individuals with severe cognitive deficits – it may not be possible to develop valid consent processes, leaving the alternatives of proxy consent or exclusion from research. Exclusion from research results in inadequate information about therapeutics for affected individuals and groups.²⁷ Rather than exposing a small number of individuals to specific risks within a regulated clinical trial, exclusion from research increases risks for all members of excluded groups who then must rely on off-label prescribing of therapies with unknown effects for their patient cohort.²⁸

Despite the labelling approach to vulnerability being criticised for being too narrow, a second critique claims the opposite: that an over-inclusive approach to identifying vulnerability leads to virtually everyone being labelled as vulnerable.²⁹ Lists compiled from research ethics guidelines by Bracken-Roche and others and Hurst leave few who are not classified as vulnerable.³⁰ This apparently over-inclusive approach to vulnerability labelling renders the notion of vulnerability ineffectual for two reasons. First, despite its breadth, there is no recognition of the features that might underpin a universal conception of vulnerability; the focus remains on special vulnerability. Second, the over-inclusiveness of the labelling limits the utility of invoking protections for special vulnerability, because the context-specific needs of individuals or groups are obscured rather than identified. Vulnerability cannot be a useful marker for providing extra protections if all research participants are deemed vulnerable. The concept becomes so broad as to be meaningless, and certainly impractical for mandating specific responses.³¹

Critics of the over-inclusive labelling of vulnerability note that this approach can lead to stereotyping, discrimination and failure to consider the specifics of each case. Stereotyping occurs when whole categories of individuals are labelled vulnerable, in contrast to the rest of the – presumably invulnerable – population. Labelling has a homogenising effect as all members of the group are assumed to be equally vulnerable. But it is unreasonable both to divide the general population into exclusive vulnerable and non-vulnerable categories, and to obliterate relevant differences between those labelled vulnerable. The impacts on individuals of factors associated with special vulnerability such as educational disadvantage, cognitive impairment or dependent relationships vary enormously. For example, young people under sixteen years are usually labelled vulnerable as they are held incapable of giving valid informed consent. However, the capacity to understand complex information and make considered decisions is highly variable in adolescents; some are capable of consenting and some are not. In addition, vulnerability is not a dichotomous state such that individuals are either vulnerable or not. Rather, vulnerability occurs on a spectrum, with different levels or degrees. But this distinction is lost when labels are applied: regarding vulnerability, ‘you are either in or you are out’.³²

Once individuals or groups are labelled vulnerable, this can be a source of discrimination, used to justify unwarranted and unjust paternalistic policies.³³ Such discrimination is evident regarding pregnant women, who have historically been labelled vulnerable and excluded from research notwithstanding their evident capacity to give informed consent. Routine exclusion of pregnant women from research undermines their autonomy by removing the opportunity to make decisions about research participation. Further, exclusion of pregnant women from research is harmful as it results in a lack of information about safe and effective treatments in pregnancy, especially when pregnant women who are given untested treatments mistakenly believe they are receiving a therapeutic intervention.³⁴ Exclusion from research is especially problematic where access to treatment is premised on evidence of the safety and efficacy of that treatment. Where groups are systematically excluded from research, the evidence base is correspondingly meagre and their treatment options limited.³⁵

Given these problems with the labelling approach to vulnerability in research ethics, I now turn to analytic approaches to conceptualising vulnerability.

1.5 Analytic Approaches to Defining Vulnerability

Analytic approaches to vulnerability seek to explain the concept in ways that foster understanding of what vulnerability is, and what moral responses are owed to the vulnerable. One analytic approach to conceptualising vulnerability identifies characteristics that serve as criteria for vulnerability,³⁶ while a second examines what is owed to the vulnerable.³⁷ More recently, Luna has proposed the metaphor of layers to explain multiple forms of vulnerability,³⁸ while Lange and others focus on sources of vulnerability in their taxonomy.³⁹

Kipnis takes vulnerability to be ‘a certain precariousness’ that leaves the individual open to being harmed or taken advantage of by researchers. Rather than labelling, he asks researchers to consider participants’ circumstances as sources of vulnerability. His taxonomy focuses on circumstances – medical, cognitive, deferential, juridic, allocational, infrastructural and social – that threaten the validity of consent.⁴⁰ Similarly, Rogers and Ballantyne identify extrinsic and intrinsic – to the individual – sources of vulnerability that render participants unable to safeguard their own interests.⁴¹ On their account, extrinsic vulnerability arises from power inequalities in the researcher–participant relationship.

Nickel argues that in addition to consent considerations, vulnerability is a justification for special protections in research for two fairness-related reasons. The first concerns the unfair burden of research participation imposed on disadvantaged or dependent groups who lack the power to refuse participation. The second concerns the unfair distribution of research benefits, especially those arising from exclusion from research. Consent-based and justice-based reasons for special protections can be mutually reinforcing, as, for example, members of dependent groups may be targeted for inclusion in research because they have limited ability to refuse participation, but the research may not address problems relevant to the groups to which they belong. In considering what is owed to those who are especially vulnerable on these grounds, Nickel appeals to the principle of equal respect, asking researchers to engage in empathic consideration of the circumstances of the vulnerable to better understand their viewpoint.⁴²

Luna proposes the metaphor of ‘layers of vulnerability as an alternative to labelling. She argues that this relational and dynamic conception of vulnerability avoids the stereotyping and essentialism of the labelling approach. Her account is relational in that it identifies each layer of an individual’s vulnerability by closely examining the context in which she is situated, and dynamic because it recognises that layers of vulnerability may come and go as the context changes. Since Luna’s account assumes that vulnerability is dynamic and inessential, it does not stereotype or stigmatise individuals when describing them as vulnerable, while the complexity of a person’s situation can be recognised through considering multiple layers of vulnerability.⁴³ More recently, Luna specifies that layers of vulnerability may be related to ‘physical problems, consent, dependency, exploitation, [and] socioeconomic situations’, and that these layers should be understood as dispositions for harm or exploitation. Finally, she introduces the notion of cascade vulnerabilities that can trigger a series of events with harmful consequences.⁴⁴

Lange and others propose a taxonomy of vulnerability.⁴⁵ Their taxonomy aims to reconcile universal and particular conceptions of vulnerability by postulating inherent and situational sources of vulnerability.⁴⁶ Inherent sources include human corporeality, our affective and social natures and neediness, and our dependence on others. These sources of vulnerability are ineliminable features of the human condition. Inherent sources of vulnerability produce variable risk of harm or wrongs depending on age, health, gender and disability, as well as individuals’ capacities for resilience and the presence of social supports. Situational sources of vulnerability are context specific and include personal, social, political, economic or environmental features affecting individuals or social groups. As they are context specific, situational sources of vulnerability may be persistent or fluctuate over time. Situational vulnerability has a sub-category of pathogenic vulnerability. Pathogenic vulnerability refers to vulnerability arising from dysfunctional relationships characterised by prejudice, abuse, neglect or disrespect, or from political situations characterised by injustice, persecution or political violence. In addition, pathogenic vulnerabilities arise when policies designed to protect against existing vulnerabilities have the perverse effect of exacerbating existing, or generating new, vulnerabilities. On this account, the exclusion of pregnant women from research creates the pathogenic vulnerability experienced by ill pregnant women for whose conditions there is inadequate evidence about safe and effective treatments. These three types of vulnerability may co-exist, overlap and be occurrent (immediate and present) or dispositional (latent or background).

Like Kipnis’s approach, this taxonomy provides a systematic way to identify existing and potential sources of vulnerability in order to put in place mitigating strategies. By identifying different sources of vulnerability, researchers are required to attend to the wider context of the research as well as to the characteristics of participants. Vulnerability considerations are not limited to consent, exploitation, or unduly circumscribed safety assessments, but must take account of a full range of harms that research participation may involve. As well as protecting against harms, Lange and others postulate a positive duty to foster participants’ autonomy that is more onerous than seeking informed consent or avoiding unjust paternalism.⁴⁷ On their account, the duty to respect autonomy requires engaging with and furthering the interests of participants such that research participation actively supports autonomy and promotes resilience. Finally, the notion of pathogenic vulnerability grounds a duty to ensure that well-intended extra protections do not exacerbate existing or create new vulnerabilities.

1.6 What Work Can the Concept of Vulnerability Do?

To be useful in research ethics, the concept of vulnerability should draw attention to a set of concerns that are distinct from other moral concerns, and which are identifiable in non-stigmatising or stereotyping ways. The first step is to settle on a definition of vulnerability that encompasses both the universal and special conceptualisations, and is not implicitly comparative regarding some unstated norm of invulnerability. That is, we need a definition that acknowledges a universal capacity to be harmed or wronged in various ways related to our embodiment, shared needs and relationality, but that also recognises that the likelihood of those harms and wrongs does not fall equally on all individuals due to varying capacities and circumstances. Based on this understanding, vulnerability in research can be thought of as a disposition to, or risk of, suffering harms or wrongs arising from a range of inherent and situational factors. As these factors will independently vary across individuals and groups and change over time, vulnerability ascriptions must be specific, contextualised and reviewed.

The question remains as to whether vulnerability raises its own moral concerns, or is merely a marker for existing harms or wrongs. Hurst adopts the latter view, defining vulnerable research participants as those who are especially likely to incur an already defined research-related wrong. On her account, vulnerability does not generate new moral obligations, but rather serves as a heuristic to draw attention to existing obligations. The point of drawing attention to vulnerability is to flag that special steps may be necessary to fulfil existing research obligations for those who are especially vulnerable.⁴⁸ However, this does not seem to be a very satisfactory approach as it effectively renders the concept of vulnerability redundant. Despite the difficulty of defining vulnerability, the concept is valuable in highlighting morally salient features of our humanity that are central to everyday practices and notions of obligation.⁴⁹ The concept of vulnerability triggers us to think empathetically and humanely about others in a holistic way; to consider their situation, their strengths and weaknesses, and their liability to harm. These are important moral considerations that can readily be obscured by a procedural focus on informed consent or balancing research benefits and burdens.

In the context of research, researchers have a duty to consider the vulnerability of participants in a systematic and comprehensive way. This includes not only identifying specific risks of harm that may arise from experimental interventions, but also investigating the interactions and potentially cumulative effects of different sources or layers of vulnerability. In this process, researchers should be aware of the dangers of stereotyping and discrimination and actively seek to avoid labelling groups or individuals in essentialist ways. In addition, there is a need to consider the potential for protections in research to have counter-intuitive effects and create their own pathogenic vulnerabilities. Often, these may be more visible to participants than researchers, creating an obligation for meaningful consultation with participants and their communities. As being vulnerable implies a lack of power in some regard, this creates a duty for researchers to foster and support autonomy to the extent possible. This is more onerous than respecting autonomous decisions, as it requires investigation on the part of researchers and more comprehensive actions in terms of capacity building.

1.7 Conclusion

The recognition that vulnerable research participants should be protected is longstanding. However, difficulties have arisen in conceptualising vulnerability, reconciling universal and special notions of vulnerability, and identifying distinct duties and obligations triggered by vulnerability. The dominant approach of labelling vulnerable participants and groups is subject to increasing critique, because it can lead to stereotyping, discrimination and exclusion, and fail to be action-guiding. Newer analytic approaches conceptualise vulnerability as relational and dynamic, and identify multiple potential sources of vulnerability. These approaches offer a more nuanced way of thinking about vulnerability and protections against the risks of research-related harm and wrongs. Further work is needed to bridge the gap between these newer conceptualisations of vulnerability and practical guidance for research.

2 Autonomy Relational Conceptions

Natalie Stoljar

2.1 Introduction

This chapter argues that the conception of autonomy implicit in the current health research ethics guidelines is an individualistic one. On this model, people are autonomous when they voluntarily exercise their deliberative capacities to make choices based on their values. They can fail to be autonomous either for psychological reasons, when their deliberative capacities are impaired; or for epistemic reasons, when they do not have access to all the information that is relevant to a decision. The chapter outlines a contrasting ‘relational’ approach to autonomy. On the relational model, people can fail to be autonomous for social reasons. The social conditions that a person inhabits, including their available options, and the interpersonal context – such as family relationships and social expectations – all affect the ability to make autonomous decisions. The chapter also identifies some of the implications of the relational model for health research ethics. It examines two issues implicit in the guidelines: the relationship between autonomy and vulnerability, and the claim that informed consent is sufficient for autonomy.

2.2 Autonomy in Health Research Ethics Guidelines

Autonomy is a one of the fundamental ethical principles identified in documents that provide ethical guidelines for health research using human subjects. Section 9 of the Declaration of Helsinki states that ‘[i]t is the duty of physicians who are involved in medical research to protect the life, health, dignity, integrity, right to self-determination, privacy, and confidentiality of personal information of research subjects’.¹ The Belmont Report outlines three fundamental ethical principles: respect for persons, beneficence and justice. Respect for persons requires that ‘individuals should be treated as autonomous agents’ and that ‘[a]n autonomous person is an individual capable of deliberation about personal goals and of acting under the direction of such deliberation’.² These documents propose explicitly or implicitly that the principle of respect for autonomy can by satisfied by obtaining informed consent. Section 25 of the Declaration of Helsinki states that ‘no individual capable of giving informed consent may be enrolled in a research study unless he or she freely agrees’.³ The Belmont Report provides that respect for persons requires that ‘subjects be given the opportunity to choose what shall or shall not happen to them. This opportunity is provided when adequate standards for informed consent are satisfied’.⁴ And the Government of Canada guidelines state that ‘[i]ndividuals are generally presumed to have the capacity and right to make free and informed decisions. Respect for persons thus means respecting the exercise of individual consent’.⁵

As has been often noted, these ethical codes were initially formulated in response to notorious examples of exploitation and autonomy-denial of vulnerable people in medical research, such as the Nazi medical experiments and the Tuskegee syphilis study of the US Public Health Service. The latter took place from 1932 to 1972 to ‘record the natural history of syphilis’. Researchers recruited 600 Black men – sharecroppers who had not received medical care before – of whom 399 had syphilis and 201 did not. The men consented to participate in exchange for free medical care but were never informed about the true purpose of the study – they were told they were receiving medical treatment for ‘bad blood’. They did not receive penicillin when it became available in 1947, nor were they offered the opportunity to leave the study.⁶ The ethical guidelines aim to protect such vulnerable populations. The Belmont Report states that ‘[c]ertain groups, such as racial minorities, the economically disadvantaged, the very sick, and the institutionalized may continually be sought as research subjects, owing to their ready availability in settings where research is conducted. Given their dependent status and their frequently compromised capacity for free consent, they should be protected against the danger of being involved in research’.⁷ The document also associates vulnerability with a lack of autonomy: ‘Respect for human dignity entails high ethical obligations towards vulnerable persons – to those whose diminished competence and/or decision-making capacity make them vulnerable’.⁸

The ethical guidelines therefore raise three important questions. First, what is the understanding of autonomy implicit in the guidelines? Second, is the ‘autonomy of the vulnerable’⁹ possible, and, if so, how does vulnerability intersect with autonomy? Third, what is the connection between informed consent and autonomy? The next section claims that the conception of autonomy implicit in the guidelines is an individualistic one on which it is sufficient for autonomous decision-making that people have deliberative capacities and exercise them in appropriate ways. However, the individualistic model is limited both in its conception of autonomy and its recommendations regarding the ethical obligations of health researchers. I therefore outline a contrasting ‘relational’ model of autonomy that has been developed over the last several decades by feminist philosophers and others.¹⁰ The final sections examine the implications of relational autonomy for the notions of vulnerability and informed consent. The relational model suggests that the ethical obligations of health researchers to respect autonomy are more demanding than the requirement to secure informed consent.

2.3 The Individualistic Conception of Autonomy

Autonomy is the ability of people to ‘fashion’ their own destiny through ‘successive decisions throughout their lives’.¹¹ The challenge is to identify the conditions that are required to secure this ability and to promote its exercise in particular cases. The philosophical debate about autonomy offers both individualistic and a variety of relational answers. On a standard individualistic view, the conditions of autonomy are purely internal or psychological: autonomy corresponds to a capacity for rational deliberation. Gerald Dworkin argues that a person is globally autonomous when they have a higher-order capacity for critical reflection. A particular decision is locally autonomous when the capacity is exercised to evaluate, endorse or repudiate first-order preferences.¹² This view assumes that most people are rationally competent agents who engage in critical reflection and generate endorsed or ‘authentic’ decisions.¹³ People are nonautonomous only when critical reflection is defective – when the capacity for rational deliberation is undermined by a psychological impairment such as being under the influence of drugs or alcohol, hypnosis, brainwashing or similar forms of psychological manipulation, or pathological psychological conditions, like delusions, that interfere with mental competency. The individualistic conception of autonomy has been highly influential and has led many later authors to adopt variants of Dworkin’s approach.¹⁴

The health research ethics guidelines do not explicitly elaborate a notion of autonomy. The Declaration of Helsinki refers to respecting a ‘right to self-determination’.¹⁵ The Government of Canada Guidelines state that ‘individuals are generally presumed to have the capacity and right to make free and informed decisions’.¹⁶ Only the Belmont Report adopts the individualistic model explicitly, stating that ‘[a]n autonomous person is an individual capable of deliberation about personal goals and of acting under the direction of such deliberation. To respect autonomy is to give weight to autonomous persons’ considered opinions …’¹⁷ However, all the guidelines refer to informed consent as the ethical requirement that will secure subjects’ ‘free choice’ or autonomy. This implies that, for individuals who are ‘capable of deliberation’, the only additional thing needed for autonomy are optimal epistemic conditions in which to deliberate. It is reasonable to assume, therefore, that the health research guidelines do not go beyond the individualistic model, and that the conditions that undermine autonomy are limited to the psychological (impaired cognitive capacities) or the epistemic (inadequate information). On this model, ethical obligations are also limited to ensuring that subjects are mentally competent and have all relevant information. However, this narrow account of ethical obligation does not fully explain the ethical failures of the research in cases such as the Tuskegee study. There is no reason to think that the Tuskegee subjects were mentally incompetent or lacked autonomy due to cognitive impairment. The individualistic model – and the guidelines – imply that these subjects could, in principle, have made an informed decision to participate in the study and therefore that the obligation to respect their autonomy could have been discharged. This model fails to address the social context of choice and is both an implausible characterisation of autonomy and an impoverished account of the researchers’ ethical obligations. I now turn to the relational model of autonomy to address these problems.

2.4 Relational Autonomy

The starting point of relational approaches to autonomy is the observation that people are not disembodied rational subjects who, absent psychological impediments, function as fully autonomous agents. Rather, autonomy is an ability that develops and is exercised in a social context. Social relationships, as well as social and historical conditions – such as gender socialisation – foster or impede autonomy.¹⁸ In short, the social conditions that a person inhabits, the interpersonal context of family and community relationships, as well as norms and interpersonal expectations, can all promote or undermine the ability to make autonomous decisions.

Relational autonomy is a popular framework in clinical ethics¹⁹ but so far has received little attention in health research ethics. One exception is offered by Edward S. Dove et al., who advocate a relational approach to understanding consent to placental sampling in pregnant women.²⁰ Some women request to consult their partners before providing consent to placental sampling. The authors propose that such requests are better accommodated within a relational model than an individualistic model because the former acknowledges the impact of social relationships on agency and hence recognises that ‘joint decision-making [could] be built into the consent process’.²¹ This illustration of an application of the relational model is quite limited, however. First, it seems to conflate joint decision-making with relational autonomy. Joint decision-making implies that the author of a decision is a collective (the woman and her partner), whereas the relational model aims to characterise individual autonomy that takes into account the decision-maker’s social context. Second, the potential of the relational model, both to identify social conditions that hamper individual autonomy, as well as to identify strategies to promote autonomy, is considerably richer than suggested here. There is a complex set of factors elaborated by relational theories that potentially affect subjects’ autonomy and hence should be considered in the context of health research ethics. In what follows, I sketch three varieties of relational approach.

The first set of views argues that although autonomy is primarily a psychological capacity, social conditions play a significant causal role in its development. Marilyn Friedman offers a psychological test of autonomy on which being autonomy requires ‘self-reflective affirmation’ with respect to one’s preferences and values.²² She emphasises, however, that family relationships serve to either promote or hamper the development of the critical reflection that allows people to engage in self-reflective affirmation.²³ Similarly, Diana Tietjens Meyers argues that a person has ‘autonomy competency’ when she exercises a ‘repertoire of skills’ of critical reflection to engage in ‘self-discovery, self-definition and self-direction’.²⁴ Social conditions, particularly lack of education or oppressive socialisation, can affect whether or not such skills develop in the first place, erode them even if they do develop or fail to promote a person’s exercise of these skills. For instance, gendered expectations might deter the development or exercise of autonomy skills in girls but promote it in boys. Boys are more likely to be encouraged to debate their points of view and adopt an attitude of ‘authority over their own voice’. For Meyers, well-honed autonomy competency can also counter oppressive socialisation. She comments that autonomy skills ‘endow women with the capacity to fashion self-portraits and self-narratives in their own voices and to lead their own lives’.²⁵ Meyers’s account is relational because it recognises the many ways in which deliberation is itself moulded by socialisation. She also identifies a rich array of deliberative skills that are required for autonomy competency including those of introspection, imagination, communication and memory.²⁶ These include emotional, bodily and interpersonal components and hence, on Meyers’s approach, relationality is built into critical reflection itself.

A second category of relational views claims that the presence of certain external options is a necessary condition of autonomy. Joseph Raz argues that autonomy – and hence a flourishing life – will be thwarted if a person lacks a sufficient range of meaningful external options.²⁷ For instance, a ‘hounded woman’ on a desert island, who is pursued by a carnivorous wild animal that is hunting her down – and whose every move is determined by the need to protect herself from the beast – cannot be autonomous, even if her powers of critical reflection are fully intact. Marina Oshana also advocates an externalist approach, on which autonomy is incompatible with external conditions that interfere with the ‘de facto power and authority’ that is necessary for people to direct their own lives.²⁸ For example, when young African American men are routinely subjected to police profiling or arbitrary stop-and-search policies, this is autonomy-undermining because it unjustifiably constrains their options and diminishes their power to direct the course of their daily lives. Therefore, on externalist views, autonomy is not only a matter of genuine self-affirmation or of exercising skills of critical reflection; it also requires meaningful external options and the absence of severely constraining external conditions.

Finally, there is a category of relational approaches that focuses on the importance for autonomy of self-regarding attitudes, such as self-trust and self-esteem. Carolyn McLeod identifies the role of self-trust – an ‘attitude of optimism about our own competence and moral integrity’²⁹ – in autonomy through the example of ‘Anna’, who suffered a miscarriage at six weeks’ gestation and afterwards felt considerable emotional turmoil. McLeod analyses Anna’s sense of incompetence to articulate her emotions as in part a result of others’ lack of sympathy for her grief and corresponding failure to reinforce her feelings: ‘[O]ften women and their partners are pressured not to grieve after miscarriage because people tend not to view the fetus’s death as an event that warrants grief’.³⁰ Comments such as ‘it was a blessing in disguise’ or ‘it could have been worse; you could have lost a baby’ fail to ‘give uptake to [women’s] feelings’.³¹ Self-regarding attitudes – people’s sense of their own competence, sense of self-worth and self-trust – are thus particularly sensitive to interpersonal conditions and the attitudes of others.

Catriona Mackenzie has extended this proposal, arguing that self-regarding attitudes such as self-trust are constitutive components of what she calls the ‘normative authority’ required for autonomy.³² Mackenzie elaborates contrasting examples from the clinical context of decisions to withdraw life-preserving medical treatment. Ms B. was a forty-three-year-old single woman who had had a career in social work in a hospital before her illness. Ms B. suffered from a spinal condition that left her paralyzed from the neck down. After an attempt to improve the condition through surgery had failed, she requested that her ventilator be switched off. She was fully aware of all treatment options and thoughtful and articulate about her decision to have treatment withdrawn.³³ Mackenzie claims that she ‘clearly regarded herself as a moral equal’ and that ‘she was motivated by an attitude of self-worth; that is, by a sense that her life should be meaningful, worthwhile, and valuable, and by a conviction that it could no longer continue to be so given her medical condition’.³⁴ Contrast Mrs H., who suffered an aggressive form of bone cancer that required amputation of a leg below the knee. Her husband left her as a result of her disability. Mackenzie comments that ‘she does not have the kind of strongly developed sense of basic self-confidence that might enable her to envisage a different future for herself than the one she had anticipated; and she does not have a sense of her life as worth living in its own right, independently of her relation to her husband’.³⁵ By hypothesis, Mrs H.’s capacity for self-trust and self-esteem was damaged by both her socialisation – Mrs H. was influenced by cultural norms and gender stereotypes that seemed to suppress her sense of self-assurance – as well as gendered relationships within her family – her husband left her when she was no longer able to occupy the traditional role of a supportive wife. Mackenzie argues that self-regarding attitudes are not just the outcomes of the social context; they are also ‘intrinsically relational’ because they are ‘enmeshed in interpersonal relationships and social structures of mutual recognition and it is for this reason that our autonomy can be impaired by failures of recognition’.³⁶

The different relational conceptions of autonomy that I have just outlined convey the complexity of the social–relational factors that potentially affect whether subjects of health research will be autonomous or not. In the next two sections, I outline the implications of the relational model for vulnerability and informed consent. I suggest that the relational model generates more demanding ethical obligations on health researchers than those derived from the individualistic model.

2.5 Relational Autonomy and Vulnerability

Relational conceptions of autonomy help to illuminate the connection between autonomy and vulnerability (see Rogers, Chapter 1 of this volume). Recall that the Belmont Report says that vulnerable persons are ‘those whose diminished competence and/or decision-making capacity make them vulnerable’.³⁷ Vulnerability is here being characterised as (individualistic) non-autonomy, namely as non-autonomy due to (psychologically) diminished competence. This suggests that vulnerability and autonomy are incompatible and that therefore that it is unethical for vulnerable agents to be the subjects of healthcare research. However, rather than treating vulnerability and autonomy as always opposed, the relational model potentially provides a more nuanced account of how to evaluate and ultimately promote the autonomy of members of vulnerable groups.

Theorists of vulnerability have observed that vulnerability can be inherent or situational.³⁸ Certain vulnerable groups, such as pregnant women, are vulnerable due to intrinsic, bodily aspects of being human. Others, such as the subjects of the Tuskegee study, are situationally vulnerable. As impoverished Black sharecroppers in the American South who had never received adequate education or medical care, these subjects were situationally vulnerable. Poverty and racist discrimination afforded them few real options and disposed them to manipulation by the medical personnel conducting the study. The situational vulnerability of subjects of the Tuskegee study generates global non-autonomy. Even if such subjects have the deliberative competence to make informed, locally autonomous decisions, they lacked minimally adequate options, and could not direct the course of their lives. Is health research using subjects who are situationally vulnerable and globally non-autonomous always ethically indefensible? Mackenzie disputes this conclusion. Vulnerability and autonomy are not always opposed, because ‘ethical responses to vulnerability should be guided by the value of autonomy … to counter the sense of powerlessness and loss of agency that is often associated with vulnerability … and to counter the risks of objectionable paternalism’.³⁹ If she is right, it would not be impossible to conduct ethical studies using subjects who are situationally vulnerable, but the obligation to respect autonomy would be demanding in such cases. Researchers would have to actively ameliorate subjects’ autonomy by, e.g. ensuring that participation in the research itself provides meaningful options and fosters positive communication to promote subjects’ self-trust and sense of self-worth.

The situation is different for groups that are defined as vulnerable due to their inherent characteristics, such as pregnant women. It has been pointed out that the assumption that there is homogeneity among members of vulnerable groups is problematic (see Brassington, Chapter 9 and Rogers, Chapter 1 in this volume). Labelling a group a ‘vulnerable population’ can lead to ‘unfettered protectionism’ with respect to health research on this population, which in turn can have adverse consequences for their health.⁴⁰ For instance, there is no effective malaria drug that can be used by pregnant women, thus leading to unnecessarily high rates of death from the disease.⁴¹ The relational model of autonomy challenges the assumed homogeneity of groups that are characterised by inherent vulnerabilities. Although the conditions causing situational vulnerability often coincide with those that undermine autonomy, this is not the case for inherent vulnerability. Individual women do not have diminished mental competence due to pregnancy; nor do social conditions render pregnant women as a group globally non-autonomous. From the perspective of relational autonomy, there is no impediment to ethical research using pregnant women merely due to the inherent vulnerability of pregnancy.

2.6 Relational Autonomy and Informed Consent

The ethical guidelines mostly treat informed consent and autonomy as interchangeable and suggest that securing informed consent is both necessary and sufficient to respect autonomy. The relational model challenges both assumptions.⁴² On the relational model, it is possible to respect and promote autonomy without obtaining informed consent. Consider children or people with a cognitive disability who cannot provide genuine informed consent. Relational conceptions of autonomy imply that there are many ways in which autonomy can be fostered, such as providing meaningful options and reinforcing self-regarding attitudes of self-confidence, self-trust and self-worth. The ethical obligation of respect for relational autonomy is therefore distinct from the requirement to obtain informed consent. If health research on children or people with cognitive disabilities were envisioned, it may be necessary to secure informed consent from a parent or guardian. But this would not imply that the ethical obligation of researchers to respect the autonomy of subjects themselves has been discharged. Researchers would have an additional obligation to foster the relational autonomy of the subjects of the research.

The relational model also questions the assumption that informed consent is sufficient for autonomy.⁴³ The obligation to obtain informed consent requires health researchers to provide relevant information and an opportunity for subjects to weigh up costs and benefits on the basis of the information. This obligation is not onerous, as McLeod comments vis-à-vis informed consent in clinical contexts: ‘[r]arely does significant communication about a patient’s options occur’ besides either recommending the procedure, or in cases in which a recommendation is not possible, ‘[saying] to the patient that she has to choose based of her beliefs and values’.⁴⁴ As we have seen, an informed decision is not always an autonomous one. The Tuskegee subjects did not suffer cognitive impairments, so in principle they had the capacity to evaluate information and decide to participate in the study. On the relational model however, this decision is unlikely to be even locally autonomous. The Tuskegee subjects lacked recognition by others of their status as moral equals; they likely also lacked self-trust or a sense of themselves as the equals of the researchers conducting the study. If so, even informed decisions would not count as autonomous. The relational model implies therefore that health researchers have additional ethical obligations than simply to obtain informed consent. Consider again the example of placental sampling. Dove et al. suggest that requests by some women to consult their partners indicate a form of relational rather than individualistic decision-making. This may be true, but it would be wrong to assume that the requirement to respect relational autonomy would be discharged by providing subjects an opportunity to consult family members. Indeed, it has been argued that respecting relational autonomy implies quite stringent obligations, for instance an epistemic obligation to understand the woman’s subjective point of view.⁴⁵ For instance, in the placental sampling case, will consulting a partner foster the woman’s autonomy, or does it indicate a disposition to defer to a male partner’s wishes due to lack of self-trust or self-worth? (Recall the case of Mrs H. above.) In general, respecting relational autonomy requires the provision of meaningful options – including the option to say no – and an environment that promotes rather than undermines the necessary competencies and self-regarding attitudes.

I have argued that securing informed consent, as understood on the individualistic model, is insufficient to respect relational autonomy. However, could the obligation to secure informed consent itself be revised in ways congenial to the relational approach? Guideline 9 of the 2016 International Ethical Guidelines for Health-Related Research Involving Humans states that informed consent should be a ‘two-way communicative process’, and that persons obtaining the consent should ensure that subjects comprehend the relevant information. For example, the information must be in ‘plain language understandable by the applicant’, and ‘the researchers must make themselves available to answer questions’.⁴⁶ These latest guidelines therefore adopt some of the lessons of relational autonomy. They focus attention on the ways on which the availability of researchers to answer questions could facilitate a better-informed process and hence on the effects of interpersonal attitudes on individuals’ decision-making. Yet they do not fully articulate the potential for understanding informed consent in relational terms. One promising avenue of research is the framework of ‘epistemic injustice’, which argues that people can suffer injustice due to unjustifiable disregard for their status as full epistemic agents.⁴⁷ For example, due to their social subordination, the epistemic agency of the Tuskegee subjects was ignored and there was no attempt to secure informed consent. Epistemic injustice can also occur in cases in which the need for informed consent is recognised. Indeed, it may be especially likely to occur in the context of health research due to the privileged epistemic position that researchers occupy as experts in the research field. As José Medina has argued, having privileged status – epistemic or otherwise – tends to make one more likely to display epistemic vices such as epistemic arrogance, or the expectation of ‘not being called into question in one’s opinions, that is having an undisputed cognitive authority’.⁴⁸ Such epistemic vices potentially will interfere with obtaining appropriate consent. Attention to the relational context in characterising informed consent reinforces the conclusion of previous sections that the ethical obligations of researchers who study human subjects are more demanding than set out in the current guidelines. In addition to the obligation to foster the social and interpersonal conditions that promote autonomy, they may have special epistemic obligations, namely to cultivate epistemic virtue and attempt to eliminate epistemic vice.

2.7 Conclusion

This chapter has argued that that the current guidelines for health research ethics are implicitly committed to an individualistic conception of autonomy. This is an implausible conception because it does not recognise ways in which people’s capacities for autonomy depend on social conditions, family and community relationships, and interpersonal attitudes. The chapter outlined a competing relational model of autonomy and argued that the ethical obligations derived from the relational model are more stringent than those derived from the individualistic conception.

3 Proportionality in Health Research Regulation

Owen Schaefer

3.1 Introduction

Proportionality in health research regulation can, at its broadest level, be understood as an attempt to balance two considerations that sometimes compete: the protection of individuals affected by research – especially, but not limited to, human subjects – and the promotion of socially valuable research. This chapter will explore the concept of proportionality through three sections: First, a clarification on what I mean by proportionality in this context and why it is important; second, an exploration of how particularly challenging it is to assess proportionality; and third, a proposal for a procedural approach to proportionality that may assist with those challenges. In particular, I will propose that adopting a facilitative attitude, undertaking rigorous justification, ensuring transparency and engaging with relevant stakeholders may be effective procedural means of overcoming the challenges of proportionality.¹

3.2 What Is Proportionality?

The term ‘proportionality’ has several meanings even within the context of health research regulations. We can roughly distinguish between the first-order or study-level sense of the term, and second-order or policy-level sense.

First-order proportionality refers to the benefits of a study – inclusive of benefits to the subjects as well as society as a whole – being proportionate to its risks and burdens. It is interchangeable with ‘favourable risk-benefit ratio’ as found in the classic article ‘What makes clinical research ethical?’,² and a variety of authors have followed suit.³ On this understanding, the benefits of a given study need to be of sufficient strength or magnitude to justify the risks individuals are exposed to. Research Ethics Committees (RECs), Institutional Review Boards (IRBs) or equivalent are routinely tasked with making such assessments on a case-by-case basis for human subjects research.

I will set aside assessment of first-order proportionality in this chapter, as risk–benefit ratios will be covered elsewhere in this volume.⁴ Instead, I will focus on second-order proportionality in health research, which operates primarily at the policy level (inclusive of national and institutional policies).

Second-order proportionality refers to whether the burdens of a given rule or policy governing research are proportionate to its benefits. The burdens and benefits can further be delineated along two axes: effects on the process of research including monetary/staffing costs, researchers’ time, efficiency and scientific validity; and the effect on protection afforded to individuals affected by research (including, but not limited to, human subjects). As will be discussed below, this is not only limited to physical effects, but extends to other impacts such as the wrongdoing of privacy violations.

Proportionality assessments then, will involve evaluating the benefits of a regulation in terms of both protection and promotion, and weighing those against the burdens, also in terms of protection and promotion. While we might typically expect research regulations to impose burdens in terms of process while affording benefits in terms of protections, we should keep in mind that regulations can also have beneficial effects on processes, and deleterious effects on protections, as seen in Table 3.1.

Table 3.1 Matrix of regulations’ potential effects on research

	Process (monetary/staffing costs, researchers’ time, efficiency, and scientific validity)	Protection (relating to the interests of human subjects or other individuals directly affected by research)
Benefits (positive effects)	e.g. streamlining research review	e.g. data security protocols to protect confidentiality of subject data
Burdens (negative effects)	e.g. substantial time from researchers to ensure compliance	e.g. retaining sensitive study data for years, increasing risk of breach

Like first-order proportionality, there is a justificatory relationship: the benefits of a rule or policy must be sufficient to justify the burdens imposed. But unlike first-order proportionality, second-order proportionality is not evaluated on a case-by-case basis. Rather, it concerns the total effect a given policy has on the research enterprise. It is the responsibility of policymakers – including regulators and institutional leaders – along with institutional bodies like RECs and IRBs, to ensure that their policies are proportionate in this way.

Still, context will be important in assessing the proportionality of a given policy. Rules will have different impacts on different institutions, fields of study, countries and cultures. For example, a rule requiring written informed consent from subjects – which ensures consistent provision of information and ease of auditing – may be quite proportionate in societies with high literacy. But in societies with low literacy, the requirement would lead to the exclusion of many subjects, potentially endangering the scientific validity and depriving already-marginalised groups of potentially beneficial interventions. This could tip the rule from being proportionate to being disproportionate.

3.3 Proportionality of Review and Proportionality of Harms

Discussions around second-order proportionality typically focus on two related aspects: proportionality of review, and proportionality of harms. Proportionate review involves tailoring the degree of scrutiny to the amount of risk subjects may be exposed to.⁵ Proportionality of harms defines those risks in terms of probability of physical or psychosocial harms.

In regard to proportionality of review, low-risk research may be reviewed under expedited or exempted pathways, where only one or two members of a REC are directly involved in assessing and approving a study. Higher-risk research would instead go to a full board.

Full board reviews take more time, potentially reducing the efficiency of research with potential social benefits. However, they also are more likely to pick up on potential ethical failings, due to both the larger number of eyes on a proposal, and the greater diversity of expertise brought to bear on it. This will be more proportionate for studies with higher risks to subjects, and so in need of closer attention. For low-risk studies, there may not be much reason to apply that extra scrutiny, as the marginal benefit to subjects of correcting a failing is relatively small. At the same time, the study will still consume resources, which may be separate grounds for some scrutiny.⁶

However, ethics review is just one component of research oversight. Many policies governing health research operate using different mechanisms, including rules that bind researchers directly, regardless of the scrutiny applied. These include policies delineating the contents of informed consent, confidentiality protections, documentation and authorisation. All these requirements have the potential to slow down research or increase its costs, and so must be justified in terms of the benefits they afford. Stratifying the stringency of a wider variety of rules is more common in Europe, while US regulations only stratify the review process.⁷

Additionally, approaches centred around proportionality of harms capture only part of the justifications for rules governing health research. The four principles approach can help illustrate this. Beauchamp and Childress identified four central mid-level ethical principles that underpin bioethics in general, and research ethics in particular: beneficence, non-maleficence, respect for autonomy and justice.⁸

Beneficence in this context relates to the impetus to ensure that socially beneficial research is conducted in an efficient manner; inefficiencies resulting from over-regulation increase the overall cost of research, in turn reducing the amount of research – and thus social benefits generated – that can be conducted on a given budget. Non-maleficence refers to the potential harms of research mentioned above. On the above understanding, proportionality would primarily involve balancing beneficence – in terms of promoting valuable research – against non-maleficence (avoiding harms caused by research).

This leaves out autonomy and justice, which are also relevant to proportionality assessments. Consider the following examples of informed consent – as an illustration of respect for autonomy – and subject selection (illustrating justice concerns).

Informed consent may in part be aimed at harm mitigation, so subjects can avoid participating in trials whose risks are unacceptable to them personally. But it also aims at respecting the ability of subjects to govern their own lives – here, to ensure that participating in a study is in accordance with their values. This includes risks, but may also relate to other factors such as how much they identify with the aims of the study, trust researchers, or believe it will produce social benefits.⁹

Fair subject selection might also have a risk-mitigation aim, insofar as subjects particularly vulnerable to harm from a study may be excluded. But this must also be balanced by justice considerations in excluding certain groups from a study. A case in point is the routine exclusion of pregnant women from research. This is done in the name of non-maleficence, as fetuses are frequently thought to be at higher risk of harm from experimental interventions. But the result is a lack of evidence for the safety of a wide variety of drugs on pregnant women, forcing them into an uncomfortable dilemma: accept substandard care with a more proven safety record, or go for proven interventions that have an uncertain risk to their children. As such, a rule meant to protect pregnant women arguably perpetuates injustices against them.¹⁰

Assessments of proportionality should go beyond benefits and harms to incorporate considerations of justice and respect for persons. These considerations may factor in on both sides of the proportionality equation: the burden of regulation may be necessary to prevent an injustice and promote autonomy, or – as with the case of preventing research with pregnant women – a regulation’s burdens may be shown to be unjustifiable by virtue of the injustice and disrespect that it promulgates.

3.4 The Challenge of Social Value

From the preceding discussion, it should already be evident that assessments of proportionality of a given policy governing health research will be quite complex. Further challenges emerge upon closer analysis, one of which is how to integrate the social value of research into proportionality assessments.

Up until this point, it has been assumed that greater efficiency, lowered cost and improved scientific validity in health research are unquestionably valuable. This is predicated on a potentially contestable notion – that the outputs of research have substantial social value. If policies slow down research, then in turn society’s access to valuable outputs – more effective treatments, better prevention of disease, mitigation of symptoms and side-effects, etc. – will slow down. Increased costs mean less research can be done, and thereby fewer valuable outputs are produced. Further, detriments to scientific validity – such as limitations on the use of placebo-controlled trials – may undermine the robustness of those outputs.

It is often held that all health research must contribute to social value in order to be ethically justified.¹¹ For present purposes, it is sufficient to note that if a given study really has no social value, proportionality is irrelevant – it should not be permitted in the first place.

Meeting the minimal threshold of social value masks the larger, much more intractable issue of assessing the magnitude of that social value. This magnitude is important in proportionality assessments to get an understanding of how problematic a given inefficiency or other detriment really is. For low social value research, barriers to research may more easily be outweighed by ensuring protections for subjects; vice-versa for high social value research.

But there is no reliable formula for quantifying the social value of a given study. The results of all studies are by nature uncertain – if we knew the results ahead of time, there would be no need to engage in a study in the first place. Much health research is not directly translatable; it instead builds a base of understanding that over time, in combination with other studies, will eventually lead to improved practice down the road. It is also unclear how we should judge the impact of health research. Some measures like Quality-Adjusted Life Years are relevant here, but these have been disputed as too abstracted a way from patient experiences and values,¹² and being potentially discriminatory.¹³

Still, some reasonable estimates of possible social benefits of a given study must be possible. This is the routine task of agencies that disburse grants for research, after all. Moreover, research regulation can itself play a role in improving the social value of a study. For example, beyond risk assessment, a REC/IRB review of scientific validity and value can play a role in promoting sound knowledge generation from studies.¹⁴

3.5 Complexities of Analysis

The evaluation problem is compounded at the policy level, where broad categories of research are being considered, rather than individual studies. Furthermore, the social value of research is just one piece of a proportionality assessment. We need to ascertain not only the social value of the research, but the extent to which a given policy will detrimentally impact this research. While it is routine for policymakers and academics to perform economic impacts of regulations, these analyses have been questioned in terms of their rigor and real-world validity.¹⁵ In the research context, a full analysis would have to take into account the extent to which increased cost of research would crowd out further socially valuable studies, assuming total budgets for research are independent of regulatory policy.

There are similar uncertainties and complexities in relation to assessing protections. Only a few policies will be relatively easy to assess: prohibitions of activities that would almost certainly harm or wrong participants. Failure to obtain informed consent from competent adults for interventional research may be an example, insofar as it would be a clear violation of autonomy.

But often, harms or wrongs are probabilistic, with the probabilities themselves unknown or uncertain. Policies may only reduce the likelihood of such harms or wrongs, rather than prevent them entirely. For example, a requirement that researchers provide their CVs to review boards has some use – ethics board members can ensure that they have the relevant qualifications to carry out the study procedures. At the same time, it would be difficult to articulate exactly how much harm (qua adverse outcomes) is actually prevented by such a policy. In theory, one could perform a comparative analysis of researchers with different qualifications and assess correlations with adverse study outcomes. But such analyses have not been done for the wide array of subfields and procedures that ethics committees may encounter, so those setting policies on the matter must instead rely on personal judgment.

Finally, there is the question of how to bridge the two sides of proportionality, namely, burdens and benefits. That is, how to determine if a given burden on research is justified by the benefits it affords? A potential approach is to leverage decision theory, where the gains and losses to individuals’ well-being from a regulation are quantified and aggregated, and a determination is made as to whether a given regulation or policy overall improves expected utility.¹⁶

Here, incommensurability is a particular issue. Even if one side or the other can be somehow defined and explicated, the values on either are likely not commensurate – they are not easily compared and weighed up against each other. Perhaps the burden of requiring informed consent for some secondary data research can be quantified in terms of the increased cost, delays and potential bias of only including those who would consent. But those measures are of an entirely different nature from the autonomy interests of individuals to maintain control over data about them, one of the main values being protected by informed consent requirements. Unfortunately, again, there is no formula to make such assessments, and a good deal of individual judgment on the part of policymakers is necessary.

3.6 Procedural Approaches to Proportionality Assessments

There is an old joke that philosophers like to kick up dirt in front of their eyes, then complain they can’t see. The above may appear to be like so much dirt, pointing out all the difficulties in doing a proper proportionality assessment of health research regulations. So, at this point, I will be somewhat more constructive and propose some ways that proportionality assessments can be made more reliable and legitimate.

The following will primarily be procedural proposals. That is, they are not explications for how to determine whether a given regulation is proportionate. Rather, they are a series of structures and systems governing the process of assessing proportionality that should have two desirable features: they will improve the reliability of proportionality assessments, by prompting systems that are better able to assess whether a given rule’s protections really justify the burdens imposed; and they will help engender legitimacy in those assessments, by adopting systems that can earn the confidence of stakeholders – including researchers and participants – in the proportionality of rules that are ultimately produced.

At the most general level, it is important for those involved in regulation to have a facilitative attitude towards proportionality. A view that their role is solely to protect subjects may engender the perception that they are there to get in the way of valuable research.¹⁷ This will not only engender hostility towards regulations and regulators but is also fundamentally mistaken. It ignores the crucial consideration of proportionality; to ensure that any burdens are adequately justified by the benefits they bring about. As a result, proportionality requires consideration of both the positive and negative sides of the research enterprise.

A more useful way for regulators to frame their role would be as facilitators of responsible research. This is not to say that they are there to make research easier than it might otherwise be; almost any regulations will have some costs in terms of efficiency, expense or validity, and regulators should be up front about that. However, enforcing reasonable rules that are proportionate to the burdens they accrue is a means to ensure that the research that does occur is responsible in terms of the benefits to those affected. The term ‘facilitation’ gives explicit emphasis on the need to ensure that the regulations are as minimally intrusive as necessary to achieve a given protective aim.

This framing has both inward-facing and outward-facing benefits. Looking inwards, regulators are reminded of the need to consider burdens of regulation along with benefits, and the balancing effort between the two in the proportionality assessment. This will help avoid blatantly one-sided approaches to regulation. Also, looking outwards, expressing this attitude in engagement with stakeholders can help assure them that their interests are being adequately accounted for. Such engagement is not merely limited to top-down communication of regulatory decisions, but active engagement as will be discussed further below.

A related procedural approach is actually doing the work of a proportionality assessment – that is, providing rigorous justification of a rule or policy’s proportionality. It may be tempting to give up in the face of the uncertainties and ambiguities discussed. Nevertheless, responsible regulation must proceed. Ignoring proportionality can lead to one-sided policies, which either produce overly protective regimes with unacceptably burden research, or overly permissive regimes that do not adequately provide protections out of fear of inhibiting research.

And it will be work, indeed. When a given rule is under consideration, a non-trivial amount of research and analysis will be needed. Is there evidence on the magnitude of the harms or wrongs being prevented? What about the effectiveness of the proposed rule? And on the flip side, what effects will it have on the research enterprise? What are the quantifiable and non-quantifiable costs? Finally, when all those considerations are taken into account, can the regulation’s protective effects truly justify the burdens imposed? And if not, can it be refined so that it does?

The final justificatory step may be the most uncertain and challenging. In some ways, it is an ethical or normative question relating to the values promoted and inhibited by a given policy. Regulators are not typically trained in philosophical analysis that may assist here, but some features of decision-making can be highlighted. These include articulation of the competing values at stake; scrutiny of any empirical evidence adduced; consistency between different judgments; clarity in terms of the reasons a given rule is justified, or not.

There is not space to elaborate here on such analytical tools. Indeed, no single article could adequately do so. Instead, it may be that regulators – or at least, some individuals in the regulatory process – should receive training in these analytical tools. As it stands, many relevant degrees like Masters of Public Policy or Masters of Public Administration do not routinely integrate such analytical training into their curriculums, focusing instead on social sciences. Reform of these curriculums might help boost competence in performing proportionality assessments. Alternative educational systems should also be considered, such as short courses, blended learning modules and ad hoc training workshops that may be more practicable for working professionals.

Especially because of the difficulty of making proportionality assessments, transparency in justificatory analyses will be crucial. Transparency here refers to some public promulgation of the reasoning process behind the decision that is reached. This would not only be easily accessed by stakeholders, but promulgated to relevant stakeholder groups so they are aware it exists in the first place.

Almost any rule will involve some trade-offs between protection of individuals and minimising burdens on research. As such, criticism from some affected stakeholders is inevitable. Having the reasoning and evaluation of a proportionality assessment will not eliminate that criticism, but it can go some way towards blunting suspicion that such an assessment was one-sided or ignored their concerns.

Moreover, there is good reason to suppose that stakeholders are owed this sort of transparency. For researchers, regulations have coercive force – failure to abide by them will result in penalties, whether criminal, civil, or – in the case of instructional policies – professional. It is a matter of respect to those individuals who are liable to such punishments that the reasoning process behind the rules is laid out in full. Other individuals like research subjects have a different relationship with regulations; while regulations do not directly bind them, they are carried out in their name. And if a regulator decides against enacting a given protective rule, that regulator is deciding to permit a certain degree of risk of harm to accrue to participants or others. Those affected individuals deserve to know the reasoning process behind this decision, as they may well be harmed by it.¹⁸

Another benefit of transparency is that it can prompt regulators to ensure their reasoning is truly defensible. Behind closed doors, there may be a temptation to wave away concerns that are too difficult or complex. By making their reasoning public, they are compelled to seriously reckon with all the considerations that stakeholders may find relevant. If not, they will be open to – legitimate – scrutiny and critique for inadequate analysis that will undermine confidence in the rules that are put forth.

Promulgation of reasoning and justification from regulators to stakeholders is important, but limited insofar as it is top-down and one-way. A more thoroughgoing and robust way to ensure adequate consideration of competing interests and earn public trust in proportionality assessments is to directly engage with those groups, to allow the co-creation of rules and collaborate assessment of the thorny issue of proportionality.

There are a myriad of ways that stakeholders can be engaged in proportionality assessments. For more details on approaches to and justifications for public engagement, see Aitken and Cunningham-Burley, Chapter 11 in this volume (on public engagement and access), and Burgess, Chapter 25 in this volume, (on public engagement and health research regulation).

These approaches are especially valuable for complex and uncertain issues like proportionality assessments. A small group of regulators may have parochial approaches or biased analyses that can be avoided by the involvement of a larger body of stakeholders. It may also relieve some of the pressure to make such complex judgments on their own, by soliciting assistance from a wider group.

This engagement should not be seen as one-off, or only occurring prior to rulemaking. A truly proportional approach to regulation must recognise the potential fallibility of initial judgments, and the fact that the situation on the ground may change. Protections previously seen as adequate could become threatened. For example, DNA profiles have recently been shown to be re-identifiable, which means previous protections merely stripping names and other extraneous information from such profiles are no longer sufficient to guarantee anonymity.¹⁹ Previously burdensome compliance can be made easier by new technologies, as arguably occurred with the advent of digital compilation of ethics review documents allowing for more rapid collation and assessment.

For this reason, engagement should be a continual process, with the proportionality of a given rule periodically up for review and re-evaluation. Regulators may not be equipped to maintain such active review, so instead being open to updates and comments from stakeholders may be optimal. This both relieves regulators of some burden to keep regulations’ proportionality up to date, and ensures stakeholders have a continued ability to positively impact the rules that affect them.

To be sure, there are limitations on how much engagement can do. It was noted earlier that regulators may need additional training to adequately undertake proportionality assessments. This would already be practically difficult with regulators; with broader stakeholder groups, it is probably impossible. As such, there may be some limit on the extent to which co-creation is achievable for matters as complex as proportionality assessments. Still, we should not allow the perfect to be the enemy of the good; engagement has substantial value, as explained, that can supplement the deep analysis that regulators are responsible for.

3.7 Conclusion

In this chapter, I have explored the notion of proportionality in the context of health research regulation. Proportionality was defined in terms of a justificatory relationship: the benefits afforded by a given rule must serve to justify the burdens imposed by it. Assessing proportionality is no easy task; it is beset by uncertainties and challenges of analysis at a variety of levels, and involves weighing of different values – relating to beneficence, non-maleficence, justice and autonomy – that are non-commensurate and often non-quantifiable. The task of proportionality assessment is not impossible, however. Indeed, it is a necessary part of responsible regulation of health research. I have suggested several procedural approaches that can help improve the reliability and legitimacy of those assessments: a facilitative attitude; rigorous justificatory analysis; transparency in reasoning; and engagement in decision-making. These procedures recognise that we cannot formulaically produce an answer as to whether a given regulation is proportionate, and judgement is required. Hopefully, the contents of this chapter – in conjunction with the other material in this volume – can go some way to assisting those involved in regulation in understanding the nature, importance and practice of proportionality assessments.

4 Social Value

Johannes J. M. van Delden and Rieke van der Graaf

4.1 Introduction

This chapter starts from the assumption that science is a matter of co-creation. To open up science to democracy means that we have to think about the social value of research, which in itself we cannot leave to science to evaluate. This raises detailed questions around patient and public involvement (PPI) in deciding which research to perform, and about how to handle conflicts between individual and public interests. These are addressed elsewhere in this volume.¹

In this chapter we focus on social value in health-related research involving humans, including data driven research. We first describe the background to the concept of social value and its meaning. Then we examine the concept itself and define the social value of an intervention as the value that an intervention could eventually have on the well-being of groups of patients and/or society. We also discuss some of the open issues in the scholarly debate about the concept of social value.

We find that to state a requirement for social value is one thing; to actually evaluate the social value of a research project in a Research Ethics Committee (REC) is another. We therefore elaborate on how the requirement of social value can be applied. We argue, first, that it is important to have this requirement as a separate condition. To increase systematisation, we further discuss how social value can be assessed in the steps that together constitute the risk-benefit task of RECs.

Returning to our opening statement, we argue that the addition of the requirement of social value can be seen as a consequence of a change in the sociology of science. It illustrates the move away from a science–internal understanding of scientific validity into an inclusive understanding of social value. Accepting social value as a requirement for research to be evaluated by a REC means that social value has matured from an attractive but illusive idea into something that has to be assessed, evaluated and optimised and can be used to address some of the justice issues in healthcare.

4.2 Social Value in the 2016 CIOMS Guidelines

Social value is a key principle in the 2016 version of the International Ethical Guidelines for Health-related Research prepared by the Council for International Organizations of Medical Sciences (CIOMS) in collaboration with the World Health Organization (WHO). The account of social value in this chapter has been largely influenced by the wording in the 2016 CIOMS Guidelines. Its very first guideline reads:

The ethical justification for undertaking health-related research involving humans is its scientific and social value: the prospect of generating the knowledge and the means necessary to protect and promote people’s health. Patients, health professionals, researchers, policy-makers, public health officials, pharmaceutical companies and others rely on the results of research for activities and decisions that impact individual and public health, welfare, and the use of limited resources. Therefore, researchers, sponsors, research ethics committees, and health authorities, must ensure that proposed studies are scientifically sound, build on an adequate prior knowledge base, and are likely to generate valuable information.

Although scientific and social value are the fundamental justification for undertaking research, researchers, sponsors, research ethics committees and health authorities have a moral obligation to ensure that all research is carried out in ways that uphold human rights, and respect, protect, and are fair to study participants and the communities in which the research is conducted. Scientific and social value cannot legitimate subjecting study participants or host communities to mistreatment, or injustice.²

The entry of the requirement of social value in the 2016 CIOMS International Ethical Guidelines for Health-related Research involving humans was certainly not unprecedented. Many scholars trace its origins back to the Nuremberg Code of 1947, which states that ‘The experiment should be such as to yield fruitful results for the good of society’.³ Also, it is commonly understood that the social value of a research project may be part of the evaluation of risks and benefits of such a project.⁴ The concept also plays a key role in the Belmont Report, the World Medical Association’s Declaration of Helsinki, and the Common Rule. Furthermore, social value is considered to be of relevance when international collaborators are conducting health research in resource-limited settings. The concept also plays a key role in frameworks for research ethics, such as the ‘7- principle-framework’ of Emanuel and colleagues⁵ and the component analysis framework of Weijer and Miller.⁶

4.3 Social Value as Indication for a Change in Sociology of Science

The addition of social value to the 2016 CIOMS International Ethical Guidelines at this point in history can be understood as part of a broader movement within the sociology of science, which describes how people come to accept certain scientific statements. Elements of this movement can also be seen in other guidelines within the 2016 CIOMS Guidelines, such as those on Community Engagement (7) and Public Accountability for Health-related Research (24). A first example of this broader movement within the sociology of science is the current critique of science and scientific knowledge.⁷ Part of the critique concerns the replicability of research results, which in some areas is disturbingly low. Another part concerns the way in which scientists are evaluated: in many areas of science this is done, at least until recently, by looking at the number of articles produced and/or the number of times an article is cited – e.g. combined into the Hirsch-index – creating an incentive to produce enormous quantities of papers. But the most important critique – also implied in the former point – is that science appears to be concerned more with producing science as such, than with furthering socially valuable goals through research. The term ‘research waste’ was coined to describe the result of this way of doing research.

In response, we currently see programmes such as the EU programme on Responsible Research and Innovation, movements such as that for Open Science – which is certainly about more than just open access publishing – and Science in Transition.⁸ These programmes try to reinvent the sociology of science in order to enable it to perform the tasks society has entrusted to scientists. They also encourage the involvement of all stakeholders in the production of science, including patients and publics, in order to increase the relevance of research results. Present-day problems in society are simply too complex to think we can solve them without cooperating across borders. Science cannot continue to take its own interests as primary, instead of living up to its societal task. Science needs to earn and deserve a so-called social licence for research.⁹ PPI in research is an essential means to mitigate concerns on research waste.

There are a number of reasons why we need PPI in research – as addressed in more detail elsewhere in this volume.¹⁰ First, this is because research is about all of us! And nothing should be done ‘about us, without us’. We therefore need a model in which patients consider themselves as partners in a trustworthy system, not just passive sources of information. Second, the purpose of patient involvement is ultimately to improve our health. By this we do not mean through individual healthcare. Rather, we suggest that this can come about by ensuring that those who conduct research projects ask the right questions, use the right endpoints, make the right choices and effectively implement their findings. This illustrates the efficiency argument as applied to input from patients – and wider publics – who are similarly motivated to find answers to health and disease-related questions. It is believed that this will help science to become more socially valuable and thus to reduce research waste.

These developments also point to important questions in the area of the philosophy of science. It is common to think that science produces facts that are independent of public preferences. Shouldn’t science inform democratic decision-making rather than being influenced by it? What is left of scientific independence if we allow PPI in research? It is generally understood why democracies need science, but why would science need democracy?¹¹

To answer these questions we turn to Science and Technology Studies (STS) where several schools of thought can be discerned. The first (1900–1960) was a positivistic one: it was believed that science was a way of knowledge-making and that its knowledge was absolute and universalistic.¹² The correctness of scientific research needed no social explanation, it was simply true. What needed explanation was how false beliefs were mistakenly taken to be correct, typically by pointing at prejudice, bias and so on. This is what Nowotny calls Mode 1 research.¹³ Although this view is no longer supported by social science, it remains the common-sense view of many scientists and the public. One needs only to watch an episode of CSI to see how a forensic scientist reveals ‘the truth’ about the case.

The second school of thought (1960–2000) started when others took the work of Kuhn and other researchers to show that scientific truth is best seen as an outcome of negotiation and agreement located within social groups. Science is a human activity subject to all the strengths and flaws of humans. Nowotny speaks about Mode 2 research in which interaction between science and society is taken as a starting point and science has become a matter of co-creation.¹⁴ Science needed to be democratised. This second school illuminated the constructivist side of science, in order to deconstruct science, but did less to provide an alternative.¹⁵ A risk of this type of thinking is that this may produce the kind of relativism in which scientific claims have become ‘just another opinion’ and alternative facts are as good as any other account.

To counter this, the third school (after 2000) emphasises that we do not need to end up in relativism, and that there are more arguments in favour of some claims about states of the world than there are for others. Textbook science is not perfect, and remains open to revision, but is more reliable than primary research, because we have more reasons to accept the claims in a textbook than in primary research. In ethics, the Rawlsian understanding of ethical claims as provisional fixed points captures the same idea: claims are always open to revision (hence ‘provisional’) but we have good reasons to accept them (hence ‘fixed’). It is important to note that the last school of thought accepts the rationale established by the former, but tries to make the next, constructive step.

We think that the addition of the requirement of social value into the CIOMS Guidelines can be seen as a consequence of this change in the sociology of science. It clearly illustrates the move away from a science–internal understanding of scientific validity into an inclusive understanding of social value. It sends the message that science needs to be cognisant of its societal role and should explain how it aims to fulfil that role. That message is reinforced by guidelines on community consultation and public accountability. Placing social value as a requirement in a list of conditions to be evaluated by a REC means that social value has matured from an attractive but illusive idea into something that has to be assessed, evaluated and optimised. In other words: social value has gained ‘teeth’.

4.4 Meaning of Social Value

We will now zoom in on the meaning of the concept ‘social value’ itself. According to Wendler and Rid, the standard view on social value is that ‘it is an ethical requirement for the vast majority of clinical studies’.¹⁶ They also argue that there is ‘strong support’ that social value of research is important ‘for protecting participants who cannot consent, preventing inappropriate research that poses high net risks, and promoting appropriate investigator behaviour’¹⁷ (see also below).

Here is the description of the meaning of the term social value according to the 2016 CIOMS Guidelines:

Social value refers to the importance of the information that a study is likely to produce. Information can be important because of its direct relevance for understanding or intervening on a significant health problem or because of its expected contribution to research likely to promote individual or public health. The importance of such information can vary depending on the significance of the health need, the novelty and expected merits of the approach, the merits of alternative means of addressing the problem, and other considerations.¹⁸

We next examine separately the concepts of value and social value. We understand value to mean the potential of a study to improve health, broadly construed as biological, psychological or social well-being.¹⁹ Health value can be categorised along two dimensions: immediate versus future health value, and the population that receives this value.²⁰ It is also important to note that social value is attributed both to information that has direct relevance in promoting health, and to the contribution this information may have for subsequent valuable research.

The concept ‘value’ has been scrutinised in many different research fields such as sociology and philosophy. However, little agreement exists on how ‘value’ should be defined. Consensus does exist on the fact that values arise out of human experience. Whereas the term ‘benefit’ refers to an advantage or profit gained from something, the concept of value refers to the regard that something is held to deserve. The latter is thus a relational concept; both the object to be valued, and an evaluator are necessary preconditions for value to exist.²¹

Turning next to ‘social value’, this functions in two main ways in our everyday use. First, social value can be seen as values shared by a community of individuals; they are values held by society and are contrasted with individual (non-shared) values. By social value, we refer to socially collective beliefs and systems of beliefs that operate as guiding principles in life. Second, besides values of society, the concept can also be used to refer to values for society. Here, social value is an assigned predicate or property of an object, and, in our case, of health-related research.²² This implies that we have to assess the importance of the information in terms of the nature and magnitude of the expected improvement an intervention – as assessed in the study – is expected to have on society. Note that benefit for the individual research participant would be called a direct benefit. Social value is not about rewarding careers for scientists, employment for citizens or a sense of fulfilment for participants.²³

We conclude that the social value of an intervention encompasses the value that an intervention could eventually have on the well-being of groups of patients and/or society. In case of early phase trials, this value may lie in the distant future; in those cases, RECs may also assess the ability of trials to promote progression to later stages of research in which successful clinical translation becomes more likely.

It is important to note that the CIOMS guideline on social value also explicitly talks about what social value cannot do, as follows:

Although scientific and social value are the fundamental justification for undertaking research, researchers, sponsors, research ethics committees and health authorities have a moral obligation to ensure that all research is carried out in ways that uphold human rights, and respect, protect, and are fair to study participants and the communities in which the research is conducted. Scientific and social value cannot legitimate subjecting study participants or host communities to mistreatment, or injustice.²⁴

This provision is a reformulation in human rights language of the so-called primacy principle. This is the ethical principle stating that the individual shall have priority over science, found, for instance, in guideline 8 of the 2013 Declaration of Helsinki: ‘While the primary purpose of medical research is to generate new knowledge, this goal can never take precedence over the rights and interests of individual research subjects’.²⁵ There is an ongoing debate about the tenability of this primacy principle²⁶ which deserves a separate discussion.

4.5 Social Value in Scholarly Debate

Whereas the merits of the social value requirement have been largely uncontested, over the past few years the concept of social value has received increasing scholarly attention. Among others, the journal Bioethics launched a Special Issue (2017, 31(2)) on social value. Also Danielle Wenner’s²⁷ analysis of social value in the Hastings Center Report led to several responses.²⁸ The attention has not only led to improved understanding of the meaning and scope of social value but also to more critique. Next, we will consider some of the key points from this ongoing debate.

Traditionally, social value has been located in the context of clinical research, but more recently the concept has also been introduced in health systems research and into the global health ethics debate.²⁹ Whereas the concept, as discussed above, in clinical research focuses on the knowledge to be gained for society in general, in public and global health ethics the requirement seems to have a different role. For instance, according to Nicola Barsdorf and Joseph Millum, social value should be seen as ‘a function of expected benefits of the research and the priorities that beneficiaries deserve’.³⁰ Social value then also becomes a means to address questions of priority setting,³¹ promotion of health equity and addressing health inequality.³² At the same time, in the context of health systems research, some argue that its social value can also be justified ‘in pragmatic systems rather than linked only to priority setting’.³³

Further discussion centres on whether the concept of social value should be located in the traditional account of research ethics that has a focus on clinical trials and observational research. According to Wendler and Rid, there are eight reasons that ‘taken together provide strong support’ that social value must be obtained in the context of clinical research: (1) to protect participants who cannot consent; (2) to ensure the acceptability of high-risk research with competent adults; (3) to maintain researcher integrity; (4) to avoid participant deception; (5) to safeguard against exploitation; (6) to exercise stewardship of public resources; (7) to promote public trust; and (8) support for clinical research.³⁴ Others, like Wenner,³⁵ Wertheimer³⁶ and Resnik,³⁷ ground the social value requirement in other principles and outside of the traditional scope of research ethics. According to Wenner, the current view on research ethics is primarily about protection. Instead, she believes it should be grounded in justice-based considerations. She argues that certain developments in research, such as the inclusion of pregnant women, cannot be understood only from a protectionist view towards research subjects but has to be explained from underlying issues of justice.³⁸

Whereas some, like Wertheimer and Resnik, argue that studies must have ‘significant’ social value, Wendler and Rid³⁹ argue that studies should have ‘sufficient’ social value. The first group of authors expresses concern that without the qualification of significance, the concept becomes too weak, whereas Wendler and Rid argue that their understanding is also able to distinguish between studies with and without social value. Whether a study has sufficient social value should always be determined in relation to the risks of research. In some cases participants may face significant risks. However, if there is no social value to be gained, they argue that the study should not be approved even if participants consent to participation. At the same time, if the social value is limited but the risks are minimal as well, they argue it is not unethical to offer participation.

4.6 Application

In the preceding analysis we have considered both what the term social value means and the discussions that it has sparked. As such, we can now go on to look at its role in the set of requirements for acceptance of a research protocol. First, we would like to point to the importance of having this as a separate requirement. It could be argued that the social value of a research project is already being taken into account in the classical requirement in research ethics to have a favourable balance of benefits over risks and burdens. The 2013 version of the Declaration of Helsinki for instance reads: ‘Medical research involving human subjects may only be conducted if the importance of the objective outweighs the risks and burdens to the research subjects’.⁴⁰ One could conclude from this that it would not be necessary to have a separate guideline on social value. However, the problem with including social value in this so-called risk/benefit ratio is that in research projects without risks or burdens, a lack of anticipated benefit would not be sufficient grounds for a REC to deny approval of the project. If one thinks that the main aim of research ethics guidelines is to protect the individual, then one might be satisfied. If one takes a broader view and includes justice among the ethical principles that are relevant to such a deliberation, then allowing a project without benefit is unacceptable from a societal perspective. Projects still use time, money and energy in addition to contributing to more research waste. Therefore we argue that it is necessary to have social value as a separate requirement.

Some might object on the basis that social value cannot be a necessary requirement for research to be ethical since certain medical discoveries have been made by coincidence, and that requiring social value may limit medical advancement. However, accidental findings cannot be planned, nor does requiring social value mean that we will no longer find accidental findings by restricting clinical research to interventions with expected social value.

Having made the preceding claim, we now turn to the role of RECs, which are currently tasked with judging whether a favourable risk-benefit balance is achieved to ultimately decide whether a research project can proceed. This judgement has to be systematic, transparent and grounded in evidence. Evaluating the social value of a particular research project can be seen as part of this task. To increase systematisation we draw upon insights from decision-theory and propose that the risk-benefit tasks are divided into the following steps: (1) analysis; (2) evaluation; (3) treatment; and (4) decision-making.⁴¹

4.6.1 Benefit Analysis

It is the primary responsibility and expertise of investigators to map and characterise benefits, including the social value of research. However, evaluators should be able to judge whether they agree with the reasoning that supports the presented characterisation of benefits.⁴² To map benefits, we divide these into direct, collateral and aspirational benefits.⁴³ Social value can be regarded as one of the aspirational benefits. We further divide social value into: (1) the direct social value of the intervention; (2) the progressive value; and (3) the translational value of a trial.

In characterising the social value of an intervention we draw upon the proposal by Habets and colleagues.⁴⁴ They argue that at least three steps should be followed. First, the nature and magnitude of efficacy of the intervention studied in humans has to be critically assessed. Second, the anticipated clinical improvement in actual patients should be assessed, assuming that the intervention is efficacious. This means that it has to be asked whether treatment effects are meaningful, both from a medical and individual perspective, and that they have to be weighed against factors that may hamper beneficial effects, such as adverse effects and ease of use. Third, the nature and magnitude of the anticipated improvement on the well-being of patients, individuals in society and society should be evaluated. This assessment is contextual: the social value of the intervention is the expected improvement relative to other considerations, such as treatment alternatives, number of patients and costs etc. Ultimately, determining what has social value constitutes a moral judgment.⁴⁵

To characterise progressive value⁴⁶ we argue that at least two elements should be evaluated: (1) whether there is a reasonable probability that an intervention could progress to the next stages of research at all; and (2) whether the trial is designed such that the yielded results can contribute to progression to the next stage of research (typically Phase II). The assessment of estimated efficacy can contribute to the assessment of both elements. Evaluators should therefore judge whether they find the estimated efficacy as presented by investigators to be substantive.

For trials to have translational value they should be hypothesis-driven. Preclinical and reference class evidence form the basis for the generation of hypotheses and the context for the subsequent interpretation of both positive and negative findings.⁴⁷ For instance, if a positive result in animals is followed by a negative result in humans, this difference can lead to further explorations of this difference and/or which modifications to the intervention have to be made to overcome translational hurdles. Furthermore, the determination and evaluation of reference class evidence helps researchers to put their findings in a broader context and to communicate their findings to other areas of research. Evaluators should thus judge whether investigators base their hypotheses on a solid assessment of preclinical and reference class evidence.⁴⁸

4.6.2 Benefit Evaluation

We contend that investigators and evaluators should be transparent about the weight they ascribe to the different types of benefits (and harms). Progressive and translational value are not necessarily mutually exclusive, however, they may require a different trial design.⁴⁹ Therefore, it should be made explicit how a trade-off between different types of benefits and harms are made.

4.6.3 Benefit Treatment

After benefit assessment, RECs need to judge whether measures need to be implemented to modify – and ideally to maximise – benefits. The following measures can be taken to enhance the translational value of a trial. If hypotheses are insufficiently supported by evidence, investigators can be prompted to conduct additional preclinical testing. Alternatively, evaluators can demand more thorough gathering and assessment of existing preclinical and reference class evidence. Methods of PPI can show whether or not patient-relevant outcome measures have been used. Furthermore, open sharing of the assessed preclinical and reference class evidence can enhance the collateral value of a trial. Additionally, amendments to the trial design can spur the translational value.

4.6.4 Decision-Making

Finally, RECs have to decide whether benefits truly outweigh the risks. The three steps of benefit analysis, evaluation and treatment contribute to the transparency of decision-making. It has been claimed that it matters whether the research is funded with public money or not. We disagree: even when privately funded, we can see no justification for burdening participants with research that has no social value.

4.7 Conclusion

The term ‘social value’ strikes the necessary balance between scientific advancement, equitably responding to human conditions and realising the human right to health. The requirement of social value bridges the gap between conducting commendable science and making a contribution to the health of the populations where health research is being carried out. The concept of social value is the ethical justification for doing health research involving humans.

5 Solidarity in Health Research Regulation

Katharina Kieslich and Barbara Prainsack

5.1 Introduction

This chapter explores the analytical and normative roles that solidarity can play when designing health research regulation (HRR) regimes. It provides an introduction to the meanings and practical applications of solidarity, followed by a description of the role solidarity plays in HRR, especially in fostering practices of mutual support between patient organisations and between countries. We illustrate our argument in a case study of HRR, namely the European Union (EU) regulatory regime for research on rare diseases and orphan drugs. The current regime aims to decrease barriers to research on orphan drugs by creating, predominantly financial, incentives for research institutions to take on the perceived increased risks in this area. We show how the concept of solidarity can be used to reframe the purpose of regulation of research on orphan drugs from a market failure problem to a societal challenge in which the nature of barriers is not just financial. This has specific implications for the types of policy instruments chosen to address the problem. Solidarity can be used to highlight the political, social, economic and research value of supporting research on rare diseases and orphan drugs.

5.2 The Meaning of Solidarity

The concept of solidarity underpins many social and healthcare systems in Europe.¹ While it could be argued that solidarity – in the form of policies and institutional structures facilitating mutual support, with special emphasis on supporting the vulnerable – has come under pressure with the spread of nativist and other sectarian political ideologies, there are also forceful counter-movements under way. These include people standing up with and for others,² may it be newcomers to our society, victims of wars and natural disasters or people who suffer from our economic and political system. As such, it is fair to say that solidarity is seen by many as having a lot to offer to how we frame and address societal challenges.

What is solidarity? At first sight, it might seem an elusive concept. For decades, solidarity has been used to justify a wide variety of policies and practices ranging from vaccination programmes to biobanks to the penalisation of undesirable behaviours. Another reason for the elusiveness of solidarity lies in the practical and embodied nature of solidarity. Solidarity is, first and foremost, a relational practice: its full meaning unfolds only when it is enacted, in concrete practice, by – at least one – giver and a receiver, and its nature cannot be exhaustively captured by language. For the same reason that poetry, art or nature are so much more powerful in conveying the meaning of love or friendship, words alone struggle to convey the full meaning of solidarity.

Acknowledging that part of the meaning of solidarity resides in its embodied- and enactedness does not mean, however, that we cannot spell out what makes solidarity different from other types of prosocial practice. Building upon a long history of scholarship on solidarity we have, in our own work, proposed that solidarity is best understood as a practice that reflects a person’s – or persons’ – commitments to support others with whom the person(s) recognise(s) similarity in a relevant respect.³ The similarities with others that people recognise are, however, not ‘objectively’ existing properties, but they are characteristics that we have learned to attribute to ourselves and to others. The first step in this process is that we use categories that have been developed to sort people in different groups, such as separating them into women and men, children and adults, Jews, Buddhists and Muslims, or Koreans and Croatians. While these categories clearly have an expression in material reality, such as the correspondence of national labels with specific territories, or – in the case of children and adults, even stages in human biology – these categories are not merely material. To whom the label of ‘Korean’ or ‘Croatian’ is applied has not been stable in history but it has depended on changing territorial rule, changing understandings of nationality and different perspectives on who can legitimately claim belonging to such a label. Similarly, the notions of children and adults are not clearly delineated in biology in the sense that every person neatly fits into one or the other category. In this way, the categories that we use to describe characteristics that we and others hold are lenses through which we have learned to see reality.

For solidarity this means that when a woman supports another person because she recognises her as a fellow woman, then ‘being a woman’ is the ‘similarity in a relevant respect’ that gives rise to solidaristic action – despite the fact that the two people in question are many more things than women. They may be different in almost every other way. In this sense, the recognition of similarities in a relevant respect is a subjective process – I recognise something in you that you may not recognise in yourself because you have not learned to see it. At the same time it concerns shared social meaning – as societies have shared conventions about how they classify people.

Solidarity happens when people are guided in their practices by the similarities they recognise with each other, despite everything that sets them apart. It is the similarities, and not the differences, that give rise to action in the sense that they prompt people to do something to support somebody else. This ‘doing something’ could consist of something big – such as donating an organ – or something small, such as offering somebody a seat on a bus.

In sum, what makes solidarity different from other pro-social practice is the symmetry between people in the moment of enacting solidarity. This symmetry is not an essentialist ontological statement that glosses over claimed or ascribed differences and structural inequalities. Instead, it is the description of a relational state in the moment of enacting solidarity. In this way, solidarity is distinct from other pro-social supportive behaviours such as cooperation and charity, for example. The notion of cooperation describes pro-social supportive behaviour without saying anything about how and why people engage in it. The notion of charity describes an asymmetrical interaction between a stronger entity giving something and a weaker entity receiving something. In contrast, solidarity refers to entities that are different in many respects but make the thing they share in common the feature upon which they act: I do something for you because I recognise you as a fellow woman, a co-worker who struggles to make ends meet, as I do, or a fellow human in need of help.

5.3 The Three Tiers of Solidarity: Applicability and Adjustments in the Context of Health Research Regulation

Having defined solidarity as practices that reflect commitments to support others with whom a person – or persons – recognise(s) similarities in a relevant respect, in previous work one of us identified three main tiers of solidarity, capturing the societal levels where solidaristic practice takes place.⁴ Tier 1 is the interpersonal level where solidarity is practised between two or more people without that practice having become more widespread. An example from the field of health research would be a person with diabetes signing up to a biobank researching the disease because she wants to support others with similar health problems.

If this practice were to become more widespread, so that it became common or even normal behaviour within a group, then we speak of solidarity at Tier 2 solidarity, which is solidarity at the group level. The group within which solidarity is practised could be a pre-existing group – such as a self-help group around diabetes where it becomes normal practice, for example, to also volunteer for disease research – or a group that is created through the solidaristic practice itself. An example for the latter would be a patients’ rights organisation created in response to the effects of harmful medical practices such as the blood contamination scandal in the 1970s and 1980s in the United Kingdom (UK).

If solidaristic practices become so commonplace that they are reflected in legal, administrative or bureaucratic norms, then we speak of Tier 3 solidarity. This is the ‘hardest’ form of solidarity because it has coagulated into enforceable norms. Tier 3 solidarity could be seen to contradict the idea held by many scholars in the field that solidarity cannot be demanded, but only appealed to.⁵ In this understanding, contractual and legal obligations are incompatible with solidarity. While we agree with these authors that solidarity is typically a more informal, voluntary ‘glue’ between the bricks of formal institutional arrangements, we also believe solidarity to be a toothless, if not empty, concept if it cannot also denote practices that are so deeply engrained in society that they become legally enforceable in some cases.

Ruud ter Meulen and colleagues very helpfully distinguish between solidarity as a community value and solidarity as a system value:⁶ the latter can contain articulations of solidarity in formal, often legal arrangements. The key here is to consider enforceable – and thus not always voluntary – solidarity in conjunction with more informal, voluntary forms of solidarity, and not see them as isolated from one another. An example would be tax or contribution-based financing of universal healthcare where those with higher incomes contribute more than others.

A problem arises when legally enforceable solidarity is still in place while the actual practices that used to underpin them are breaking away. This is becoming apparent at the moment in many countries where certain features of welfare states, such as transfer payments in the form of as child allowances or income support for those considered undeserving, have come under attack. The argument is often that the people benefitting from this are ‘free riders’ as they have not contributed towards the system that they are now using – perhaps because they are new immigrants or people who have never been in paid employment. What is happening here is that the basis for solidaristic practice – namely the ‘recognition of similarity in a similar respect’ (see above) – is breaking away. The people who are receiving financial support, or benefitting from a solidaristic healthcare system, are no longer seen as belonging to ‘us’ – because of something that they supposedly did, or failed to do, or because they do not have the same passport as we do.

While it will often be the case that solidarity prescribed at Tier 3, in the form of legal, contractual, bureaucratic and administrative norms, will have evolved out of solidarity practised at group (Tier 2) and interpersonal (Tier 1) levels, the reverse is not necessarily true: interpersonal solidarity can, but does not necessarily, scale upwards. The ‘higher’ the level of solidarity, the more important reciprocity becomes. Here we refer not to direct reciprocity, where one gives something in return for something else – this would be a business transaction instead of solidaristic practice – but indirect, systemic reciprocity. Institutional arrangements of solidarity work best when people give because they want to support others, but they also know that when they are in need they will be supported as well.

5.4 Solidarity in Health Research Regulation

How do the aforementioned conceptualisations of solidarity apply to HRR regimes? The first aspect we need to acknowledge is that HRR regimes are complex and varied. There is no such thing as one regime that applies to all areas of HRR, but rather there are multiple and sometimes overlapping legal and ethical requirements that need to be fulfilled by those planning, funding, supporting and undertaking research. HRR is a multidisciplinary endeavour that involves different actors such as policymakers, researchers, health professionals, industry and patients. HRR also spans a large variety of ‘objects’ that are regulated, such as data, tissue, embryos, devices or clinical trials.⁷ This means that it occupies regulatory spaces beyond health, such as in data regulation, research financing, in fostering innovation and in the obligation to protect research recruitees.

At the start of this chapter we suggested that solidarity can be thought of as ‘enacted commitments to accept costs to assist others with whom a person or persons recognise similarity in one relevant respect’.⁸ Thus the question arises: what are the shared practices that reflect a commitment to carry costs – emotional, financial, societal – in HRR, and what are the similarities that give rise to these practices? The two tiers of solidarity most relevant in HRR are Tiers 2 and 3. Tier 2, or group solidarity, is reflected, for example, in the way patients, patient groups and other stakeholders advocate for, inform about, and partake in research endeavours and regulatory steps to make them happen. The question of who partakes in research is not just important for methodological reasons but is also connected to the concept of solidarity. It is considered good scientific practice to carry out research in the populations for whom an intervention is intended, but there may be instances in which it is justified to conduct research in populations other than the intended beneficiaries. According to the Council for International Organizations of Medical Sciences (CIOMS) and the World Health Organization (WHO) such instances are ‘important demonstration[s] of solidarity with burdened populations’,⁹ for example in 2014 when Ebola vaccines were tested in communities not affected by the Ebola outbreak.

The costs and the similarities that are at the heart of these – predominantly clinical – research practices are comparatively easy to identify. The costs commonly consist of individuals giving up their time to become research participants or to become involved in a patient advocacy group. They accept the burden of cumbersome regulatory steps to partake in research, such as navigating consent forms, risk assessments, data ownership and other issues. The similarity that motivates people to assist others despite the costs they incur is often the experience of suffering from a particular disease or the acknowledgement that we, as members of society or those close to us, all run the potential risk of illness in the future. It is a recognition that temporary sacrifices can result in long-term gains from the generation of new knowledge about health conditions and treatments.

A feature that distinguishes HRR from other areas of policy, regulatory and societal processes is that group solidarity is often not just confined to a small group of patients who are afflicted by the same illness. Rather, other members of the public – so-called healthy recruits – partake in the solidaristic practice of research and are directly affected by the associated regulatory procedures. The underlying ‘similarity in a relevant respect’ that, in Prainsack and Buyx’s definition of solidarity gives rise to solidaristic practice, is then typically a broad sense of human vulnerability that we all have in common. In other words, the nature of Tier 2 solidarity in HRR is not necessarily restricted to suffering from the same illness, but it can arise from the recognition that in a universally funded healthcare system, we all carry a commitment to carry costs because we all carry the risk that we might one day become ill.

To explore how Tier 3 solidarity, or institutional solidarity, is reflected in HRR, we trace the logic that forms the basis for understanding HRR through the lens of solidarity. The logic runs something like this: A solidaristically financed healthcare system is built on the principles of fair access to healthcare, protection against financial risks due to illness and quality. Ensuring access, provision and high-quality healthcare requires efforts to advance knowledge through research. Implicitly entailed in the social contract between governments, citizens and residents is the acceptance that mandatory financial contributions – i.e. costs – in the form of taxes or health insurance contributions will not only be used for the day-to-day provision of services but also for the fostering of research activities. With this implicit acceptance of carrying costs collectively comes a recognition that the health research area needs to be regulated to safeguard against unethical, harmful, and wasteful practices, and to foster innovation. This recognition translates into public policies that regulate the field.

But there are also regulatory burdens arising from such public policies that might negatively affect solidarisic practices in HRR. For example, the cumbersome, and often time-intensive, process of giving consent for a research participant’s data to be used for research purposes might deter some people from taking part in a study, especially if the use of the data is not explained or communicated clearly. Moreover, the predominant lens through which data ownership – in a moral and in a legal sense – is currently viewed is that of the rights of individuals, who, in turn, are conceptualised as bounded and independent entities.¹⁰ This view is problematic because it fails to acknowledge the deeply engrained relational characteristics of data. This is so because the meaning of most data only unfolds once the data is interpreted in relation to other data, and that this meaning is often relevant for a wider range of people than only the person from whom they came. Currently, this relational nature of data is not reflected in most data governance frameworks in the health domain; even those frameworks that give people more control over how their data is used typically give this control to individuals. Instruments of collective control and shared ownership of personal data are rare. The ‘individualisation’ of data governance sits squarely within a system that relies on people’s willingness to make data about themselves available for research. It is a missed opportunity for showing how control and use of data can reflect both personal and collective interests and rights.

5.5 Solidarity in Research on Rare Diseases and Orphan Drugs

An example of how solidarity can be used to change the way we approach a policy problem in HRR can be found in rare diseases and orphan drugs research. The European Commission (EC) defines a rare disease as ‘any disease affecting fewer than 5 people in 10,000 in the EU’.¹¹ It estimates that there are approximately 5,000–8,000 rare diseases in the world. The challenge around rare diseases is that the comparatively small numbers of people affected by them translate into the neglect or the unavailability of diagnoses and treatment options. It can be explained by drawing on the notion of issue characteristics, famously developed by political scientist Theodore Lowi.¹² Lowi posited that different types of policies – e.g. regulatory, distributive or redistributive policies – give rise to different policymaking or decision-making processes through which distinct patterns of political and societal relationships and behaviours emerge. Just as the categories we use to describe characteristics that we hold – women and men, adults and children, Koreans or Croatians – we can use categories to describe characteristics that policies or policy fields hold. For example, the depiction of European healthcare and welfare systems as solidaristic has arisen from their embeddedness in redistributive policies that allow the state to redistribute taxes and other welfare contributions in the pursuit of policy goals. Different types of policies give rise to different forms of state action, but also to different types of public participation, or even political controversy and contestation. The latter is what we frequently observe when a change in redistributive policies is suggested. Following Lowi’s rationale, the key to understanding patterns of behaviours, in this case the lack of attention given to rare diseases, is to identify the characteristics of the issues to which they give rise. The more complex the regulatory or policy area, the more difficult it is to develop policy solutions.

The issue characteristics for rare diseases are complex. We know relatively little about the factors and processes that underlie these diseases. This stems from a lack of basic research into rare diseases¹³ which is mostly due to a lack of available funding for research that a relatively small number of people suffer from. From a public policy perspective, the question of how and if to prioritise research for rare diseases is an intrinsically complex issue because of the low numbers of patients and the high costs for research and treatment. It begs the (redistributive) policy question how spending a large proportion of overall research or healthcare budgets on a few patients can be justified if the opportunity costs are such that other patients may lose out as a result. The low patient numbers also result in difficulties in the design of clinical trials that meet the evidentiary hurdles of most regulatory agencies in Europe.¹⁴

Solidarity offers a lens through which these difficult questions surrounding research on rare diseases can be reframed. Patients suffering from rare diseases are characteristically vulnerable (please see Rogers’ Chapter 1 in this volume for more detail on the concept of vulnerability). Their vulnerability results from the severity and the chronicity of their conditions, the inadequate access to appropriate diagnoses and treatment options, societal isolation and a lack of representation of their interests.¹⁵ Coming back to the importance of Tier 3 solidarity in HRR (the institutional and legal level), the solidaristic principles upon which healthcare systems in Europe rest suggest a duty to care for society’s most vulnerable members, which patients with rare diseases undoubtedly are. Policies or regulations to support research and service provision for patients with rare diseases can therefore be viewed as solidaristic practices.

However, despite initiatives such as the introduction of Regulation (EC) 141/2000 on orphan medical products, access to adequate services and research for patients is still falling short of expectations. Following Lowi’s approach, as outlined above, we can observe that the more complicated the issues to which a regulatory or policy area give rise, the less policymakers are inclined to act because of the perceived lack of policy options. This might also explain why the challenges around fostering research activity on rare diseases are predominantly framed as a regulatory policy problem rather than a distributive or redistributive one. Interestingly, the perceived lack of policy options and responses corresponds with a flourishing of solidaristic practices below the level of public policy that span borders and countries at the EU level. For example, there seems to be an emerging recognition of ‘similarity in a relevant respect’ among EU countries in the sense that the issue characteristics of rare diseases are such that no country can stem the challenge of protecting vulnerable patients suffering from rare diseases on its own. Here, Tier 2 solidarity does not just apply to the level of interaction and collaboration among patient groups, but also to the level of cooperation between nation states. The similarity is the recognition that all countries face the same challenge in finding adequate research and treatments on rare diseases – the policy problem – and that countries are similar in their failure to find policy solutions. This can lead to the fostering of solidaristic practices such as the EC’s advocacy for a European Platform on Rare Diseases Registration that would bring together patient registries and databases to encourage and simplify clinical research in the area.

An unresolved question in the application of a solidarity-based approach to the field of HRR is the role of industry, especially in fostering or hindering solidaristic practices. It is frequently argued that pharmaceutical manufacturers do not invest enough resources into the research and development of rare diseases and orphan drugs because the small patient numbers lead to a low return on investment (RoI).¹⁶ The response of EU member states has been to create incentives through policy instruments such as fee waivers for regulatory procedures or a 10-year market exclusivity for authorised products.¹⁷ The introduction of such measures in the Regulation (EC) 141/2000 on orphan medical products has increased the number of orphan drugs being authorised. But is it also a sign that pharmaceutical industries are engaging in solidaristic practices to benefit some of the most vulnerable patients?

We argue that it is not. We must assume that pharmaceutical companies are motivated by the incentives offered through this regulation rather than a recognition of similarity with entities that seek to promote public benefit, or with people suffering from illness. The perception that some people, as taxpayers or patients, are expected to contribute to supporting others who suffer from rare diseases, while some corporate actors do the bare minimum required by law, may have a significant negative effect on the people of other actors to contribute. This may be exacerbated by the payment by corporations of hefty dividends to their shareholders. Institutionalised solidarity requires some level of reciprocity – the understanding that each actor makes a contribution adequate to their nature and ability. As a result, if large multinational companies are seen to get away with ‘picking the raisins’ this is a serious impediment to solidarity.

In a field that is still very dependent on the investment of pharmaceutical companies into drug research, resolving this challenge of asymmetry is not easy to rectify in the short term. Its solution would require legislation that forces companies to cut their profits and support rare disease patients in more significant ways than they are doing at present. A for-profit company cannot reasonably be expected to be motivated by the desire to help people; it is to be expected, and justified, that they put profits first. This is why it is the role and responsibility of legislators to ensure that companies are contributing their fair share. This is not only a necessity for moral and ethical reasons, but also to avoid the hollowing out of solidaristic practices among people who may, as argued above, be deterred by the expectation to accept costs to help others, while others are making huge profits.

The concept of solidarity can and should be used to reframe the regulation of research on orphan drugs from a market failure problem that requires financial incentives, to a societal problem that requires more than market measures. This will require a reframing of the issue as a redistributive policy problem rather than a purely regulatory one, in the hope that this will instigate political debates, as well as patient and public participation that would help bring the challenges of research on rare diseases and orphan diseases more to the centre of the policy process. Using the concept of solidarity to help reframe the policy issue has the potential to draw it out of the comparatively confined policy spaces it currently occupies. This helps to illuminate its political and public salience. The joined-up working of patient groups for rare diseases and the mutual efforts of EU member states – also as regulators that impose rules of fair play on pharmaceutical companies – are needed to facilitate – and where they already exist, stabilise – solidaristic practices. To make these practices more powerful and meaningful, priority-setting mechanisms for the prioritisation of research funding need to be developed,¹⁸ and more public money should be invested, especially into basic research, in an effort to decrease the dependence on the pharmaceutical industry.

5.6 Conclusion

In this chapter, we have used research on rare diseases and orphan drugs to highlight the application of solidarity to HRR. It is an example of a space where solidaristic practices are already taking place, but also illustrates that there is room for improvement. Solidarity is an integral part of health research, and it is enacted every time a person takes part in a clinical trial or other research because they want to support the creation of public benefits. Regulation is important to ensure that research is carried out in an ethical manner, but, equally, it is important that decision-makers who define the regulatory spaces for HRR recognise the need to support solidaristic practices rather than undermine them through overly cumbersome bureaucratic hurdles to enrol in research.

6 The Public Interest

Annie Sorbie

6.1 Introduction

This chapter provides an introduction to the concept of ‘the public interest’ in health research regulation (HRR). It considers two key ways that the public interest is constructed in HRR: namely as a legal device and through empirical evidence of the views of publics. To appreciate the scope of this concept, the public interest is set in its broader context, i.e. beyond HRR, highlighting that, historically, it has been a contested concept that is difficult to define in the abstract. Next, the public interest is situated within HRR, paying attention first to how it features in the HRR legal landscape and then how this is constructed through the views of publics (with specific reference to the use of identifiable health data for research). Both conceptualisations are analysed with reference to the key challenges and opportunities that they present before a holistic concept of the public interest in HRR is proposed and consideration given to how this may be operationalised in practice.

6.2 The Public Interest: A Contested Concept

Although the public interest is fully embedded in HRR, it is by no means exclusive to this context. The following brief consideration of wider perspectives on this contested concept point to persistent debates not only on what the public interest ‘is’, but also to tensions as to how this concept should be understood. Appeals have been made variously to the values it invokes, the process it requires, and/or the views of (some or all) of ‘society’ at large that it reflects.¹

Political and social scientists, philosophers and lawyers, among other disciplines, have contemplated this elusive concept without reaching consensus on its meaning or usefulness. During a period of scholarly interest in the public interest in post-World War II America, it was both lauded as ‘a central concept of a civilised polity’² and dismissed as a concept so vague and ambiguous that it is no more than a rhetorical device.³ This ambivalence can be seen in Sorauf’s work in which, despite his scepticism, he initially concedes a ‘modest conception’ of the public interest that is rooted in ‘our interest in the democratic method and its settlement of conflict by orderly rules and procedures’.⁴ He recognises too the potential function of the public interest as a ‘hair shirt’ that serves as ‘an uncomfortable and persistent reminder of the unorganized and unrepresented (or underrepresented) interests of politics’.⁵ Over time, however, his position hardens and becomes more negative. He later posits that the public interest promotes ‘oversimplification’, as it purports to “solve” the dilemmas of … pluralism’.⁶ Turning to the regulatory role of the public interest, Feintuck also points to a continued reluctance to define the public interest beyond what ‘will vary according to time, place and the specific values held by a particular society’.⁷ He characterises the public interest as an ‘empty vessel’ and argues for an account that looks ‘to the fundamental value laden, democratic imperatives that underlie society: human dignity, parity of esteem, and the ability to participate actively in society’.⁸

Whether the public interest is best understood modestly as a procedural mechanism, ambitiously as protecting fundamental values in society including those that may otherwise be overlooked, or in utilitarian terms as the views of the majority, there is little doubt that this is a contested concept that is ‘much used but ill defined’.⁹ This chapter proposes that while there is need for further conceptual clarity here, there is also value to be found in such contestation and flexibility.

6.3 Appeals to the Public Interest in HRR

In HRR, the concept of the public interest is embedded in law and in policy, often as a counterpoint to individual interests. In medical research involving human subjects – including research on identifiable human tissue and data – consideration of the relationship between individual and public interests can be traced back to the original Declaration of Helsinki.¹⁰ More recently, the legal mandate of the Health Research Authority (HRA) in the United Kingdom, as set out in the Care Act 2014, prescribes twin objectives to protect and promote the interests of both individual participants (and potential participants) and the interests of wider publics in safe and ethical health and social care research.¹¹

However, reflecting the broader literature on public interest, Taylor notes in his consideration of genetic data and the law, that the public interest remains a ‘notoriously uncertain idea’.¹² This chapter proceeds with an account of two key ways in which the concept of the public interest appears in HRR (with a focus on the use of identifiable health data for research), as constructed in law and through publics’ views. It considers the key challenges and opportunities presented by the public interest in each framing. Having identified the benefits and shortcomings of each, a holistic concept of the public interest is proposed, the relationship between the public interest as constructed within and beyond the law is examined, and consideration is given to how, in a more concrete way, public interest might be operationalised in HRR practice.

6.4 The Public Interest as Legal Device

When health research is conducted on identifiable personal data, the public interest is a striking feature of the legal landscape. For example, in the realm of data protection, the public interest forms one of the routes to the lawful processing of personal data in health and social care research. Thus, the General Data Protection Regulation¹³ (GDPR) provides a lawful basis to process personal data where this is a ‘task in the public interest’.¹⁴ Health Research Authority (HRA) guidance confirms that, for the purposes of the GDPR, this is the appropriate legal basis that should be used by public authorities, such as NHS bodies or universities, in order to process data for health and social care research.¹⁵ In UK law, the Data Protection Act 2018¹⁶ (DPA 2018) purports to add further detail to the interpretation of ‘a task in the public interest’, although concerns have been raised that the drafting of this legislation does little to add clarity to how this concept should be understood in practice.¹⁷ A late addition to the Explanatory Note to the Act indicates, by way of an example, that ‘a university undertaking processing of personal data necessary for medical research purposes in the public interest should be able to rely on [a task in the public interest]’¹⁸, thus providing some guidance on the context, if not the content, of the public interest in these circumstances.

Two other prominent features of the health data legal landscape are: (i) the common law duty of confidentiality and (ii) the legislative regime which established the predecessor body to the HRA’s Confidentiality Advisory Group (CAG). The common law duty of confidentiality provides that where confidential information is imparted to another person, in circumstances giving rise to an obligation of confidentiality, this must not be disclosed without consent or justification.¹⁹ One such justification is where disclosure is ‘in the public interest’. This duty, and its exceptions, apply not only in the context of the traditional doctor/patient relationship, but also where it is proposed that the information in question may be used for purposes beyond direct care, such as for health or social care research. The interpretation of this duty of confidentiality (and, importantly for this chapter, the meaning of the public interest) has emerged as a result of decisions made on the facts of cases that have come before the courts. These judgements indicate, for example, that there is not only a personal interest in an individual’s confidentiality being maintained, but also a wider public interest in doing so in order that patients (in general) are not discouraged from consulting with healthcare practitioners.²⁰ Case law, in relation to whether disclosure of deceased patients’ records to a public inquiry was in the public interest,²¹ recognises that the public interest (which was distinguished from ‘what the public found interesting’)²² is multifaceted and can encompass both individual and collective interests. These include interests in: disclosure, maintaining the patient’s confidentiality and maintaining confidence in the institutions under investigation.²³

As with the legislation, there is no fixed definition of the public interest in case law; where this lies must be decided on the individual facts of each scenario. This perception of a lack of certainty led to concerns from some clinicians that routine activities, such as providing information to registries that collect and analyse data on specific diseases, might be vulnerable to challenge in the absence of specific consent.²⁴ These worries about the legality of such practices, among other matters, led to the enactment of legislation in England and Wales in 2001 that forms another key feature of the data sharing landscape, namely the establishment of the predecessor to the CAG. In summary, this legislation allows the Secretary of State for Health to make regulations to explicitly ‘set aside’ the common law duty of confidentiality for defined medical purposes, including medical research, where this is ‘in the interests of improving patient care, or in the public interest’. These powers are now found in Section 251 of the NHS Act 2006 (as enabled by the Health Service (Control of Patient Information) Regulations 2002) and referred to colloquially as ‘s251 support’. In sum: where seeking consent is neither possible nor practical, researchers can obtain s251 support to use confidential patient information for medical research by make an application to the HRA’s CAG. The effect of such an application is that, if granted, the researcher need not be concerned whether (in the admittedly unlikely event of litigation) a court would agree that their use of identifiable patient information without consent was indeed in the public interest.

In common with the broader literature on the public interest, the preceding whistle-stop tour of the public interest in law reveals anxieties around how this concept is interpreted in practice. It also speaks to the strengths and limitations of a narrow legal construction of the public interest decided on a case-by-case basis, but for which precedents can be established over time. These are explored further in the passages that follow.

We return first to Taylor’s description of the public interest as a ‘notoriously uncertain idea’.²⁵ It is of note that Parliamentary debate on the DPA 2018²⁶ on this topic resurrected many of the concerns around the public interest that had arisen some fifteen years previously, at the time of the promulgation of the CAG regime. These included the potential for the public interest to be interpreted widely to deliver ‘sweeping powers’.²⁷ Nonetheless the CAG regime, which was first proposed as a temporary solution as the NHS geared up to apply a ‘consent or anonymise’ binary to its use of health data, has become an example of good governance and established itself as part of the data sharing landscape.²⁸ This can be attributed, in part, to a growing recognition from stakeholders in HRR – including researchers and publics – that consent is not necessarily the ‘magic bullet’ to legitimise HRR governance that it might once have been presumed to be. For example, Wellcome’s research, as commissioned from Ipsos MORI, on public attitudes to commercial access to health data for research purposes found that, when considering data uses, ‘a strong case for public benefit is the most important factor for many people: without it, data use by any organisation is rarely acceptable’.²⁹ This tends to suggest that while concerns about the uncertainty of the application of the public interest in HRR persist, it is a concept that also, in some ways, benefits from its inherent flexibility and its ability to adapt to changing interests over time.

A further critique that arises from this legal construction of the public interest is that this looks inwards to derive its legitimacy from its institutional origins and is disconnected from actual publics’ views. For example, in the case of legislation – such as the DPA 2018 and the legislation underpinning the CAG regime – legitimacy comes from Parliament. Notwithstanding, the public interest in (legal) text tells us little about its context. Even when amplified by its Explanatory Note, the DPA 2018 does not elaborate on the legitimate content of the public interest in HRR.

Turning to case law, the public interest is conceptualised by the courts on the facts of each case, following precedents in previous decisions. This inward-looking legal construction of the public interest is consistent with the long established ‘intellectual tradition’³⁰ within the law of invoking fictional persons to provide a barometer of what ‘reasonable’ members of the public would expect in any given situation. The paradigm is the fictional ‘man on the Clapham Omnibus’,³¹ who in English law is deployed to represent the reasonable person. Elsewhere in the law, other fictional reference points include the ‘right-thinking member of society’ (in defamation law) or even the ‘officious bystander’ (in contract law).³² It has thus been confirmed by the Supreme Court that: ‘The spokesman of the fair and reasonable man, who represents after all no more than the anthropomorphic conception of justice, is and must be the court itself’.³³ This underlines why the law historically has not been centrally concerned with empirical evidence of the views of actual members of the public when it deploys the legal notion of the public interest in civil law cases.

However, this legal self-referential conception of the public interest in HRR is increasingly under pressure, as exemplified by the high-profile failure of care.data. As described more fully in this volume by Burgess (Chapter 25), this was an NHS England initiative that sought to make patient data available for specified purposes, including audit and research, in a format that was stripped of identifiable information. However, following widespread concerns about the scheme – including around its transparency and oversight – the programme closed in 2016.³⁴ Here, a legal framework was in place to facilitate data sharing but, as argued by Carter et al.,³⁵ the social licence to do so was not. This failure underlines the message that ‘legal authority does not necessarily command social legitimacy’.³⁶ It follows that where the law alone is unable to fully legitimise and animate the public interest, something else must fill this void. The following section suggests that a richer relationship between this legal concept and the views of publics could be a worthy candidate.

6.5 The Public Interest as the Views of Actual Publics

The potential benefits of responsible access to health data by researchers, as well as the perils of getting this wrong, have led to a renewed focus on the public acceptability of data sharing initiatives and a growing body of literature that explores public attitudes towards sharing health data for research purposes.³⁷ Aitken et al. note the desire of stakeholders in HRR to optimise the use of existing data in health research and: ‘the recognition of the importance of ensuring that data uses align with public interests or preferences’.³⁸ This commitment to using patient data responsibly is shared by funders, as exemplified by Wellcome’s ‘Understanding Patient Data’ initiative, which works to champion responsible uses of data and improve stakeholder engagement around how and why data is used for care and research.³⁹

Consider too the call in HRR for more and better public and patient involvement (PPI). The National Institute for Health Research (NIHR) recently issued ‘Standards for Public Involvement in Research’, which provide ‘a framework for reflecting on and improving the purpose, quality and consistency of public involvement in research’.⁴⁰ In particular, Standard 6 on Governance states that ‘[w]e involve the public in our governance and leadership so that our decisions promote and protect the public interest’. Here, the role of publics is positioned not only as shaping and supporting research, but also as a means of legitimising HRR and grounding the broader public interest.

This approach has the benefit of being anchored to actual publics’ views, something that is lacking from the narrow legal account set out above. In this way, it has the potential to provide at least some of the social legitimacy that was lacking in care.data. However, public engagement activities also attract criticisms of exclusivity and tokenism,⁴¹ raising ‘questions of representativeness, articulation, impacts and outcomes’.⁴² Thus, to simply equate these outputs with ‘the public interest’ more broadly also runs the risk of reinforcing underlying inequalities in the delivery of a majoritarian account of the concept. Reports of instances of ‘personal lobbying by volunteers for pet causes’⁴³ point to the dangers of ‘assuming that the perspectives of a small number of involved patients necessarily reflect the perspectives of a larger patient community’.⁴⁴ Indeed, McCoy et al.’s analysis of the recent NIHR ‘Standards for Public Involvement’ suggests that ‘it is simplistic to assume that including public representatives on governance and leadership bodies will necessarily promote the public interest’.⁴⁵ They highlight the likelihood that the interests of differing ‘publics’ will, in any event, diverge, and call for more attention to be paid to who is being asked to contribute, at what stage in a research project, and for what purpose.

This is not, of course, to discount the important contributions that can be made to shaping and delivering responsible HRR through the thoughtful involvement of patients and wider publics.⁴⁶ However, whereas it is advanced above that the law alone is not enough to legitimise the public interest, this analysis also suggests that an additive approach to publics’ views in HRR is also insufficient to provide a lasting and justifiable account of this concept. Something more is required.

6.6 The Public Interest: A Holistic Concept

Taken together, the preceding examples illustrate the prevalence of the public interest in HRR and how this concept may be constructed both through the law and through the views of publics. On the one hand, the tendency of the law to approach the public interest as a legal test draws the criticism that this narrow notion of what purports to be in the public interest is wholly disconnected from the views of publics and can lack social legitimacy. On the other, to claim that the public interest can simply be extrapolated from the outputs of public involvement work is equally problematic. Nonetheless, despite this disjuncture, common themes emerge and, in this section, two further contributions to the debate on the role of the public interest are offered. The first is a proposal for a holistic concept of the public interest that is able to account for a plurality of interests and views. The second is that, despite the apparent impasse, legal and empirical notions of the public interest are not mutually exclusive. It is argued that these do bear upon one another and that if the public interest is to be effectively deployed in HRR, this relationship should be both acknowledged and made more overt.

The first proposal is to recognise that both the legal and empirical constructions of the public interest call for a conception of the public interest that is able to account for a range of diverse interests. In law, the potential for this approach is evident in an arc of case law that emphasises that the public interest is a multifaceted and flexible concept that is able to account for both individual and collective interests, including wider publics and institutional stakeholders. Similarly, the analysis above suggests that the value of public involvement is optimised when attention is paid to the multiple interests of differing patients and publics, including who is being asked to contribute, when, and for what purpose. This also tracks a move in HRR literature away from a narrow account of the public interest that pits individual interests against collective benefits. For example, Rid describes this ‘pluralistic conception of public interest’ as an account that is capable of recognising that multiple interests are in play.⁴⁷ Taylor’s work also proposes that individual and public interests need not be balanced against one another, but rather that the need for legitimacy requires that each should account for each other.⁴⁸ Together, this forms the basis for a holistic concept of the public interest in HRR that is able to account for multiple interests and views. This approach does not, in the words of Sorauf, aim to ‘solve’ pluralism. Quite the opposite: it embraces the messy realities and subjectivities, both of the law, as broadly conceived, and of outputs from public involvement activities.⁴⁹

The second contribution is to suggest that, despite the messiness, these accounts are not mutually exclusive and do, in fact, bear upon one another (though this relationship is far from clear). For example, I suggested earlier that shifting public views on health data sharing (and a move away from a ‘consent or anonymise’ binary) have contributed to the longevity of the CAG, which was originally proposed only as a temporary measure. Similarly, I have referred to how lobbying from the HRR community during the promulgation of the DPA 2018 led to an amendment of the Explanatory Note to clarify that ‘a task in the public interest’ is an appropriate route for public authorities such as universities to use when processing health data for research purposes. Lessons from care.data exemplify the importance of ‘social licence’ to the success of otherwise legal data sharing initiatives. In turn, there is an on-going need for deeper understanding of public acceptability to realise the potential of new and novel uses of health data.⁵⁰ Given the impetus to deliver clear and transparent governance of health data, it is proposed that this this relationship ought to be both acknowledged and made more overt, in order that it may be exposed to debate in HRR. Three concrete suggestions are made in this regard. The first is that the public interest, along with other concepts that operate at the intersection of public involvement and governance in HRR, should be examined to identify their potential to bridge the divide between the outputs from public engagement and the implementation of these in practice. The second is that initiatives such as CAG, where there is ‘evidence’ of the public interest being given effect to facilitate responsible HRR, should be further mobilised. The third is that instances where appeals to the public interest are made in HRR should be captured and articulated publicly, in order to promote transparency and accountability around how and why these have (or indeed have not) been justified.

6.7 Concluding Remarks

This chapter advocates for a holistic conception of the public interest, where interests are accounted for, rather than polarised. HRR governance has moved on from a ‘consent or anonymise’ binary and now needs novel and bold mechanisms that do not seek to over-play the role of legal mechanisms, nor suggest that public views alone can deliver good governance solutions. While the concept of the public interest remains contested and highly contextual, there is an increasing drive towards maximising the potential of this embedded concept in order to deliver a step-change in HRR.

7 Privacy

David Townend

7.1 Introduction: The Modern Difficulty¹

Privacy is a well-established element of the governance and narrative of modern society. In research, it is a mainstay of good and best practice; major research initiatives all speak of safeguarding participants’ rights and ensuring ‘privacy protecting’ processing of personal data. However, while privacy protection is pervasive in modern society and is at the conceptual heart of human rights, it remains nebulous in character. For researchers who engage with people in their studies, the need to respect privacy is obvious, yet how to do so is less so. This chapter offers first an explanation of why privacy is a difficult concept to express, how the law approaches the concept and how it might be explored as a broader normative concept that can be operationalised by researchers. In that wider scheme, I show how individuals respond to the same privacy situation in different ways – that we have a range of privacy sensitivities. I think about four privacy elements in the law: human rights, privacy in legal theory, personal data protection and consent. Finally, I consider how law participates in the broader normative understanding of property as the private life lived in society.

7.2 Privacy as a Normative Difficulty

A good starting point is to ask: what do we mean when we talk about ‘privacy’? It would be difficult for a modern research project to suggest that it was not ‘privacy respecting’ or ‘privacy preserving’. However, the concept is somewhat ill-defined, and that claim to be privacy respecting or preserving might, in reality, add little to the protection of individuals. In part, this problem stems from the colloquial, cultural aspect of the concept: we each have our own idea of what constitutes our privacy – our private space.

Imagine setting up a new data sharing project. You hypothesise that linking data that different institutions already gather could address a modern health problem – say, the growth of obesity and type 2 diabetes. Such data, current and historical, could be used by machine learning to create and continuously revise algorithms to help identify and ‘nudge’ those at risk of developing the condition or disease. The already-gathered data could be from general practitioners and hospitals, supermarkets and banks, gym memberships, and health and lifestyle apps on smart phones, watches and other ‘wearables’. But how would individuals’ privacy be protected within such a project? Many will be uneasy about such data being stored in the first place, let alone retaining it and linking it for this purpose. Many will see that there might be a benefit, but would want to be convinced of technical safeguards before opting into such a project. Many will be happy, having ‘nothing to hide’ and seeing the benefits for their health through such an app. Some would see this initiative as socially desirable, as part, perhaps, of one’s general duty and the basis of personalised medicine, so that such processing would be a compulsory part of registration for healthcare; an in-kind payment to the healthcare system alongside financial payments, necessary for the continued development of modern healthcare that is a general societal and personal good.

Our difficulty is that each one of the people taking these different positions would see their response as a ‘privacy preserving’ stance.² As explored elsewhere in this volume, this observation underlines the diversity of ‘publics’ and their views (see Aitken and Cunningham Burley, Chapter 11, and Burgess, Chapter 25, in this volume). Under the label ‘privacy’ there is a wide spectrum of conceptualisations, from the enthusiastic adopter and compulsion for all, through allowing people to opt-out, generally leaving participation to opting in, to wanting nothing to do with such projects. How then can a researcher frame a ‘privacy’ policy for their research? Are we creating the problem by using the term ‘privacy’ informally and colloquially? Does the law provide a definition of the term that avoids or militates against the problem?

7.3 Privacy as a Human Right

A logical starting point might be human rights law. Privacy and the right to respect for private life is enshrined in human rights law. Unfortunately, it does not give much assistance in the definition of those rights. Two examples show the common problem clearly.³ Article 12 of the Universal Declaration of Human Rights states:

No one shall be subjected to arbitrary interference with his privacy, family, home or correspondence, nor to attacks upon his honour and reputation. Everyone has the right to the protection of the law against such interference or attacks.⁴

Article 8 of the European Convention on Human Rights creates the right in this way:

1. 1. Everyone has the right to respect for his private and family life, his home and his correspondence.

2. 2. There shall be no interference by a public authority with the exercise of this right except such as is in accordance with the law and is necessary in a democratic society in the interests of national security, public safety or the economic well-being of the country, for the prevention of disorder or crime, for the protection of health or morals, or for the protection of the rights and freedoms of others.⁵

Two observations can be made about these ‘privacy’ rights: (1) privacy is not an absolute right, i.e. there are always exceptions and (2) ‘privacy’ and ‘respect for private life’ require a great deal of further definition to make them operational. As to the first observation, the rights are held in relation to the competing rights of others: a right against ‘arbitrary interference’ and ‘no interference … except such’. The concepts of privacy in human rights legislation acknowledge that the rights are held in balance between members of society; privacy is not absolute, because on occasion one has to give way to the needs of others.

As to the content of privacy – and reflecting the broad conceptualisations in the research project example above – we see that what is available from the human right to privacy is international recognition of a space where an individual can exist, free from the demands of others; there is a normative standard that recognises that people must be respected as individuals.

The European Court of Human Rights has ruled extensively on the human right to respect for private life, and a line of caselaw has been created. This produces a canon of decisions where particular disputes have been settled where the particular parties have been unable to resolve their conflict between themselves. However, does that line of cases produce a normative definition of privacy, i.e. one that sits with and accommodates the range of sensitivities expressed above? I think not. A courtroom determination arguably defines a point on the range of sensitivities as ‘privacy’, pragmatically for the parties. Our problem comes when we try to use caselaw as indicative of more than how judges resolve conflicts between intractable parties when a privacy right is engaged. Does this mean the law adds little to the broader normative question about how we, as researchers, should respect the privacy of those with whom we engage in our work?

Two North American contributions could help to understand this. The first expression of the legal right to privacy is usually recognised as Warren and Brandeis’ 1890 idea that we can agree that individuals have the right to be left alone.⁶ Reading their paper today, it resonates with current concerns: technological developments and the increasing press prurience required a right to be ‘left alone’. In the modern context of genetics, Allen proposes a broader typology of privacy: ‘physical privacy’, ‘proprietary privacy’, ‘informational privacy’, and ‘decisional privacy’.⁷ The first two, which seem strange ‘privacies’ today, are where Warren and Brandeis clearly see the Common Law as having reached in 1890. Law protects individuals’ physical privacy through consent; private property law is equally well established. Warren and Brandeis identified ‘informational privacy’ and what might be described as ‘reputational privacy’ as the area where the law needed to develop in 1890. Allen pointed to the vast and compelling literature around the woman’s right to choose in discussing the right to ‘decisional privacy’. Legal theory, in part, responds to current privacy issues. Today, two major privacy issues in research are the protection of personal data protection and informed consent.

7.4 Privacy in Specific Legal Responses: Personal Data Protection and Informed Consent

The development of the automated processing of personal data has focussed privacy, at least in part, around ‘informational privacy’.⁸ The Organization for Economic Co-operation and Development OECD Guidelines on the Protection of Privacy and Transborder Flows of Personal Data set an international standard in 1980 that remains at the core of data protection law.⁹ The guidelines are transposed into regional and national laws.¹⁰

Data protection, as an expression of an area of privacy, seeks to balance a variety of interests in the processing of personal data within the non-absolute nature of privacy; the object of data protection is to create the legal conditions under which it is possible and appropriate to process personal data. Taking the European Union General Data Protection Regulation 2016/679 (GDPR) as an example, there are four elements in data protection law: data protection principles;¹¹ legal bases for processing personal data;¹² information that must be given to the data subject;¹³ and, rights of the data subject.¹⁴ Each element contains a balance of interests.

For stand-alone research with human participants directly contacted by the researcher, the route through the GDPR is clear. Security and data minimisation standards (i.e. only gathering, analysing and keeping data for such a time necessary for the purpose of the project) are clear; data subjects can be informed about the project fully, and data subjects rights can be respected. More complex data-sharing methodologies – perhaps the project envisaged in Section 7.2 above – are more difficult to negotiate through the GDPR. Are original consents valid for the new processing? Was the consent too broad for the new GDPR requirements? Might the processing be compatible with the original purpose for which the data were gathered? Could the new processing be in the public interest? How should data subjects be informed about the proposed new project? Each of these questions is open to debate in the GDPR. And the problem is how can the lack of definitional clarity in the rights be resolved in such a way that it accommodates all the positions on the spectrum of interests indicated in Section 7.2 above? One could say, law must produce a working definition and in a democracy, all differences cannot be accommodated, so some will be disappointed. However, the sensitivity of the data in the example above shows that there is a danger that those who are not within the working definition of privacy will be alienated from participating in key areas of social life, perhaps even avoiding interaction with, say, health research or medical services to their detriment.

If one of the current legal discussions is around informational privacy, the other is around decisional privacy. Informed consent is a legal mechanism to protect decisional privacy, not just in research, but across consumer society. The right of individual adults to make their own choices is largely unchallenged.¹⁵ The choices must be free and informed. The question is, how informed must a choice be to qualify as a valid choice from an individual? This, in many modern biomedical research methodologies, is contested. A biobank, where data are gathered for the purpose of providing datasets for future, as yet undefined, research projects, depends on creating the biobank at the outset through a ‘broad’ consent. How though can an individual be said to give ‘informed consent’ if the purposes for which the consent is asked cannot be explained in detail? How can a consent that is ‘for research’ be specific enough to be an adequate safeguard of privacy interests? (See, for example, Kaye and Prictor, Chapter 10, in this volume for a specific discussion of consent in this context.)

The privacy issue is: what constitutes sufficient information upon which a participant can base her choice? Two conditions have to be satisfied: the quality of the information that will be made available; and (who determines) the amount of information that is necessary to underpin a decision. A non-specialist participant is not necessarily in a position to judge the first of these conditions. That is the role of independent review boards, standing as a proxy for the participant to assess the quality and trustworthiness of the scientific and methodological information that will be offered to the participant. For the second condition, what is sufficient information upon which to make a decision and who determines that decision, is a matter for the individual participant, and should not be seen as part of the role of the ethics committee, researcher or other body. The purpose of informed consent is to protect the individual participant from, essentially, paternalism – the usurping of the participant’s free choice of whether or not to participate (unless the decision is palpably to the detriment of an individual who is not deemed competent to make a choice). Therefore, in the general case, it is inappropriate to remove the determination of what is sufficient information to inform the particular person from that person, or to determine for them what are appropriate or inappropriate considerations to bring to the decision-making process. This would seem to be crucial in ensuring an individual’s decisional privacy – the extent of the right to make decisions for oneself.

7.5 Realising Privacy in Modern Research Governance

So far I have made two claims about privacy. First, individuals hold a range of sensitivities about their privacy (and we could add that this is a dynamic balance depending also upon the relationships between individuals and the emotional setting or moment of the relationship.¹⁶ Second, the law produces a mechanism for resolving conflicts that fall within its definition of privacy, but it does not provide a complete normative definition of privacy that meets all the social functions required of the concept (that will confront researchers negotiating privacy relationships with their participants). Two observations might help with locating our thinking at this point. First, there is not a complete, normative definition of privacy in any discipline that satisfactorily meets the dynamic nature of privacy. There are many different definitions and conceptualisations, but there is no granular agreement on the normative question – what ought I to understand as ‘my permitted private life’.¹⁷ Second, the presentation so far might appear to suggest that privacy is a matter of individual autonomy, in opposition to society. This, in the remainder of the chapter, I will argue is not the case, by exploring how privacy might be operationalised, in our case, in research. The question is: what tools can we use to understand our relationships as individuals in society?

To do this, I suggest that there are three areas that can usefully be considered by both researchers (and participants) in the particular circumstances of a research project and by society in trying to understand the conceptualisation of privacy in modern society: the public interest, confidentiality and discourse.

The public interest, the common good, as a measure of solidarity is very attractive. It addresses directly the range of sentiments problem to which I refer to throughout the chapter (see Kieslich and Prainsack, Chapter 5, and Sorbie, Chapter 6, in this section, and Taylor and Whitton, Chapter 24, later in this volume). Appealing to the public interest is a practical mechanism that answers the individual’s privacy sensitivities with the following: whatever you believe to be your privacy, these are the supervening arguments why, for example, you should let me stand on your land or use your personal data (your privacy has to accommodate these broader needs of others). The difficulty with the public interest is that it seems itself to have no definition or internal rules. Appeals to the public interest seem to be constructed loosely through a utilitarian calculus: the greatest utility for the greatest number. Mill himself identifies the problem: the tyranny of the majority. The problem has two elements. The claim to ‘supervening utility’ could seem itself to be a subjective claim, so those in the minority, suffering the consequences of a loss of amenity (in this case the breach of their privacy), are not immediately convinced of the substance of the argument. The construction does not balance the magnitude of the loss to the individual with the benefit (or avoided loss) of another individual; rather, the one stands against the many. This is not particularly satisfactory, especially when one links this back to the fundamental breach and the sense of the loss of privacy cutting to the personhood of the individual. Adopting the arguments of Arendt, we might phrase this more strongly. Arendt identifies the individual as constituted in two parts: the physical and the legal. In her studies of totalitarianism, she finds that tyrannies occur where the two parts of the individual are separated by bureaucracies and the legal individual is forgotten. Left with only the physical individual, the human is reduced to an expendable commodity.¹⁸ Simple appeals to the public interest could be in danger of overlooking the whole individual and producing an alienation of those whose rights are removed in the public interest or common good.

Another way of constructing the appeal to the public interest can be through deontological rather than ontological theories, particularly those of Kant and Rawls. Taking Kant,¹⁹ a first step would be to consider the losses to individuals – the person who stands to lose their privacy rights, and a person who would suffer a loss if that privacy was not breached. A second step would be to require each of those individuals to consider their claim to their privacy through the lens of the second formulation of the Categorical Imperative – that one should treat others as ends in themselves, not merely as means to one’s own ends.²⁰ Because privacy is not an absolute right, when making such a claim, we must each ask: do I merely instrumentalise the other person in the balance by making this privacy claim? This is a matter of fact: which of us will suffer most? The third element is to acknowledge that the law can require me to adopt that choice if I fail to make it for myself, as it is the choice I should have made unprompted (I can argue that the calculation on the facts is incorrect, but not that the calculation ought not to be made). Rawls might construct it slightly differently: whereas I might prefer a particular action preserving my privacy, I must accept the breach of my privacy as reasonable in the circumstances. Using his ‘veil of ignorance’, when I do not know my potential status in society, I must adopt this measure to protect the least-well-off member of society when the decision is made.²¹

In the example raised in Section 7.2, using this public interest consideration helps to reconcile the range of sensitivities problem. As a researcher trying to design privacy safeguards, I can use the calculations to evaluate the risks and benefits identifiable in the research, and then present the evaluation to participants and regulators. The public interest creates a discourse that steps outside self-interest. However, this sets off a klaxon that the public interest is not antithetical to privacy, as presented here; public interest is part of privacy. And I agree. Here, I am suggesting that using public interest arguments is a mechanism for defining the relationship of the individual to others (that is, to other individuals). The result is not saying that the public interest ‘breaches’ the privacy of the participant, but that it helps to define the individual’s privacy in relation to others (for the individual and for other people and institutions). It brings to the subjectivity of the dynamic range of sentiments (that I identified as an issue at the outset) the solidarity and community that is also part of one’s privacy.²² This holistic understanding of privacy as a private life lived in community not reducible to a simply autonomy-based claim is best explored by Laurie’s ‘spatial’, psychological privacy.²³

Confidentiality is a second legal tool to ensure participants’ rights are safeguarded. Arguably, it is a more practical tool or concept for researchers than privacy. Confidentiality earths abstract privacy concepts in actionable relationships and duties. Taking Common Law confidentiality as an example, it is constructed either expressly, as a contractual term, or it is implied into the conduct of a contract or through equity into the relationship between individuals.²⁴ Confidentiality depends on concrete, known parameters of the relationship, or parameters that one ought, in good conscience, to have known. Like data protection, it does not prohibit behaviour; rather confidentiality creates an environment in which particular behaviours can occur. This is important in the context of health research regulation because many potential research participants will be recruited through the professional relationships they enjoy with healthcare professionals; it is a tool that can be extended into other researcher–participant relationships. Confidentiality and the trust-based nature of that relationship can both help with recruitment and provide a welcome degree of reassurance about privacy protection.

Finally, and implicit throughout the operationalisation of privacy, privacy is a negotiated space that requires public engagement through discourse. Discourse ethics has a modern iteration, but a long history. The virtue ethics of Aristotle and Ancient Greek philosophy is dependent on the identification of the extent and nature of the virtues and their application in human life; Shaftesbury’s early enlightenment ‘politeness’²⁵ and the salons of the Age of Reason again ground the discussion of the questions, ‘who are we, and how ought we to behave?’ in public, albeit intellectual, discourse; today, Habermas et al. advocate this inclusion as a part of a participative democracy, perhaps reiterating the central arguments of the early Frankfurt School against the false consciousness of the Culture Industry.²⁶ The thrust of this whole chapter is that privacy must be debated and understood in the lives of individuals; universities, professional bodies, and ethics committees must facilitate conversations that empower individuals to realise their decisional privacy in making choices about the nature of their participation in society.²⁷

7.6 Conclusion

This chapter has focused on different aspects of a conceptual problem raised in relation to a modern research dilemma: how do we negotiate privacy-protecting research where individuals hold a dynamic range of sensitivities about their relationships to others in society? We have seen that whereas human rights law does not present granular definitions of privacy and courts use privacy concepts to resolve disputes in the area, attempts in legal theory and specific areas of law (personal data protection and informed consent) do not fill the conceptual gaps. The argument I advance is that using the public interest, confidentiality and public engagement discourse in constructing research protocols will go some way to address those gaps. It will also strengthen the relationship between researchers and the public they seek to engage and to serve, and could facilitate a greater understanding of the methods and objectives of science.

8 Trustworthy Institutions in Global Health Research Collaborations

Angeliki Kerasidou

8.1 Introduction

Trust is often cited as being fundamental in biomedical research and in research collaborations. However, despite its prominence, its specific meaning and role remain vague. What does trust mean, and is it the same whether directed towards individuals and research institutions? What is it about trust that makes it important in global health research, and how can we effectively promote it? This chapter analyses the meaning of trust and discusses its importance and relevance in the context of global health research collaborations.

In recent decades, biomedical research has moved away from a one-researcher-one-project model to adopt a more collaborative way of working that brings together researchers from different disciplines, institutions and countries. Global health research, a field that has emerged as a distinct area of biomedical research, exemplifies this trend towards collaborative partnerships. Global health, as ‘collaborative trans-national research …’,¹ often relies on collaborations between researchers and institutions from high-income countries (HIC) and low-and-middle income countries (LMIC). LMICs still carry the highest burden of disease globally and have a high prevalence of many illnesses that pose global threats (e.g. infectious diseases). This has motivated a number of national and international funders to support global health research. The redirection of funds towards global health has resulted in increased interest among HIC researchers in working on diseases such as malaria, tuberculosis, HIV–AIDS and conditions such as malnutrition, and also a new impetus in forming partnerships with colleagues from LMICs.

Global health research is seen as a natural field for collaborative work for two reasons. First, by definition, the problems that global health research is trying to answer are complex, multifaceted and transcend borders and boundaries; tackling such problems requires collaborations between disciplines, countries and institutions. Second, most of the issues global health is concerned with affect less affluent parts of the world. Building health research capacity in these countries, and strengthening their public health systems, is seen as the most effective and sustainable way to ensure the successful progress and implementation of global health research, and to meet global health priorities. The importance of trust and the role of institutions in establishing and promoting trust relationships are often noted in discussions regarding global health research collaborations. Trust is often presented as a foundational element of research participation,² data sharing³ and sharing of samples and other resources.⁴

Here, I give an account of what it means for an institution to be trusted and be trustworthy in the context of global health research. I employ the example of data sharing to illustrate the importance and value of trustworthiness as an institutional moral characteristic. I use the term ‘institution’ to refer to groups or collectives that actively undertake research, such as universities and research centres. I conclude that trust is important in global health research collaboration because of the power imbalance between partners that often characterises such collaborations. In order to promote trust, institutions need to focus on being trustworthy by developing a behaviour that corresponds to the aims, principles and values they profess to uphold, and by demonstrating that they have incorporated into their functions, rules and regulations the particular needs of their partners and collaborators.

8.2 What Is a Collaboration?

We use the term ‘collaboration’ in our everyday language to signify many different types of partnerships. Yet, not all ways of working together are collaborations. The term denotes a particular type of partnership where two or more partners come together to achieve a common aim or goal.⁵ Collaborations are non-hierarchical structures, based on the sharing of decision-making and responsibility that rely more on capacity and expertise, rather than on functions or titles.⁶ Consider, for example, a collaboration between a statistics unit and an epidemiology unit working together on a population health project investigating lung cancer. The two groups are committed to the aim of the project and share equal responsibility for its successful completion. They bring different expertise into the project, participate equally in decisions regarding its running and direction and share ownership for its outputs (e.g. authorship on academic publications). Collaborations are characterised by transparency, openness in communication, synergy and honest appreciation of each other’s positions. Transparency facilitates a collective awareness of the project, its structure, strengths and weaknesses, and promotes collective ownership. Open communication allows for the free flow of information and exchange of ideas, but also for the expression of concerns. The easier it is for people to talk to each other and share their thoughts and viewpoints, the easier it is for a project to stay on track and reach its goals. Understanding each other’s positions and particular circumstances is also important, as it helps with setting expectations at the right level, anticipating problems and foreseeing areas where conflict may arise. Finally, synergy, which describes the drive and desire to achieve the common goal and recognition of the partners’ interdependence in fulfilling it, is what drives such partnerships.⁷

Other types of partnerships or co-working include cooperation – which brings partners together who do not share the same goal, but who need each other’s skills and expertise to reach their individual aims – and hiring or commissioning someone to do a specific job. For example, someone can be brought into a project to complete a very specific task, such as to conduct a systematic literature review, collect samples or develop informed consent forms for a clinical trial. Once the task is completed, the person’s involvement in the project is ended. None of these types of partnerships can be described as collaborations as they lack the fundamental characteristics of non-hierarchical and synergistic co-labouring.

8.3 Trust and Trustworthiness

Alongside synergy and horizontal organisation, trust is regularly cited as a fundamental characteristic of collaborations.⁸ D’Amour notes that ‘the term collaboration conveys the idea of sharing … in a spirit of harmony and trust’.⁹ One empirical study that investigated what underpins successful collaborations in global health research from the perspective of scientists and other research actors, identified trust between partners as one of the major contributing factors.¹⁰ But why is trust so crucial for collaborations? Trust is yet another term that is often used in our everyday language but not always to describe the same thing. A short analysis would help to better define this term and see how it applies to the context of global health research.

Trust is an attitude towards a person whom we hope, and have good reasons to believe, will behave in a way that confirms our trust. This attitude can take different forms. People can trust others wholeheartedly and perfectly (A trusts B), for example their mother or spouse. But most commonly, trust is perceived as a three-part relationship (A trusts B to x). There are three main attributes of a trust relationship: vulnerability, assumption of good will from the trustee towards the trustor and voluntariness.¹¹ Vulnerability stems from the fact that when trusting, the trustor becomes vulnerable to the trustee as they acknowledge and accept that the trustee can decisively affect the outcome of the entrusted action.¹² This is what justifies feelings of gratitude or of betrayal when trust is confirmed or broken.¹³ Vulnerability is not, however, a personal characteristic of the trustor. Rather, it is a relational property that emerges from the act of trusting. Consider the following example: a researcher shares some potentially significant pre-publication findings with a colleague who also works in a similar area. She has previously worked with this colleague and trusts him. In her correspondence she stresses the importance of these findings and asks the colleague to keep them confidential. If the colleague confirms her trust, and keeps the findings confidential, she will feel her trust is confirmed; if, however, the colleague ignores her trust and publishes the findings or shares them with others (e.g. at a conference), she will justifiably feel betrayed. The feeling of betrayal is predicated on the fact that she has no assurances, other than her trust, to protect her from the colleague’s decision and behaviour, and this is what makes her vulnerable towards her trustor. Seeking assurances, by trying to constrain someone’s behaviours as a way of limiting one’s vulnerability would indicate that the trustor mistrusts, or lacks trust for that person. This is why trusting requires some level of optimism about the trustee or a normative attitude that the trustee ought to do what the trustor wills them to do.¹⁴

The second characteristic of trust is the belief that the trustee has good will towards the trustor.¹⁵ It is this belief that counterbalances vulnerability and provides a reasonable justification for trusting someone. To return to the previous example, the researcher reveals the pre-publication findings to her colleague because she has good reasons to believe that he has good will towards her and will not intentionally harm or hurt her. If she did not have good reasons to believe this, then choosing to reveal her findings and make herself vulnerable towards him, all things being equal, would be unjustified. Some challenge the importance of good will in trust, by suggesting that trust may be warranted when we believe that those we trust (trustees) will conform to social constraints and norms, or that they will act in the ways we expect out of self-interest.¹⁶ Yet, while social constrains and self-interest could increase people’s reliability, it is questionable whether such motives can underpin trust. A belief in the good will of the trustee signifies that the trustor has good reason to assume that the trustee cares about her and/or about the things about which she also cares. Although this could be problematic in situations where one does not have insight into the ‘psychology of the one-trusted’,¹⁷ one could still justify trust on the belief in the other’s good will by adopting a wide notion of good will, which includes commitment to benevolence and conscientious moral attitude.¹⁸

The third characteristic is that trust is voluntary, insofar as it cannot be forced or demanded. As Baier notes: ‘“Trust me!” is for most of us an invitation which we cannot accept at will – either we do already trust the one who says it, in which case it serves at best as reassurance, or it is properly responded to with, “Why should and how can I, until I have cause to?”’¹⁹ Trust takes time to establish, and requires an expectation that people will behave not only in the way we assume they will, but rather in the way we assume they should.²⁰ A consistent demonstration of good will, as well as capacity to perform the entrusted action, can provide a good reason for trust. And those who want to be trusted can help generate such relationships by fostering and increasing their trustworthiness.

Trustworthiness is a moral characteristic of the trustee and signifies that they have an attitude of good will towards the trustor by being responsive to the trustor’s dependency upon them.²¹ The motivation for behaving trustworthily also matters. Trustworthiness signifies something more than just the mere observation of rules and regulations out of self-interest or duty. It is not just a tactic to avoid punishment or penalties, or to fulfil one’s sense of duty. Potter describes trustworthiness as a virtue. ‘In evaluating someone’s trustworthiness’, she argues ‘we need to know that she can be counted on, as a matter of the sort of person she is, to take care of those things with which we are considering entrusting her’ (emphasis added).²² However, it is important to note that expectation for one to behave in a certain manner does not compel the trustee to behave in the expected way. The fact that one is being counted on forms an important consideration to be taken into account but does not force one to act in a certain way – otherwise one would be forced to act in a ‘trustworthy’ way even when the trust placed on one is unjustified or misguided.

So far, I have argued that trust is a relational mode predicated on a reasonable belief in the trustee’s skill to perform the entrusted action and also good will towards the trustor. Trust cannot be forced or demanded, and by trusting, one makes oneself vulnerable toward the person they choose to trust. Trustworthiness is a moral characteristic that indicates that someone can be counted on. It is not necessary that a trustworthy person is automatically trusted, but trustworthy behaviour can illicit trust. In the context of these definitions we can explore this chapter’s main questions: what is the role of trust in global health research collaborations and can institutions be trusted? If trust is commonly perceived as a characteristic of interpersonal relationships and trustworthiness as a personal quality or virtue, is it possible to talk meaningfully about trusting institutions, or to ascribe moral characteristics such as trustworthiness to collectives?

8.4 Trust in Global Health Research

In 2013, the Council on Health Research for Development (COHRED) published a report on fair collaborations in global health.²³ It noted that relying on HIC collaborators’ good will has not been sufficient to ensure fair and just collaborations between partners. What was needed instead, the report recommended, was to build LMIC institutions’ capacity in contract negotiations. The implication seems to be here, that instead of just trusting people to behave fairly and justly and thus opening oneself up to having their trust betrayed, one needs to ensure that people will behave this way. This could be achieved by putting in place contracts that direct and set the parameters of right behaviour. One way of understanding this contractual relationship is as relationship of reliance. Relationships of reliance are based on proven capacity and clear systems of accountability. In such relationships, the expectation is that the partners will act based on self-interest. Collaborators can ensure successful partnerships by aligning their interests and by putting in place rules to secure against defection. What makes relationships of reliance preferable to relationships of trust is that the former do not require an assumption of good will, nor do they require the trustor to become vulnerable to the trustee.²⁴

One important condition, however, must apply for relationships of reliance to work. Reliance requires power parity between partners.²⁵ This is because in relationships that operate on self-interest, it is far easier for the stronger partners to shift the balance to their favour. This is particularly relevant for global health research collaborations, which often bring together institutions from HICs and LMICs. Giving LMIC researchers and institutions the tools to defend and promote their own interests is one way of promoting reliable – rather than trusting – partnerships and addressing relationships of dependency, and COHRED’s efforts are a valuable step towards this. However, there are a number of reasons why building trust relationships and promoting trustworthiness remains important in this context. First, in situations where power parity between partners is lacking, trust can be an essential foundation on which to build a good and fair collaboration. For example, attitudes of good will – a crucial feature of trust – can counterbalance self-interested motivations. Second, trust could facilitate good collaborative partnerships, by creating a safe environment in which partners can focus on achieving the common goal rather than on protecting their own interests. Finally, it is common sense that everyone, given the option, would prefer to work with partners they trust and not only with those they can reliably predict their behaviour.

If we accept that trust remains relevant in global health research, what we need to consider next is how it could be promoted. One of the reasons that could justify and encourage a trust relationship is trustworthiness. Although trustworthiness cannot always and de facto guarantee trust, moral agents who want to be trusted by their partners and collaborators could do worse than to try to cultivate and demonstrate their trustworthiness. However, while trustworthiness can be attributed to individual persons, collaborations in global health research however, are not just between individuals, but also between institutions. It is important, therefore, to examine whether it is reasonable to talk about ‘trustworthy’ institutions.

8.5 Trustworthy Institutions

Prior to ascertaining whether institutions (e.g. universities, research centres) can be trustworthy, we must establish whether it is reasonable to talk about groups and collectives possessing such moral characteristics as trustworthiness.²⁶ In other words, can institutions involved in global health research collaborations be moral agents? There are two main reasons, which, I believe, give support to the view that collectives are entities that could be treated as moral agents: first, such a position chimes with the way we think about the role of collectives in public life and also the way we treat them in practical terms. For example, we expect universities to adhere to ethical principles when conducting research and we hold them responsible when they fail to do so.²⁷ In law, collectives are treated as bearers of rights and responsibilities and can be penalised for wrongdoing and for failing to meet their duties and obligations. Second, the view that institutions are moral agents reflects a growing realisation that many issues require the action of collectives in order to be resolved. Actions such as conducting large-scale research aimed at halting pandemics or reversing climate change are unavailable to individuals but possible to groups and institutions. If we accept that these actions reflect duties that ought to be met, then these duties will have to be ascribed to actors that can meet them.²⁸

Being trustworthy means that a given individual (or institution) acts not only as they are expected to, but in a way that demonstrates that they have taken into account the fact that someone is counting on them.²⁹ Trustworthiness is a characteristic or moral attitude that is revealed through one’s actions and also in one’s ‘values, commitments and loyalties’.³⁰ When it comes to institutions, their trustworthy character is revealed in their professed goals and aims, at their institutional structures, internal rules and regulations as indicators for their moral motivations,³¹ and in their reputation and track record as indicator for their skill and commitment to right action.³² Researchers and groups in global health who are looking for collaborations would perceive institutions that declare to care about things they also care as more trustworthy, rather than institutions that do not profess such interests. An institution’s track record and proven capacity in their ability to reach these shared goals would add to its trustworthiness. Importantly, being trustworthy is not about following rules but acquiring a disposition of trustworthiness. Behaving in a certain way only for fear of penalty demonstrates a self-interested orientation, rather than concern for others or about what others value. Therefore, institutions would need to demonstrate that their commitment to trustworthy behaviour is principled and corresponds with their aims and purpose, rather than motivated by a desire to avoid sanctions and penalties, including loss of future collaborations. For institutions participating in global health research, this will mean demonstrating that they have incorporated into their structures, rules and regulations central aims of global health such as addressing health inequalities, improving health through rigorous research and promoting research capacity in countries that lack it. Using the example of data sharing and open access may help to illustrate this point.

8.6 Trustworthiness in Data Sharing Collaborations

Data sharing is often presented as foundational to global health research.³³ Health and health-related data (e.g. genomic, phenotypic or clinical data) are an inexhaustible resource that could be used repeatedly to address multiple research questions, provide answers to a plethora of global health issues and thereby help reduce the global burden of disease. For example, data sharing between countries and institutions is essential in the attempt to understand and respond to epidemics and pandemics, as the cases of the H5N1 avian flu in 2007 and the outbreaks of Ebola in 2014 and ZIKA in 2015 have demonstrated.³⁴ Recognition that health data offer valuable resources with multiple applications has led to a position where data sharing is seen as both a scientific and moral imperative in biomedical research,³⁵ while failure to share has been variously described as being unscientific, contrary to research integrity, wasteful and unjust.³⁶ In recent years, a lot of effort has been put into facilitating and promoting the open sharing of data.³⁷ Progress in data sharing tools, methods and policies is seen as the ‘innovation with the farthest-reaching impacts among the global medical community’.³⁸ This has led to the wide endorsement of data sharing and open access policies by many international research bodies, funding organisations, academic publishers and policymakers.³⁹ It seems that adopting and promoting open access of data and the implementation of a robust open data sharing policy would signal an institution’s moral character as being one dedicated to open, transparent and robust science, and to maximising research benefits for all. But would this mean that such an institution is trustworthy?

In the context of global health research, data sharing can be ethically and practically complex. Despite its potential benefits, there are significant ethical and societal barriers to the wide implementation of open data sharing policies and practices.⁴⁰ Leaving aside confidentiality and consent, a significant issue in global health stems from the uneven ability of institutions in different parts of the world to utilise data.⁴¹ As Serwadda and others note, advancements in technology that make data collection, storage and sharing easier, and the shift in the social and scientific norms to support openness and sharing, is undermining equitable collaborations between HIC and LMIC.⁴² This has led to ‘a landscape, often characterised by limited capacity and deep mistrust, for acceptance and implementation of open data policies’.⁴³ Furthermore, despite claims that open data sharing could lead to advancements that would be beneficial to all, including to the communities of origin, this is not always the case. Often, the new therapeutics developed are either too expensive for LMICs to purchase, or these countries lack adequate public health structures to make use of any new actionable knowledge. For example, in 2007, Indonesia refused to share its H5N1 avian flu data and samples unless their country was guaranteed affordable access to vaccines – and researchers from other LMIC seem to think that this was a fair response to an unfair situation.⁴⁴ Although data sharing could accelerate the production of new and useful knowledge, it can also contribute to the perpetuation of global injustices and undercut the stated goals of global health research.

Adopting an open access policy to data sharing could make an institution reliable, in the sense that its partners would know what to expect and would be able to predict its behaviour and actions with a certain degree of accuracy. Would recognising this institution’s reliability in this domain, however, amount to it been perceived as being trustworthy by its partners?⁴⁵ Although adopting certain (moral) rules and acting consistently is an indication of a certain (moral) character, trustworthiness requires more than that; it requires an attitude of good will and responsiveness to the other’s needs. Hence, a trustworthy institution in global health research would not blankly endorse an open data sharing policy, but would retain a flexible stance, leaving room for adapting its policies with the specificities of its collaborators in mind. Such adaptations might include time-specific embargoes on data release to give partners a fair head start on using their data, restrictions of use to protect the stated research aims of such partners and embedding contextually meaningful capacity building activities into their collaboration.⁴⁶ Although an open data sharing policy could ensure that maximum value and utility is extracted from data, allowing for the negotiation of a managed access policy would signal an institution with good will towards its collaborators and ‘a direct responsiveness to the fact that the other is counting on [it]’.⁴⁷

8.7 Conclusion

Trust is and will remain important in global health research collaborations, at least until the power imbalance between LMIC- and HIC-based researchers and their institutions is addressed. Institutions committed to advancing the aims of global health, including helping build research capacity in LMIC, should aim to promote fair and trusting collaborations. The best way of achieving this is by cultivating and demonstrating their trustworthiness as a way of eliciting justified trust. Being trustworthy requires more than just the observation of rules or the incorporation of moral principles in policies and structures. Although such moral attitudes would likely increase an institution’s reliability, trustworthiness also demands attention to the relational aspect of trust. Trustworthiness requires that the institution is concerned with its partners and what its partners value, acknowledges its partners’ vulnerability and demonstrates ‘a direct responsiveness to the fact that the other is counting on [it]’.⁴⁸ The practice of data sharing provides a useful case to examine what being a trustworthy institution might look like in practice. Moving forward, more research will be required to fully examine the relationship between rules, regulations and policies and the moral character of institutions in global health.

9 Vulnerabilities and Power The Political Side of Health Research

Iain Brassington

9.1 Introduction

In this chapter, I will argue that there is a political dimension to research, and that accounts of health research regulation that ignore political relations between stakeholders are therefore incomplete. The concept of vulnerability – particularly vulnerability to exploitation – provides the grit around which the claims are built. This is because vulnerability is an inescapable part of human life; because research participation may magnify vulnerability, even while health research itself promises to mitigate certain vulnerabilities (most directly vulnerability to illness, but indirectly vulnerability to economic hardships that may follow therefrom); and because vulnerability is manifested in, exacerbated by, or mitigated through, inherently political relationships with others, the groups and communities of which we are a part, and in the context of which all research takes place. I shall not be making any normative claims about research regulation here, save for the suggestion that decision-makers ought to take account of latent political aspects in their deliberations. For the most part, I shall simply attempt to sketch out some of those political aspects.

9.2 Setting the Scene

Certain key terms ought to be defined at the offset.

• By vulnerability, I understand a susceptibility to harm or wrong arising from a physical or social contingency above and beyond that found in a recognisably decent human life.

• By the vulnerable, I understand those who are at an elevated risk of harm or wrong arising from such contingencies.

• By power, I understand the capacity to act, or to resist being acted upon.¹

• By power relations, I understand the interplay of agents’ relative power.

• By the political, I understand the domain in which power relations are manifested.²

• By exploitation, I understand the use of some thing or person to serve one’s ends.

Some elaboration is in order. At its most basic, vulnerability is any susceptibility to harms or wrongs; but such an understanding is generally unhelpful, because (per Rogers) ‘it obscures rather than enables the identification of the context-specific needs of particular groups’³ – plausibly, one may read this as ‘individuals and groups’ – and because (per Wrigley and Dawson) ‘if everyone is vulnerable, then no one is’.⁴ A more nuanced and useful conceptualisation of vulnerability would relate it to a susceptibility to harms or wrongs greater than is normally found in a recognisably decent human life. Correspondingly, in stating that ‘[s]ome groups and individuals are particularly vulnerable and may have an increased likelihood of being wronged or of incurring additional harm’,⁵ the Declaration of Helsinki is plainly referring to the ways in which persons may be further vulnerable above a universal baseline. That said, I will indicate below that, and how, the more basic understanding is not without utility.

Wendy Rogers provides an account of some of the difficulties of conceptualising vulnerability in this volume (see Chapter 1) and taxonomies of different kinds of vulnerability have been offered elsewhere.⁶ I will neither rehearse nor assess those accounts here, save to highlight the idea of pathogenic vulnerability, the sources of which include morally dysfunctional or abusive interpersonal and social relationships, and sociopolitical oppression or injustice,⁷ and which thereby illustrates plainly one of the political aspects of vulnerability. However, we conceptualise or parse it, though, vulnerability invites politically-informed responses. Wrigley and Dawson assert that vulnerability ‘implies an ethical duty to safeguard [the vulnerable person’s or group’s] well-being because the person or group is unable to do so adequately themselves’.⁸ For his part, ten Have claims that ‘[w]hat makes vulnerability problematic is the possibility of abuse and exploitation’; for him, vulnerability need not be eliminated, so long as it can be ‘compensated, diminished, and transformed’.⁹ Putative duties to safeguard the vulnerable, or to militate against abuse, could be discharged by individuals in some cases, and by the state in others. Venturing claims one way or the other implies a political position, because it speaks to decisions about how and by whom power may be exerted over, and on behalf of, another.

Exploitation, as defined above, implies the exercise of power over another: the exploiter is in this context more powerful than the exploited. As a manifestation of the power relations between agents, it is therefore a political phenomenon; and if exploitation violates a right of the exploited, it may be wrongful. Insofar as that vulnerability is susceptibility to certain harms or wrongs, it includes susceptibility to wrongful exploitation; and since exploitation is a political phenomenon, vulnerability to wrongful exploitation will therefore also be political. The relevance of this will become clear as we proceed.

9.3 Individuals’ Vulnerability in Research

It is in the nature of research that outcomes are uncertain; this means that healthy volunteers in medical trials might be susceptible to unexpected harms. If research concerns a treatment’s effectiveness, it will often be necessary to recruit patients into a trial; but such a cohort will, by definition, be of people with medical needs, some of which may be otherwise unmet. The prospect of a health benefit, especially if there are few other extant or affordable treatment options, may mean that this somatic vulnerability is accompanied by vulnerability to exploitation: the patient may allow herself to be enrolled into a trial into which she would not have allowed herself to be enrolled otherwise. Moreover, participants’ ability to control their exposure to risk may be limited: even without perfect knowledge, researchers are likely to have greater insight into the risks, and are able to control information in a way that participants, who rely on researchers for information, are not. This is a form of epistemic power held by researchers. Indeed, researchers may be perceived as having control over information even when they do not; and this perception may give them a ‘credibility excess’¹⁰ that is itself a source of epistemic power, insofar as that it can influence the decisions that participants make, perhaps to the extent of inhibiting their making them at all. How researchers and research managers handle the power disparity between them and participants is a political problem writ small.

Even putting the political aspect of this relationship to one side, it would be reasonable to expect that researchers address questions about the broader political context of their programme and protocol. After all, if someone enrols as a research subject because it is the only way they can access treatment, or because it is the only way they can afford it or other necessities, this tells us something about the characteristics of the state in which they live – notably, how just it is. Correspondingly, acknowledged political injustice may alter the likelihood that a person would act as a participant, how they behave as a participant, and whether their participation is voluntary. The political questions are clear. Does the political environment in which a person lives provide adequate protection against exploitation? What should be done if it does not?

At times, it may be that political circumstances make ethically acceptable research impossible. In extremis, this might be because certain people are forced to participate by an overweening government: prisoners, say, may be particularly vulnerable to this kind of pressure in some regimes. But participatory voluntariness may also be eroded by the lure of medical treatment that participants would not otherwise have, perhaps because it is not normally within the state’s abilities to provide it. On the other hand, refusing to carry out research because the context in which it is proposed creates vulnerabilities or militates against their mitigation, may simply mean that would-be participants are deprived of benefits that they might have had – Ganguli-Mitra and Hunt touch on this problem when they consider the use of experimental interventions during the 2013–2016 Ebola outbreak in Chapter 32 of this volume – and that scientific opportunities are lost as well. A further problem is that some illnesses are illnesses of poverty; it may not be possible to carry out research on those illnesses without recruiting people who are socioeconomically vulnerable, because less socioeconomically vulnerable people would be less susceptible to the illness in question. (That said, one may wonder whether prioritising poverty alleviation would dilute any imperative to research the illnesses that it causes.)¹¹

There is unlikely to be an easy way to determine whether a given political situation is conducive to ethically sound research. Possibly the best that could be said is that good research practice may require an awareness of, and sensitivity to, the prevailing political dispensation as it applies to certain individuals.

9.4 Individuals and Groups

How well do these considerations translate to groups?

For the moment, I shall assume that groups are aggregates of individuals, and that groups’ vulnerabilities are aggregates of individuals’ vulnerabilities. Admittedly, this is a simplification: something might be good or bad for the group as a whole without being good or bad for each and every member thereof; a group’s integrity, say, may be vulnerable in a way irreducible to its members’ vulnerabilities. But, for the time being, and given space constraints, I think that the simplification is not gross.

Granted that groups are aggregates of individuals, discriminatory or otherwise unjust political arrangements may exacerbate or even generate vulnerabilities in those individuals qua group-members. Most obviously, individuals may be at increased susceptibility to harm or wrong if they lack legal or political representation, education, and so on, because of their membership of a particular group. This kind of powerlessness to resist injustice is a political product generating a pathogenic vulnerability – and a state in which injustice is not addressed is itself unjust, or vicious in some other way.¹² Further, the legacy of historic injustices may linger even if the unjust policies were ditched long ago.

But even having been identified as a member of a group at all may generate vulnerabilities in individuals, irrespective of the political circumstances. To give a simple example, a public health research programme may require population-level data-gathering. Any given individual may feature in such research by dint of having been identified as belonging to a target group – but they might not be aware that the research is taking place. Already, then, we will be confronted with the possible wrong of individuals not being treated as ends in themselves. This wrong has a political dimension in that the power of research subjects is a consideration: one is powerless to withdraw from a study in which one does not know that one is a subject.¹³ We might say that researchers who think their work is worth the effort ought to approach those persons who may be captured by it – something that is in principle in their power to do – to give them a chance to opt out, and that research without this opt-out would be impermissible. This would restore to individuals some power. Yet giving people the chance to opt out of a large cohort study would be very difficult in practice, and – perhaps more importantly – would risk undermining the study’s scientific integrity, which is itself a criterion of its moral permissibility. There is no clear solution to this sort of problem, though awareness of it is an important precursor to formulating best practice.

Even if that problem is solved, others present themselves. A group might be characterised by an elevated occurrence of certain characteristics. Imagine that members of group A typically have an unusually high susceptibility to a given disease, and that members of B typically have an elevated inherited resistance to it. Facts like this would generate legitimate questions that would be worth investigating: by learning about how it is that some human bodies are more resilient or susceptible to an illness than others, we could glean insight that would help us prevent it or treat it when it occurs. Yet both groups would also be vulnerable to injustice and exploitation. Thinking about the distribution of the eventual benefits of the research will help show how.

Clearly, medical research contributes to the development of new treatments, at least some of which provide profits for the manufacturers; and the profit motive may drive socially-desirable research. However, the line between just profit and profiteering, which is by definition unjust, is crossed if the benefits of the research are not fairly distributed between researchers – and their backers – and participants. Thus, for example, if any drugs arising from research dependent on the participation of members of A are profitmaking, and those participants derive no benefit – perhaps because socioeconomic deprivation makes the drugs unaffordable – that would be a paradigmatic example of injustice. Even if A is a reasonably well-represented and educated group, it or its members might be exploited in other ways, perhaps by being targeted specifically for expensive medical interventions. B would be less vulnerable on these fronts, since its members’ need for any drugs is, by stipulation, reduced. However, again, if members of B received no benefit at all from research into which their contribution was crucial, they might still have been exploited. After all, exploitation does not always imply harm – but to have contributed to something that benefits others is to have been exploited; and if this was without recompense, or at least without the opportunity to waive recompense, it is arguably to have been treated wholly as a means to their end, and therefore to have been wronged.

That groups can be exploited or treated unjustly – such as in the ways illustrated by A and B – is sufficient to show that there is a power differential in play; and because the political domain is that in which power relations are manifested, it is also straightforward to point out that this has a political dimension. As such, a full assessment of the ethics of a given piece of research, and a convincing regulatory policy, would take into account the political situation, both locally and globally.

On the local scale, it would be important to keep in mind questions such as whether the group’s vulnerability to exploitation is exacerbated by things like systemic discrimination or economic disadvantage, which may make it difficult for members of a community to assert moral rights that themselves may not be fully reflected in law. The better protected a group is in law, the better able it and its members will be to avoid or resist exploitation in other contexts.

Globally, if research is carried out on people from low-income countries, and the benefits of that research flow overwhelmingly towards high-income countries, what we see is, in effect, a transfer of benefits from the least-wealthy to the most. In this context, the Swiss NGO Public Eye estimates that

[a]lthough most clinical trials are conducted in the United States and Europe, over the last 20 years there has been a strong tendency towards offshoring to developing and emergent countries. The proportion of testing in emerging markets increased from 10 percent to 40 percent. This continued to increase between 2006 and 2010, while the proportion of clinical trials conducted in Western Europe and the United States fell from 55 percent to 38 percent.¹⁴

Such a transfer is facilitated and guaranteed by a system of domestic and international laws through the framing of which power becomes visible; and keeping those laws in place, or altering them to reduce the chance of exploitation, is correspondingly a matter of the political will of the powerful. And though individual researchers are powerless to do much about laws that facilitate unjust exploitation derived from research, they are able to do something about the design of individual research programmes, and whether or not they go ahead to begin with.

Yet this is not the most difficult problem in the way of handling group vulnerabilities in health research: that concerns how researchers and regulators should respond when the interests, wishes, and vulnerabilities of different members of a group are in tension. It is this problem to which I turn my attention now.

9.5 Group Membership and Group Vulnerabilities

Return to groups A and B from the example above. Suppose that researchers are particularly interested in a gene that is common in A but not in B; they hypothesise that this gene is relevant to understanding the medical condition they are studying. This presents a problem for consent: because genes are not confined to one member of the group, any individual’s participation in the programme automatically recruits other members as what we might call ‘indirect participants’. It might therefore be argued that every member of the group is vulnerable to having been wronged, even if the ‘direct participant’ – the person, say, whose blood is drawn – has given full consent. How might we take account of this vulnerability in other members of the group?

It is a commonplace that full, informed consent is at least a part of protecting the rights of research participants; from that we can infer that it would be part of mitigating their vulnerability. But obtaining the consent of each member of the group before beginning the research would be wildly impractical for any but the smallest groups in the most confined geographical areas. More, we would have to decide whether assent to participation must be unanimous: whether, that is, the permission of a person who would presumably not be a direct participant in the research should be a requirement to secure the participation of those who would be. Inasmuch as that this is a question about the relationship of individuals to each other, it is political.

Another layer of complication is added if we deny that a group’s vulnerabilities are reducible to those of the aggregate of its members – and it seems as though this may sometimes be the case. Plausibly, there will be situations in which the vulnerabilities of individuals and of groups do not map onto each other particularly closely, if at all: groups can be vulnerable in their own right. For example, the size of a tribe of hunter-gatherers may fall as its members urbanise; we might therefore want to say that the group is increasingly vulnerable even as individual members, thanks to better access to things like health care and education, become less so. But if this is correct, then even addressing every individual’s vulnerability may not address wholly the vulnerabilities of the group in the abstract, and so even unanimous consent may be insufficient to prevent impersonal harms or wrongs. Yet it does not seem plausible to say that a research programme should not go ahead because it is impossible to guarantee that the vulnerabilities of the group as a whole will not be exploited. Partly, this is because it seems to sacrifice the (probably admirable) willingness to participate of identifiable members of the community on the altar of concerns about everyone and no one in particular. And partly it is because, though the vulnerabilities of identifiable other members of the community and of the community itself may be important, they are not likely to be the only relevant moral consideration. After all: everyone who stands to benefit – directly or indirectly, tangibly or intangibly – from the research has an interest in its going ahead. These are political problems: to echo Bernard Crick, ‘conflicts of interest, when public, create political activity’.¹⁵

Having a representative or representative body that can speak on behalf of the group broadly understood may be suggested as a way forward. For example, Charles Weijer argues that, although some groups and communities ‘do not possess a legitimate political authority empowered to make binding decisions on behalf of members’, which means that ‘it would be both impossible and inappropriate to seek community consent for research participation’, they ‘may nonetheless have representative groups, and researchers ought to engage these groups in a dialogue concerning study design, conduct, and research results’.¹⁶ Yet we may still wonder how we determine who represents the community and in what way, and what we should do if and when the views of members of the group or community broadly understood diverge from the views of its notional representatives. We should not forget the possibility that would-be research participants may be vulnerable to peer pressure, either to participate or not to, from the group of which they are a part. In this light, it is not obvious what should happen if one member of group A or B from the example above is willing to volunteer as a research participant when those representatives are opposed, or vice versa. Again: since these problems concern how individuals and groups interact, they are plainly political.

Neither should we forget that individuals may be members of several communities or groups simultaneously. As such, referring to membership of a community is likely to mask other problems. Accordingly, when, in the context of genetic research, Jones et al. state that, ‘depending on the research focus’,

a community may include a group sharing a common geographic location, ethnicity, disease, occupation, etc. as well as virtual communities linked regionally, nationally or internationally¹⁷

they leave open questions about whether one must specify just one of these, and which – if any – takes priority over the other. At some point, someone would have to stipulate that the ‘kind’ of community in question is this or that; but such stipulations would appear to be always disputable, and likely politicised to boot.

When considering research involving vulnerable groups then, the relative power of the researcher (and the researcher’s backers) and the participant is not the only consideration. Researchers’ power relative to that of the group as a whole would also be important to keep in mind; at the same time, so would the power of the group as a whole in relation to the individual participant. Finally, even if we think that the interests of the community are significant, there is a lingering question of where the boundaries of the community should be drawn: sufficiently cosmopolitan politics may deny that the boundaries of this or that group are significant.¹⁸ These questions are inescapably political given the understanding of the political as that domain in which power is manifested, but also political in a more everyday sense, because they speak to problems of how individuals relate to the groups and communities of which they are a part, and how we define group or community membership.

9.6 Politics and Protection

I noted earlier in this chapter Wrigley and Dawson’s claim that there is an imperative to mitigate vulnerability. Allowing that there is such an imperative, it speaks to the obligations individuals have to each other, but also to the responsibilities of the community, as expressed through the state. Either way, there is a political dimension to it. More, it is reasonable to suppose that health research is one of the things that might be enlisted as a means of mitigating universally-shared human vulnerabilities, and it is likely that a functioning state of some sort is necessary to facilitate such research. Indeed, the idea that political existence is in one way or another crucial to human flourishing has been a touchstone of western philosophy since Aristotle.¹⁹

It should also be remembered that, as well as facilitating research, protecting research subjects from harms and wrongs more generally – notably, through regulation – falls within the state’s demesne. It is in this light that we might consider moves such as the reforms to the Mexican General Health Law approved in 2008, which made ‘the sampling of genetic material and its transport outside of Mexico without prior approval […] illegal’.²⁰ The Genomic Sovereignty amendment states that Mexican-derived human genome data are the property of Mexico’s government, and prohibits and penalises their collection and use in research without prior government approval.²¹ This may be seen as an attempt by the Mexican state to protect vulnerable groups within it from the depredations of large and wealthy biotech companies. Such moves may be seen as particularly called-for when, for example, the results of genetic research might be patentable. In such circumstances, a national government can shield minority groups that might not be able to resist unjust exploitation on their own, and can work to give them authority over what happens to data derived from their members.

This is not the only way to see things, though. Cooperation with commercial research institutions could provide vulnerable groups – think again of groups A and B above – with a way to capitalise on their own genetic resources, by entering into benefit-sharing agreements that guarantee them a portion of any proceeds. Such cooperation may also provide a way for research attention to be paid to conditions that may be more prevalent in that community than elsewhere. On this basis, legislative moves such as Mexico’s may be seen as an appropriation, however well-meaning, of the rights of some of its people(s) to decide for themselves how to handle data derived from their genes. Alternatively, it may be national governments that are best able to persuade biotech companies to research certain conditions at all; and the state may be able to use its power not to prevent a group exploiting its genetic resources, or to coopt them, but to ensure that the group in question it is able to exploit them effectively, since only national governments have the heft to ensure that the exploitation is not of the objectionable sort.

9.7 Conclusion

Research promises us a way to address human vulnerabilities, but it may exacerbate others in the process. Ensuring informed consent from participants may be a means of mitigating some of these, but not others. Those that it might mitigate often have a political genesis; but the relationship between researcher and participant can only really be understood when its own inherent political dynamic is acknowledged, too. More, the complications of the political aspects of research are magnified when we are dealing with vulnerable groups and communities, and with their members.

It has not been the aim of this chapter to offer any normative suggestions; nevertheless, fully to account for individuals’ vulnerability, and reliably to avoid exacerbating or exploiting it unjustly, researchers should probably take account not just of the familiar ethical norms of health research, such as informed consent, but also of the political context in which such norms are applied.

Section IB Tools, Processes and Actors Introduction

Edward Dove and Sethi Nayha

This section of the volume explores the tools, processes and actors at play in regulating health research. Regulators rely on a number of tools or regulatory devices to strike a balance between promoting sound research and protecting participants. Some of the paradigmatic examples are (informed) consent and research ethics review of proposed projects; both are explored in this section. Other examples include intellectual property (especially patents), data access governance models, and benefit-sharing mechanisms. Much of the contemporary scholarship on and practice of health research regulation relies on, and criticises, these tools. Relatedly, and arguably, regulation itself is processual; it is about guiding human practices towards desirable endpoints while avoiding undesirable consequences. There has been little discussion of this processual aspect of regulation to date and the specific processes at play in health research. Contributors in this section explore some of the most crucial processes, including risk–benefit analysis, research ethics review and data access governance mechanisms. Further, as becomes apparent, processes can themselves become tools or mechanisms for regulation. Finally, one cannot robustly explore the contours of health research regulation without a consideration of the roles regulatory actors play. Here, several contributors look at the institutional dimension of regulatory authorities and the crucial role experts and science advisory bodies play in constructing health research regulation.

Despite the breadth of topics explored within this section, an overarching theme emerges across the thirteen chapters: that technological change forces us to reassess the suitability of pre-existing tools, processes, and regulatory/governance ecosystems. While a number of tools and processes are long-standing features of health research regulation and are practised by a variety of long-standing actors, they are coming under increasing pressure in twenty-first-century research, driven by pluralistic societal values, learning healthcare systems, Big Data-driven analysis, artificial intelligence and international research collaboration across geographic borders that thrives on harmonised regulation. As considered by the authors, in some cases, new tools, processes or actors are advocated; in other cases, it may be more beneficial to reform them to ensure remain they fit for purpose and provide meaningful value to health research regulation.

Much of the discussion focusses therefore not only on the nature of these long-standing tools, processes, and actors, but also on how they might be sustained – if at all – well into the twenty-first century. For example, the digital-based data turn necessitates reconsidering fundamental principles like consent and developing new digital-based mechanisms to put participants at the heart of decision-making, as discussed by Kaye and Prictor (Chapter 10). Shabani, Thorogood and Murtagh (Chapter 19) also speak to the challenges that data intensive research is presenting for governance and in particular the challenges of balancing the need to grant (open) access to databases with the need to protect the rights and interests of patients and participants.

This leads to another related theme emerging within this section: the need to examine more closely the participatory turn in health research regulation. Public and participant involvement is becoming an increasingly emphasised component of health research, as illustrated by public engagement exercises becoming mandatory within many research funding schemes. But, as Aitken and Cunningham-Burley (Chapter 11) note, many different forms of public engagement exist and we need to ask ‘why’ publics are engaged, rather than simply ‘how’ they are engaged. They suggest that framing public engagement as a political exercise can help us to answer this question. For Chuong and O’Doherty (Chapter 12), the process of participatory governance also necessitates unpacking, particularly due to the varied approaches taken towards embedding deliberative practices and including patients and participants as partners within health research initiatives. Both of these chapters help set up discussion and analysis to come later in this book, specifically the contribution from Burgess (Chapter 25), who makes a case for mobilising public expertise in the design of health research regulation.

Beyond the inclusion of publics and participants in decision-making, many authors in this section raise additional questions about decision-making tools and processes involving other regulatory actors. For example, Dove (Chapter 18) notes how research ethics committees have evolved into regulatory entities in their own right, suggesting that they can play an important role in stewarding projects towards an ethical endpoint. Similarly, McMahon (Chapter 21) explores the ways in which institutions (and their scaffolding) can shape and influence decision-making in health research and argues that this ought to be reflected when drafting legal provisions and guidance. On the question of guidance, Sethi (Chapter 17) lays out different implications that rules, principles and best-practice-based approaches can carry for health research, including the importance of capturing previous lessons learned within regulatory approaches. Sethi’s discussion of principles-based regulation helps round out the discussion to come later in this book, specifically Vayena and Blassime’s contribution (Chapter 26) on Big Data and a proposed model of adaptive governance. Sethi’s chapter also engages with another key theme emerging within this section: the construction of knowledge-bases and expertise. For example, Flear (Chapter 16) suggests that basing current framings of regulatory harm as technological risk marginalises critical stakeholder knowledges of harm, in turn limits knowledge-bases. Indeed, in considering how governments make use of expertise to inform health research regulation, Meslin (Chapter 22) concludes that it will be best served when different stakeholders are empowered to contribute to the process of regulation, and when governments are open to advice from the expertise of experts and non-experts alike.

Many of the authors highlight the need to analyse how we anticipate and manage the outputs (beneficial and harmful) of health research. For example, Coleman (Chapter 13) questions the robustness and objectivity attributed to risk–benefit analysis, despite the heavy reliance placed upon it within health research. Similarly, benefit sharing has become a key requirement for many research projects but, as discussed by Simm (Chapter 15), there are practical challenges to deploying such a complex tool to distinct concrete projects. Patents are also a standard feature of health research and innovation. As considered by Nicol and Nielsen (Chapter 14), these can be used both as a positive incentive to foster innovation and, paradoxically, as a means to stifle collaboration and resource sharing.

Three final cross-cutting themes must be kept in mind as we continue to attempt to improve health research regulation. First, in closing this section, Nicholls (Chapter 20) reminds us that we must be mindful of the constant need to evaluate and adapt our approaches to the varying contexts and ongoing developments in health research regulation. Second, in recognition of the fragility of public trust and the necessity of public confidence for health research initiatives to succeed, we must continue to strive for transparency, fairness and inclusivity within our practices. Finally, as we seek to refine and develop new approaches to health research regulation, we must acknowledge that no one tool or process can provide a panacea for the complex array of values and interests at stake. All must be kept under constant review as part of a well-functioning learning system, as Laurie argues in the Afterword to this volume.

10 Consent

Jane Kaye and Megan Prictor

10.1 Introduction

Informed consent is regarded as the cornerstone of medical research; a mechanism that respects human dignity and enables research participants to exercise their autonomy and self-determination. It is a widely accepted legal, ethical and regulatory requirement for most health research. Nonetheless, the practice of informed consent varies by context, is subject to exceptions, and, in reality, often falls short of the theoretical ideal.¹ The widespread use of digital technologies this century has revolutionised the collection, management and analysis of data for health research, and has also challenged fundamental principles such as informed consent. The previously clear boundaries between health research and clinical care are becoming blurred in practice, with implications for implementation and regulation. Through our analysis we have identified the key components of consent for research articulated consistently in international legal instruments. This chapter will: (1) describe the new uses of data and other changes in health research; (2) discuss the legal requirements for informed consent for research found in international instruments; and (3) discuss the challenges in meeting these requirements in the context of emerging research data practices.

10.2 The Changing Nature of Research

Health research is no longer a case simply of the physical measurement and intimate observation of patients. Rather, it increasingly depends upon the generation and use of data, and new analysis tools such as Artificial Intelligence (AI). Health research has been transformed by innovations in digital technologies enabling the collection, curation and management of large quantities of diverse data from multiple sources. The intangible nature of digital data means that it can be perfectly replicated indefinitely, instantly shared with others across geographical borders and used for multiple purposes, such as clinical care and research. The information revolution enables data to be pulled from different sources such as electronic medical records; wearables and smart phones monitoring chronic conditions; and datasets outside the health care system yielding inferences about an individual’s health. These developments have significant implications for informed consent.

New technologies have enabled the development of ambitious scientific agendas, new types of infrastructure such as biobanks and genomic sequencing platforms and international collaborations involving datasets of thousands of research participants. Much innovation is driven by collaborations between clinical and research partners that provide practical need and clinical data, and companies offering technical expertise and resources. Examples are: national genomic initiatives including Genomics England (UK), All of Us (USA), Aviesan (France), Precision Medicine Initiative (China); international research collaborations like the Human Genome Project, Global Alliance for Genomics and Health, the Personal Genome Project; and mission-orientated collaborations such as Digital Technology Supercluster (Canada) and the UK Health Data Research Alliance.

The greatest challenges emerge around informed consent in these new contexts where already-collected data can be used in ways not anticipated at the time of collection and data can be sent across jurisdictional borders. When data and tissue samples are being collected for multiple unknown future research uses, explicit informed consent to the research aims and methods may not be possible. In response to this practical challenge, the World Medical Association (WMA) adopted the Declaration of Taipei on Ethical Considerations regarding Health Databases and Biobanks (2002, revised 2016). It stipulates that instead of consenting to individual research, individuals may validly consent to the purpose of the biobank, the governance arrangements, privacy protections, risks associated with their contribution and so on. This form of ‘broad consent’ is really an agreement that others will govern the research, since determinations about appropriate uses of the data and biomaterials are decided by researchers with approval by research ethics committees or similar bodies.²

10.3 The Basis for Informed Consent

The moral force of consent is not unique to health research; it is integral to many interpersonal interactions, as well as being entrenched in societal values. The key moral values at play in medical research are: autonomy – the right for an individual to make his or her own choice; beneficence – the principle of acting with the participant’s best interests in mind; non-maleficence – the principle that ‘above all, do no harm’; and justice – emphasising fairness and equality among individuals.³ The concepts of voluntariness and transparency embedded in informed consent speak to the ethical value of respect for human beings, their autonomy, their dignity as free moral agents and their welfare. This respect for individuals has resulted in special protections for those who are not legally competent to provide informed consent. Beneficence requires that the probable benefits of the research project outweigh the harms. In the context of informed consent, non-maleficence demands that harm is minimised by researchers being attuned to participant welfare and fully disclosing likely benefits and risks to permit adequately informed choice. The principle of justice in the research setting requires that potential participants are equally provided with adequate information to make a knowledgeable decision, helping to avoid participant exploitation. Consideration of the ethical principles underpinning informed consent also requires reflection on cultural values, such as those pertaining to specific indigenous communities or ethnic groups. Cultural values may lead researchers to consider, for example, whether unique harms to cultural integrity and heritage could accrue to certain groups through specific research projects, and whether respect for human beings should be seen through a lens of collective, as well as individual, autonomy and well-being.⁴ These ethical principles underpin informed consent in health research practice, but not all of them have been implemented into law.

10.4 Legal Requirements for Informed Consent

The requirements for informed consent emerged from a range of egregious examples of physical experimentation on humans. Among the most notable examples were the Nuremberg trials following World War II, although concern about harmful research practices internationally had surfaced decades earlier.⁵ The trial of Nazi doctors produced a ten-point Code that became the foundation of modern health research ethics. Voluntary consent was its first and arguably most emphasised principle.⁶ It has since been espoused in declarations by international and non-governmental organisations. A key instrument is the WMA’s Declaration of Helsinki (1964, as amended) setting out the basic requirements for informed consent for research.

In medical research involving human subjects capable of giving informed consent, each potential subject must be adequately informed of the aims, methods, sources of funding, any possible conflicts of interest, institutional affiliations of the researcher, the anticipated benefits and potential risks of the study and the discomfort it may entail, post-study provisions and any other relevant aspects of the study. The potential subject must be informed of the right to refuse to participate in the study or to withdraw consent to participate at any time without reprisal.⁷

Crucial to this formulation is the need to communicate and provide detailed information to the ‘human subject’. While this information should be comprehensive enough for participants to make an informed decision, it positions the researcher as the information provider and the subject as a passive recipient. Yet, the Declaration also posits ongoing engagement as an essential requirement as the participant can withdraw consent at any time.

The principle of free consent also forms part of the United Nations’ International Covenant on Civil and Political Rights (Article 7). Further guidelines and conventions promulgated by international organisations such as the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH),⁸ the Council for International Organizations of Medical Sciences,⁹ and the Council of Europe,¹⁰ endorse and explain these principles. The ICH Good Clinical Practice Guideline considers consent in the context of human clinical trials; it establishes a unified quality standard for the European Union, Japan and the USA. The Oviedo Convention and the 2005 Additional Protocol relating to biomedical research similarly foreground consent, stipulating that it be ‘informed, free, express, specific and documented’.¹¹ The European General Data Protection Regulation (GDPR) has raised the bar for informed consent for data use worldwide. In Australia, the National Health and Medical Research Council’s National Statement on Ethical Conduct in Human Research (2007, updated 2018) is the principle guiding document for health research. From these documents, several key components can be discerned, such as competence, transparency and voluntariness, and that consent must be informed.

Only ‘human subjects capable of giving informed consent’ are the subject of the Helsinki Declaration statement about consent. Ethicists have described competent people as those who have ‘the capacity to understand the material information, to make a judgement about the information in light of his or her values, to intend a certain outcome, and to freely communicate his or her wish to caregivers or investigators’.¹² Special protections pertain to those not competent to give consent, such as some young children and some people who are physically, mentally or intellectually incapacitated. These protections centre upon authorisation by a research ethics committee and consent provided by a legal representative. The potential participant may still be asked to assent to the research.

Assessing competence represents a challenge in relation to biobanks and other longitudinal research endeavours where people contributing data or tissue samples may have shifting competence over time; for instance people who were enrolled into research as children will become competent to provide consent for themselves as they reach adulthood.¹³ People impacted by cognitive decline or mental illness may lose competence to provide consent, either temporarily or indefinitely. Periodically revisiting consent for participants is an ethically appropriate, yet logistically demanding, response.

As indicated above, the Nuremberg Code and the Declaration of Helsinki outline a range of information that potential research participants are to be given to enable them to be informed before making a choice about enrolment. The ICH Guideline goes into further detail regarding clinical trials, stating that the information should be conveyed orally and in writing (4.8.10), and that that the explanation should include:

• Whether the expected benefits of the research pertain to the individual participants;

• What compensation is available if harm results;

• The extent to which the participant’s identity will be disclosed;

• The expected duration of participation;

• How many participants are likely to be involved in the research.

National or regional statutes and guidelines stipulate the required informational elements for consent to health research in their jurisdictions, mirroring the elements contained in the international instruments to varying degrees.¹⁴

Limited disclosure of information may sometimes be permitted, for instance in a study of human behaviour where the research aims would be frustrated by full disclosure to participants.¹⁵ It may also be a necessary consequence of the difficulty of comprehensive disclosure in the context of Big Data science, where not all the uses of the data (that may not be collected directly from the individual) can be anticipated when the data are collected.

The Declaration of Helsinki requirement that research participants must be ‘adequately informed’ points to further consideration of how best to communicate the complex information described above. This is the focus of much recent law and guidance.¹⁶ Research has shown repeatedly that participants often do not understand the investigative purpose of clinical trials, key concepts such as randomisation and the risks and benefits of participation.¹⁷ Using simple language and providing enough time to consider the information can help, as well as tailoring information to participant age and educational level. Researchers have evaluated tools to assist with communicating information in ways that support understanding.¹⁸ Complex, heterogenous and changing research endeavours that cross geographic boundaries and blur the lines between clinical care, daily life and research pose an additional challenge to the requirement for transparency.

A consistent requirement of international conventions, law and guidelines for ethical research is that for consent to be valid, it must be voluntary.¹⁹ The Nuremberg Code obliges researchers to avoid ‘any element of force, fraud, deceit, duress, over-reaching, or other ulterior form of constraint or coercion’.²⁰ Beyond the problem of overt coercion by another person, other considerations in evaluating voluntariness include: deference to the perceived power of the researcher or institution;²¹ the mere existence of a power imbalance;²² the existence of a dependent relationship with the researcher;²³ and the amount paid to participants.²⁴ On power and vulnerabilities, see further Brassington, Chapter 9, this volume.

These concerns are largely associated with duress as a result of specific relationships developed through personal interactions. In Big Data or AI analysis, the concept of voluntariness must be reconsidered, as often the data users are not known to the data subject and the nature of the duress may not be straightforwardly attributed to particular relationships. An example is companies that provide direct-to-consumer genetic tests, where the provision of test results also enables the companies to use the data for purposes including marketing and research. This is a different kind of duress as people lured through the fine print in click-wrap contracts are then enrolled into research.²⁵

Traditionally, valid informed consent occurs before the participant’s involvement in the research;²⁶ no specific timing is recommended as long as there is time for the person to acquire sufficient understanding of the research. In selected circumstances, ‘deferred’ consent – where individuals do not know they are enrolled in a clinical trial so that the sample is not biased and they are asked for consent later on²⁷ – a waiver of consent or an opt-out approach might be justifiable. These are typically addressed within relevant guidance.²⁸ Once-off informed consent before a project starts may, however, be insufficient to acquit researchers’ responsibilities in the context of longitudinal data-intense research infrastructures. Modalities that permit ongoing or at least repeated opportunities to refresh consent, such as staged consent and Dynamic Consent, considered below, are a developing response to this issue.

It is a key principle of health research, traceable back to the Declaration of Helsinki, that potential research participants have a right to decline the invitation to participate without giving a reason and should not incur any disadvantage or discrimination as a consequence.²⁹ Further, people who have consented must be free to withdraw consent at any time without incurring disadvantage. The GDPR stipulation that ‘It shall be as easy to withdraw as to give consent’,³⁰ has energised research into technology-based tools to facilitate seamless execution of a withdrawal decision, or even to support shifting levels of participation over time.³¹

Newer research methods and infrastructures characterised by open-ended research activities and widespread data sharing add complexity to the interpretation of ‘withdrawing consent’. International guidelines have acknowledged that withdrawal in this context might equate to no new data collection while raising a question over whether existing samples and data must be destroyed or remain available for research.³²

10.5 The Limitations of Consent

In research involving human participants, the informed consent process is foregrounded.

As a legal mechanism intended to protect human subjects in the way envisaged by international instruments, it is also recognised that consent may be insufficient. People often do not understand what they have agreed to participate in, retain the information about the research or even recall that they agreed to be involved.³³ Consent is not the only legal basis for conducting health research. While there is variation between jurisdictions, broadly speaking research involving data or tissue may be able to proceed without consent in certain circumstances. These include if: there is an overriding public interest and consent is impracticable; there is a serious public health threat; the participant is not reasonably identifiable; or the research carries low or negligible risk. Many researchers have sought to augment traditional modes of consent at the point of entry to research, to support informed decision-making by potential participants. New consent processes seek to enable truly informed consent rather than doing away with this fundamental requirement.

Traditionally, consent is operationalised as a written document prepared by the researcher setting out the information described above. The participant’s agreement is indicated by their signature and date on the document. Concerns about participant problems with reading and understanding the form have led to initiatives including simplified written materials, extra time and the incorporation of multimedia tools.³⁴ More nuanced consent modalities might encompass different tiers of information – with simple, minimally compliant information presented first, linking to more comprehensive explanation – and different staging of information, for instance with new choices being presented to participants at a later time.³⁵

Scholars have also considered when and how it might be appropriate to diverge from the notion of the individual human subject as the autonomous decision-maker for health research participation, towards a communitarian approach informed by ethical considerations pertaining to culture and relationships. The concept of informed consent must, in this context, expand to incorporate the possibility of family and community members at least being consulted, perhaps even deciding jointly. Osuji’s work on relational autonomy in informed consent points to decisions ‘made not just in relation to others but with them, that is, involving them: family members, friends, relations, and others’.³⁶ This approach might particularly suit some groups, with extensive examples deriving from Australian aboriginal and other Indigenous communities,³⁷ family members with shared genetic heritage³⁸ and some Asian and African cultures.³⁹ Communitarian-based consent processes may not meet legal requirements for informed consent to research, but may nevertheless be a beneficial adjunct to standard processes in some instances.

10.6 New Digital Consent Mechanisms

The pervasion of technology into all aspects of human endeavour has transformed health research activities and the consent processes which support them. Electronic consent may mean simply transferring the paper form to a computerised version. Internationally, electronic signatures are becoming generally accepted as legally valid in various contexts.⁴⁰ These may comprise typewritten or handwritten signatures on an electronic form, digital representations such as fingerprints or cryptographic signatures. Progress is being made on so-called digital, qualified or advanced electronic signatures which can authenticate the identity of the person signing, as well as the date and location.⁴¹

Semi-autonomous consent is emerging in computer science; it refers to an approach in which participants record their consent preferences up-front, a computer enacts these preferences in response to requests – for instance, invitations to participate in research – and the participants review the decisions, refine their expressed preferences and provide additional information.⁴² This could be a way to address consent fatigue by freeing participants from the need to make numerous disaggregated consent decisions. It is a promising development at a time when increasing uses of people’s health data for research may overwhelm traditional tick-box consent.

Dynamic Consent is an approach to consent developed to accommodate the changes in the way that medical research is conducted. It is a personalised, digital communication interface that connects researchers and participants, placing participants at the heart of decision-making. The interface facilitates two-way communication to stimulate a more engaged, informed and scientifically-literate participant population where individuals can tailor and manage their own consent preferences.⁴³ In this way it meets many of the requirements of informed consent as stipulated in legal instruments⁴⁴ but also allows for the complexity of data flows characterising health research and clinical care. The approach has been used in the PEER project,⁴⁵ CHRIS,⁴⁶ the Australian Genomics Health Alliance⁴⁷ and the RUDY project.⁴⁸ It seems appropriate to have digital consent forms for a digital world that allow for greater flexibility and engagement with patients when the uses of data for research purposes cannot be predicted at the time of collection.

10.7 Conclusion

The organisation and execution of health research has undergone considerable change due to technological innovations that have escalated in the twenty-first century. Despite this, the requirements of informed consent enshrined in the Nuremberg Code are still the basic standard for health research. These requirements were formulated specifically in response to atrocities that occurred through physical experimentation. They continue to be applied to data-based research that is very different in its scope and nature, and in the issues it raises for individuals compared to physically-based research, that was the template for the consent requirements found in international instruments. The process for obtaining and recording consent has undergone little change over time and is still recorded through paper-based systems, reliant on one-to-one interactions. While this works well for single projects with a focus on the prevention of physical, rather than informational harm, it is less suitable when data are used in multiple settings for diverse purposes.

Paper-based systems are not flexible and responsive and cannot provide people with the information that is needed in a changing research environment. Digital systems such as Dynamic Consent provide the tools for people to be given information as the research evolves and to be able to change their mind and withdraw their consent. However, given the complexity and scale of research, when data are collected from a number of remote data points it is difficult for consent to effectively respond to all of the issues associated with data-intensive research. The use of collective datasets that concern communal or public interests are difficult to govern through individual decision-making mechanisms such as consent.⁴⁹

Consent is only one of the many governance mechanisms that should be brought into play to protect people involved in health research. Additionally, attention should be given to the ecosystem of research and informational governance that consist of legal requirements, regulatory bodies and best practice that provide the protective framework that is wrapped around health research. Despite its shortcomings, informed consent is still fundamental to health research, but we should recognise its strengths and limitations. More consideration is needed on how to develop better ways to enable the basic requirements of informed consent to be enacted through digital mechanisms that are responsive to the characteristics of data-intensive research. Further research needs to be directed to how the governance of health research should adapt to this new complexity.

11 Forms of Engagement

Mhairi Aitken and Sarah Cunningham-Burley

11.1 Introduction

Public engagement (PE) is part of the contemporary landscape of health research and innovation and considered a panacea for what is often characterised as a problem of trust in science or scientific research, as well as a way to ward off actual or potential opposition to new developments. This is quite a weight for those engaging in engagement to carry, and all the more so since PE is often underspecified in terms of purpose. PE can mean and involve different things but such flexibility can come at the price of clarity. It may allow productive creativity but can limit PE’s traction.¹

In this chapter we provide a synthesis of current conceptualisations of PE. We then consider what kinds of publics are ‘engaged with’ and what this means for the kinds of information exchanges and dialogues that are undertaken. Different forms of PE ‘make up’ different kinds of publics: engagements do not, indeed cannot, start with a clean sheet – neither with a pure public nor through a pure engagement.² As Irwin,³ among others, has noted, PE is a political exercise and this wider context serves to frame what is engaged about. It is therefore all the more important to reflect on the practice of PE and what it is hoped will be achieved. We argue that clarity and transparency about the intention, practice and impact of PE are required if PE is to provide an authentic and meaningful tool within health research governance.

11.2 Engaging with Critique

PE has been a subject of debate for many years, particularly in the Science and Technology Studies literature, through what is termed critical public understanding of science. From Wynne’s⁴ seminal work onwards, this critique has championed the range of expertise that can come to bear on matters scientific and has provided analytical verve to critiques of the institutional arrangements of both science and PE. Criticisms of top-down models of PE were dominant throughout the 1990s and the ‘deficit model of public understanding’ was roundly debunked not least for suggesting that public ignorance of science was a fundamental cause of loss of trust. This critique played an important role in bringing about a new emphasis on two-way processes of PE that went beyond ‘educating the public’.⁵ New commitments to dialogue and engagement – ‘the participatory turn’ – have become more commonplace and mainstream.⁶ However, as Stilgoe and colleagues have commented, the shift from deficit model approaches to dialogic PE, has been only partially successful:

It has been relatively easy to make the first part of the argument that monologues should become conversations. It has been harder to convince the institutions of science that the public are not the problem. The rapid move from doing communication to doing dialogue has obscured an unfinished conversation about the broader meaning of this activity.⁷

Herein lies further threats to the integrity of PE.

PE is now a component of much health research where engagement or patient and public involvement is often a funding requirement. This is particularly pronounced in the UK where public understanding and engagement in science has gained increasing institutional traction since the House of Lords report in 2000. For some, the deficit model of public understanding has simply been replaced with a deficit model of public trust, to which ‘more understanding’ and, even, ‘more dialogue’ remain a solution.⁸ So, on the one hand the deficit model of publics in need of education about science lingers on, sometimes under the guise of trust. Yet, on the other, we see PE being taken up across sectors – and there is evidence of PE, sometimes, bringing science and its governance to account.

PE can be productive as many commentators have posited.⁹ The task for health research governance is to ensure that participatory practices are not skewed towards institutional ends but allow diverse voices into the policy making process so that they can make a difference to how health research is conducted, regulated and held accountable to the very publics it purports to serve. As Braun and Schultz note:

The question that is increasingly discussed in public understanding of science (PUS) today is not so much whether there is a trend towards participation but what we are to make of it, how to assess it, how to understand the dynamics propelling it, how to systematise and interpret the different forms and trajectories it takes, what the benefits, pitfalls or unintended side-effects of these forms and trajectories are and for whom.¹⁰

We now turn to consider some of these questions.

11.3 Forms of Public Engagement

Enthusiasm for, and professed commitment to, PE does not easily translate into meaningful engagement in practice. This is in no small part due to the fact that the term ‘public engagement’ can be interpreted in many different ways and PE is undertaken for a variety of reasons.

Key challenges around PE are that the different ideas about its role and value manifest in a variety of purposes and rationales, whether implicit or explicit. PE can be underpinned by normative, substantive or instrumental rationales.¹¹ A normative position suggests that PE should be conducted as it is ‘the right thing to do’ – something that is part and parcel of both public and institutional expectations. An instrumental position regards PE as a means to particular ends. For example, PE might be conducted to secure particular outcomes such as greater public support for a policy or project. Such a position aligns PE closely with institutional aims and objectives: it promotes public support through understanding and addressing public concerns. A substantive position suggests that the goal of PE is to lead to benefits for participants or wider publics: this can include empowering members of the public, enhancing skills or building social capital.¹² While these varying rationales are not mutually exclusive, they lead to different understandings and expectations regarding the objectives and role of PE, as well as different ideas of what it means for such processes to be ‘successful’.

Rowe and Frewer argue that public involvement ‘as widely understood and imprecisely defined can take many forms, in many different situations (contexts), with many different types of participants, requirements, and aims (and so on), for which different mechanisms may be required to maximize effectiveness (howsoever this is defined)’.¹³ Choosing between different forms requires consideration of purpose and an awareness of the wider context within which engagement is taking place; its effectiveness is more than a matter of method. Academic and practitioner literatures on PE contain many different typologies and classifications of forms of engagement. These often take as their starting point Arnstein’s¹⁴ ladder of public participation. This sets out eight levels of participation, in the form of a hierarchy of engagement. On the bottom rung of the ladder (non-participation), engagement is viewed instrumentally as an opportunity to educate the public and/or engineer support, a common effort when seeking to fill a knowledge deficit or garner social support for a new development. In the middle of the ladder, tokenistic forms of participation include informing and consulting members of the public, where consultation does not involve a two way process, but rather positions the public as having views and attitudes that might be helpful to seek as part of policy development. Again, this is not an unusual mode of engagement in the context of health research. Arnstein suggested that both of these could be valuable first steps towards participation but that they are limited by the lack of influence that participants have. Consultation is described as being a cosmetic ‘window-dressing ritual’ with little impact, although the extent of impact would depend on how the results of any consultation are subsequently used, rather than being intrinsic to the method itself. The top rungs of the ladder, which move towards empowerment and ownership of process, require redistribution of power to members of the public; while the participatory turn gestures towards such an approach, institutional practices often militate against its enactment.

Arnstein’s model has been adapted by a large number of individuals and organisations in developing alternative classification systems and models. This has resulted in a proliferation of typologies, tool kits and models which can be referred to in designing and/or evaluating PE approaches. Aitken has observed that these models, whilst adopting varying terminology and structures, typically follow common patterns:

Each starts with a ‘bottom’ layer of engagement which is essentially concerned with information provision […] They then have one (or more) layer(s) with limited forms of public feedback into decision-making processes (consultation), and finally they each have a ‘top’ layer with more participatory forms of PE which give greater control to participants.¹⁵

Forms of engagement classified as ‘awareness raising’ are essentially concerned with the dissemination of information. Where awareness raising is conducted on its own (i.e. where this represents the entirety of a PE approach) this represents a minimal form of PE. It may even be argued that awareness raising on its own – as one-sided and unidirectional information provision – should not be considered PE. Rowe and Frewer note that at this level, ‘information flow is one-way: there is no involvement of the public per se in the sense that public feedback is not required or specifically sought’.¹⁶ Awareness raising is limited in what it can achieve, but the focus on increasing understanding of particular issues may be a prerequisite for the deliberative approaches discussed below.

Examples of PE activities focussed on awareness raising include campaigns by national public health bodies such as Public Health England’s ‘Value of Vaccines’,¹⁷ or the creation and dissemination of videos and animations to explain the ways that people’s health data is used in research.¹⁸

Consultation aims to gather insights into the views, attitudes or knowledge of members of the public in order to inform decisions. It can involve – to varying degrees – two-way flows of information. Wilcox contends that: ‘Consultation is appropriate when you can offer some choices on what you are going to do – but not the opportunity [for the public] to develop their own ideas or participate in putting plans into action’.¹⁹ Consultation provides the means for public views to be captured and taken into consideration, but does not necessarily mean that these views, or public preferences and/or concerns will be acted on or addressed.

Consultation can be either a one-way or two-way process. In a one-way process, public opinion is sought on pre-defined topics or questions, whereas a two-way process can include opportunities for respondents to reflect on and/or question information provided by those running engagement exercises.²⁰ Such two-way processes can ensure the questions asked, and subsequently the responses given, reflect the interests and priorities of those being engaged. It can also facilitate dialogue and ‘deeper’ forms of engagement with the aim of characterising, in all their complexity, public attitudes and perspectives.

It is widely recognised that consultation will be best received and most effective when it is perceived to be meaningful. This means that participants want to know how their views are taken into account and what impact the consultation has had (i.e. how has this informed decision-making). Davidson and colleagues caution that: ‘Consultation can be a valuable mechanism for reflecting public interests, but can also lead to disappointment and frustrations if participants feel that their views are not being taken seriously or that the exercise is used to legitimise decisions that have already been made’.²¹ Again, we see that choice of method is no guarantee of meaningful engagement in terms of influence on the practices of research and its governance.

Approaches taken to consultation include: public consultations where any member of the public is able to submit a written response; surveys and questionnaires with a sample which aims to be representative of the wider population (or key groups within it); and, focus groups, deliberative engagement or community-based participatory methods to engage more deeply with communities to shape both research processes and outcomes.

Approaches to PE that can be classified under the heading of empowerment are those that would be positioned at the top of Arnstein’s ladder of participation. These approaches involve the devolution of power to participants and the creation of benefits for participants and/or wider society. This can be achieved through public-led forms of engagement where public members themselves design the process and determine its objectives, topics of relevance and scope or through partnership approaches.²² It might also be achieved through engagement approaches that bring together public members in ways that build relationships and social capital that will continue after the engagement process ends.²³ Both invited and uninvited²⁴ forms of engagement can involve empowerment, so it is possible to engineer a flattening of hierarchies of knowledge and expertise as well as respond to efforts of publics to come together to define and debate issues of concern.

Empowering forms of engagement can lead to outcomes of increased relevance to communities and that most accurately reflect public interests and values. However, they can also be more expensive than traditional forms of engagement, given that they necessitate more open and flexible timeframes and may require extra skills related to facilitation and negotiation. Certainly, they may confront the more uncomfortable social, political and economic consequences and drivers of health research.

One example of engaging with some of the wider issues raised by health data and research is the dialogue commissioned by the Scottish Government to deliberate about private and third sector involvement in data sharing.²⁵

While a hierarchical classification, such as Arnstein’s, serves to highlight the importance of how the public are positioned in different modes of engagement, each broad approach described above can add different value and play important roles in PE. In practice it may be most appropriate for PE to use a range of methods reflecting different rationales and objectives. Rather than conceptualising them hierarchically, it is more helpful to think of these methods as overlapping and often working alongside each other within any PE practice or strategy.

11.4 Types of Publics

PE and involvement professionals, policy documents and critical scholars increasingly refer to ‘publics’ as a way to problematise and differentiate within and between different kinds of public. The adoption of such a term signifies that publics are diverse and that we cannot talk of a homogeneous public. However, beyond that, the term may obscure more than it reveals: what kinds of publics are we talking about when we talk about PE, and how are these related to particular forms of engagement? As Braun and Schultz note ‘“The public,” we argue, is never immediately given but inevitably the outcome of processes of naming and framing, staging, selection and priority setting, attribution, interpellation, categorisation and classification’.²⁶ How members of ‘the public’ are recruited is more than a practical matter: the process embodies the assumptions, aims and priorities of those designing the engagement.

On the whole, publics are constructed or ‘come into being’ within PE practices rather than being self-forming. As with types of PE, different categorisations of publics have been developed. Degeling and colleagues highlight three different types: citizens (ordinary people who are unfamiliar with the issues, a kind of pure public); consumers (those with relevant personal experience, a kind of affected public) and advocates (those with technical expertise or partisan interests).²⁷ And each of these was linked to different types of PE. Citizens were treated as a resource to increase democratic legitimacy; consumers were directed to focus on personal preferences; advocates were most commonly used as expert witnesses in juries – directly linked to policy processes. However, overall the ‘type’ of public sought was often not explicit, and their role not specified.

Braun and Schultz²⁸ elaborate a four-fold distinction: the general public, the pure public, the affected public and the partisan public. Different PE methods serve to construct different kinds of publics. The general public is a construct required for opinion polls and surveys; pure publics for citizen conferences and juries; affected publics for consultative panels; partisan publics for stakeholder consultations. However, as with the different types of PE, in practice there will be overlaps across these dimensions and subject positions will shift as expertise is crafted through the processes of engagement and facilitation.²⁹ Different types of expertise are presumed here too: the general public gives policy makers knowledge about people’s attitudes; the pure public creates a ‘mature’ citizen who becomes knowledgeable and can develop sophisticated arguments; affected publics bring expertise to ‘educate’ the expert – very common in health research regulation; and a partisan public may be deliberately configured to elicit viewpoints ‘out there’ in society to assess the ‘landscape of possible argument’.³⁰

Types of PE and the categorisation of different publics involve processes of inclusion and exclusion and the legitimacy of PE can easily be challenged because of who participates: some voices may be prioritised over others, and challenges may be made to participants’ expertise. We turn now to a case study of how PE is being enacted in one area of health research to explore how we might deal with these problematics of how and who.

11.5 Public Engagement in Data Intensive Health Research: Principles for an Inclusive Approach

The digitisation of society has led to an explosion of interest in the potential uses of more and more population data in research; this is particularly true in relation to health research.³¹ However, recent years have also brought a number of public controversies, particularly regarding proposed uses of health data. Two high profile examples from England are the failed introduction of the care.data scheme to link hospital and GP records³² and Google Deep Minds’ involvement in processing health data at an NHS Trust in London.³³ The introduction of Australia’s National Electronic Health Record Systems (NEHRS) also floundered, demonstrating the importance of taking account how such programmes reflect, or jar, with public values.³⁴ Such controversies have drawn attention to the importance of engaging with members of the public and stakeholders to ensure that data are used in ways which align with public values and interests and to ensure that public concerns are adequately addressed.

The growing interest in potential uses of population data, and the increasing recognition of the importance of ensuring a social licence for their use, have resulted in considerable interest in understanding public attitudes and views on these topics.³⁵ With the expansion of research uses of (health) data there has been a growing interest in public acceptability. As Bradwell and Gallagher have suggested, ‘personal information use needs to be far more democratic, open and transparent’ and this means ‘giving people the opportunity to negotiate how others use their personal information in the various and many contexts in which this happens’.³⁶ PE is seen as key to the successful gathering and use of health data for research purposes.

As a recent consensus statement on PE in data intensive health research posits, there are particular reasons to promote PE in data intensive health research³⁷ including its scale – here the wider public is an ‘affected’ public and the distance is increased between researchers and those from whom data are gathered, thus requiring a new kind of social licence.³⁸ This requires novel thinking about how best to engage publics in shaping acceptable practices and their effects.

As well as recognising diverse practices, aims and effects, and building reflexive critique into PE for health research regulation and governance, we need to articulate some common commitments that can help steer a useful path through this diversity and thereby challenge criticisms of institutional capture and tokenism.³⁹ These commitments must include clarity of purpose and transparency, which will help deal with the challenges of multiple but often implicit purposes and goals. Inclusion and accessibility will broaden reach and two way communication – dialogue – is a necessary but not sufficient condition for impact. The latter can only be achieved if there is institutional buy-in, a commitment to respond to and utilise PE in governance and research. Given the challenges of assessing whether or not PE is impactful, something we discuss in the conclusion below, PE should be designed with impact in mind and be evaluated throughout. It is clear that you cannot straightforwardly get the right public and the right mechanism and be assured of meaningful and impactful PE. The choices are complicated and inflected with norms and goals that need to be explicitly stated and indeed challenged.

We now turn, as a conclusion, to review some of the outstanding issues that a critical approach to PE brings and make the case for robust evaluation.

11.6 Conclusion

The prominent emphasis on PE in relation to health research can be seen as a reflection of a wider resurgence of interest in PE in diverse policy areas.⁴⁰ For example, Coleman and Gotzehave pointed to a widespread commitment to PE, conceived of as a mechanism for addressing problems in democratic societies.⁴¹ For Wilsdon and Willis, the emphasis on engagement represents a wider pattern whereby the ‘standard response’ of government to public ambivalence or hostility towards technological, social or political innovation is ‘a promise to listen harder’.⁴²

PE is not straightforward, and fulfilling the commitments of PE presents challenges and dilemmas in practice. There are many different ways of approaching PE, and these lead to different ideas of what constitutes success. There is no agreed best practice in evaluation; different rationales lead to different approaches to evaluation. Approaches underpinned by normative rationales will evaluate the quality of PE processes (Was it done well?); instrumental rationales lead to a focus on outcomes (Was it useful? Did it achieve the objectives?); and substantive rationales will assess the value added for participants or wider society (Did participants benefit from the process? Were there wider positive impacts?). Evaluation following substantive rationales is typically focussed on longer term outcomes, compared to evaluation following normative or instrumental rationales. Such longer term outcomes may be indirect and difficult to quantify or measure.

While the literature on methods of doing PE continues to proliferate, evaluation of PE remains under-theorised and underreported. The current evidence base is limited, but existing approaches to evaluating PE tend to reflect instrumental rationales and focus on direct outcomes of PE rather than substantive rationales and indirect, less tangible outcomes or impacts.⁴³ Wilson and colleagues⁴⁴ have observed that there is a tendency to focus on ‘good news’ in evaluating PE and that positivist paradigms shaping research projects or programmes can limit the opportunities to fully or adequately evaluate the complexities of PE as a social process.

This is significant as it means that while a variety of rationales and purposes are acknowledged in relation to PE, there is very limited evidence of the extent to which these are realised. This in turn has negative implications for the recognition – and consequently, the institutional support – that PE receives. By providing evidence only of narrow and direct outcomes, instrumental approaches to evaluation obscure the varied and multiple benefits that can result from PE. While ‘the move from “deficit to dialogue” is now recognised and repeated by scientists, funders and policymakers […] for all of the changing currents on the surface, the deeper tidal rhythms of science and its governance remain resistant’.⁴⁵ Despite growing emphasis on dialogue and co-inquiry, simplistic views of the relationship between science and the public persist⁴⁶ and PE is often conducted in instrumental ways which seek to manufacture trust in science rather than foster meaningful dialogue. Greater reflection is required on the question of why publics are engaged rather than how they are engaged.

Finally, in designing, conducting and using PE in health research, we need to be reflective and critical, asking ourselves whether the issues are being narrowly defined and interpreted within existing frameworks (that often focus on privacy and consent). Does this preclude wider discussions of public benefit and the political economy of Big Data research for health? PE can and should improve health research and its regulation by questioning institutional practices and societal norms and using publics’ contributions to help shape solutions.

12 Participatory Governance in Health Research Patients and Publics as Stewards of Health Research Systems

Kim H. Chuong and Kieran C. O’Doherty

12.1 Introduction

This chapter discusses participatory governance as a conceptual framework for engaging patients and members of the public in health research governance, with particular emphasis on deliberative practices. We consider the involvement of patients and members of the public in institutional mechanisms to enhance responsibility and accountability in collective decision-making regarding health research. We illustrate key principles using discussion of precision medicine, as this demonstrates many of the challenges and tensions inherent in developing participatory governance in health research more generally. Precision medicine aims to advance healthcare and health research through the development of treatments that are more precisely targeted to patient characteristics.

Our central argument in this chapter is that patients and broader publics should be recognised as having a legitimate role in health research governance. As such, there need to be institutional mechanisms for patients and publics to be represented among stewards of health research systems, with a role in articulating vision, identifying research priorities, setting ethical standards, and evaluation. We begin by reviewing relevant scholarship on patient and public engagement in health research, particularly in the context of the development and use of Big Data for precision medicine. We then examine conceptualisations of participatory governance and outline stewardship as a key function of governance in a health research system. Thereafter, we propose the involvement of patients and publics as stewards who share leadership and oversight responsibilities in health research, and consider the challenges that may occur, most notably owing to professional resistance. Finally, we discuss the conditions and institutional design elements that enable participatory governance in health research.

12.2 Patient and Public Engagement in Health Research

Beresford identifies two broad approaches that have predominated in public engagement in health and social research since the 1990s.¹ Consumerist approaches reflect a broad interest in the market and seek consumer feedback to improve products or enhance services; in contrast, democratic approaches are concerned with people having more say in institutions or organisations that have an impact on their lives. Unlike consumerist approaches, democratic approaches are explicit about issues of power, the (re)distribution of power and a commitment to personal and collective empowerment. Well-known examples of democratic approaches include the social movements initiated by people living with disability and HIV/AIDS, where these communities demanded greater inclusion in the development of scientific knowledge and health policy decisions.² Moral and ethical reasons based on democratic notions of patient empowerment and redistribution of power, and consequentialist arguments that patient and public engagement can improve research credibility and social acceptance, are also offered by health researchers.³ It should be noted that patient and public engagement does not, in and of itself, constitute an active role for members of the public in health research and policy decision-making. Conceptual models have often highlighted the multiple forms that engagement can take, which vary in the degree to which members of the public are empowered to participate in an active role (see Aitken and Cunningham-Burley, Chapter 11).

In recent years, the potential to link large data sources and harness the breadth and depth of such Big Data has been hailed as bringing ‘a massive transformation’ to healthcare.⁴ Data sources include those collected for health services (e.g. electronic health records), health research (e.g. clinical trials, biobanks, genomic databases), public health (e.g. immunisation registries, vital statistics), and other innovative sources (e.g. social media). Achieving the aims of precision medicine relies on the creation of networks of diverse data sources and scientific disciplines to capture a more holistic understanding of health and disease.⁵ Conducting research using such infrastructure represents a shift from individual and isolated projects to research enterprises that span multiple institutions and jurisdictions. While the challenges of doing patient and public engagement well have been widely recognised, the emergence of precision medicine highlights the stakes and urgency of involving patients and publics in meaningful ways.

Biomedical research initiatives that involve large, networked research infrastructure rely on public support and cooperation. Rhetorical appeals to democratising scientific research, empowerment and public benefits, have been employed in government-sponsored initiatives in the USA and UK in attempts to foster to public trust and cultivate a sense of collective investment and civic duty to participate, notably to agree to data collection and sharing.⁶ Such appeals have been explicit in the US Precision Medicine Initiative (PMI)⁷ since its inception, whereas they have been used post hoc in the NHS England care.data programme after public backlash. The failure of care.data illustrates the importance of effective and meaningful public engagement – rather than tokenistic appeals – to secure public trust and confidence in its oversight for large-scale, networked research. Established to be a centralised data sharing system that linked vast amounts of patient data including electronic health records from general practitioners, care.data was suspended and eventually closed in 2016 after widespread public and professional concerns, including around its ‘opt-out’ consent scheme, transparency, patient confidentiality and privacy, and potential for commercialisation.⁸ See further, Burgess, Chapter 25, this volume.

Research using Big Data raises many unprecedented social, ethical, and legal challenges. Data are often collected without clear indication of their uses in research (e.g. electronic health records) or under vague terms regarding their future research uses (e.g. biobanks). Challenges arise with regard to informed consent about future research that may not yet be conceived; privacy and confidentiality; potential for harms from misuses; return of results and incidental findings; and ownership and benefit sharing, which have implications for social justice.⁹ As cross-border sharing of data raises the challenges of marked differences in regulatory approaches and social norms to privacy, there have been calls for an international comparative analysis of how data privacy laws might have affected biobank practices and the development of a global privacy governance framework that could be used as foundational principles.¹⁰ Arguments have been made that relying on informed consent – which was developed primarily for individual studies – is insufficient to resolve many of the social and ethical challenges in the context of large-scale, networked research; rather, the focus should be on the level of systemic oversight or governance.¹¹ Laurie proposes an ‘Ethics+’ governance approach that appraises biobank management in processual terms.¹² This approach focuses on the dynamics and interactions of stakeholders in deliberative processes towards the management of a biobank, and allows for adaptation to changes in circumstances, ways of thinking, and personnel.

12.3 Participatory Governance in Health Research Systems

The concept of governance has theoretical roots in diverse disciplines and has been used in a variety of ways, with a variety of meanings.¹³ In the health sector, the concept of governance has been informed by a systems perspective, notably the World Health Organization’s framework for health systems.¹⁴ In their review, Barbazza and Tello claim that: ‘Despite the complexities and multidimensionality inherent to governance, there does however appear to be general consensus that the governance function characterizes a set of processes (customs, policies or laws) that are formally or informally applied to distribute responsibility or accountability among actors of a given [health] system’.¹⁵ Common values, such as ‘good’ or ‘democratic,’ and descriptions of the type of accountability arrangement, such as ‘hierarchical’ or ‘networked,’ may be used to denote how governance should be defined. The notion of distributed responsibility or accountability relates to the assertion that governance is about collective decision-making and involves various forms of partnership and self-governing networks of actors.¹⁶

A systems perspective allows for a more integrated and coordinated view of health research activities that may be highly fragmented, specialised and competitive.¹⁷ Strengthening the coordination of research activities promotes more effective use of resources and dissemination of scientific knowledge in the advancement of healthcare. The vision of a learning healthcare system, which was first proposed by the US Institute of Medicine (IOM), illustrates a cycle of continuous learning and care improvement that bridges research and clinical practice.¹⁸ The engagement of patients, their families and other relevant stakeholders is identified as a fundamental element of a learning healthcare system.¹⁹ Engaging patients as active partners in the cycle is argued to both secure the materials required for research (i.e. data and samples) and enhance patient trust.²⁰

Pang and colleagues propose stewardship as a key function within a health research system that has four components: defining a vision for the health research system; identifying research priorities and coordinating adherence to them; setting and monitoring ethical standards; and monitoring and evaluating the system.²¹ Other key functions of a health research system include: financing, which involves securing and allocating research funds accountably; creating and sustaining resources including human and physical capacity; and producing and using research. An important question is therefore how to engage and incorporate the perspectives and values of patients and publics in governance, particularly in terms of stewardship.

Internationally, participatory governance has been explored in multiple reforms in social, economic, and environmental planning and development that varied in design, issue areas and scope.²² Fung and Wright use the term ‘empowered participatory governance’ to describe how such reforms are ‘participatory because they rely upon the commitment and capacities of ordinary people to make sensible decisions through reasoned deliberation and empowered because they attempt to tie action to discussion’.²³ They outline three general principles: (1) a focus on solving practical problems that creates situations for participants to cooperate and build congenial relationships; (2) bottom-up participation, with laypeople being engaged in decision-making while experts facilitate the process by leveraging professional and citizen insights; and (3) deliberative solution generation, wherein participants listen to and consider each other’s positions and offer reasons for their own positions. A similar concept is collaborative governance, which is defined by Ansell and Gash as ‘a governing arrangement where one or more public agencies directly engage non-state stakeholders in a collective decision-making process that is formal, consensus-oriented, and deliberative and that aims to make or implement public policy or manage public programs or assets’.²⁴ The criterion of formal collaboration implies established arrangements to engage publics. Participatory governance is advocated to contribute to citizen empowerment, build local communities’ capacity, address the gap in political representation and power distribution, and increase the efficiency and equity of public services. Unfortunately, however, successful implementation of participatory governance ideals is ‘a story of mixed outcomes’ with the failures still outnumbering the successful cases.²⁵

Yishai argues that the health sector has remained impervious to the practice of participatory governance: patients have not had a substantial voice in health policy decisions, even though they may enjoy the power to choose from different health services and providers as consumers.²⁶ Professional resistance to non-expert views and marginalisation of public interests by commercial interests are cited as some of the reasons for the limited involvement of patients. Similarly, there are concerns that public voices are not given the same weight as those of professionals in health research decision-making. Tokenism, engaging patients as merely a ‘tick-box exercise’ – for funding or regulatory requirements – and devaluing patient input in comparison to expert input are common concerns.²⁷ Furthermore, most engagement efforts are limited to preliminary activities and not sustained across the research cycle; the vast majority of biomedical research initiatives do not engage publics beyond informed consent for data collection and sharing.²⁸

Deliberative practices, such as community advisory boards and citizens’ forums, have been suggested as mechanisms to allow public input in the governance of research with Big Data.²⁹ Public deliberation has been used to engage diverse members of the public to explore, discuss and reach collective decisions regarding the institutional practices and governance of biobanks, and the use and sharing of linked data for research.³⁰ However, in many instances, public input is limited to the point in time at which the deliberative forum is convened. One example of ongoing input is provided by the Mayo Clinic Biobank deliberation, which was used as a seeding mechanism for the establishment of a standing Community Advisory Board. To address the challenge of moving from one-time input to ongoing, institutionalised public engagement, O’Doherty and colleagues propose four principles to guide adaptive biobank governance: (1) recognition of participants as a collective body, as opposed to just an aggregation of individuals; (2) trustworthiness of the biobank, with a reflexive focus of biobank leaders and managers on its practices and governance arrangements, as opposed to a focus on the trust of participants divorced from considerations of how such trust is earned; (3) adaptive management that is capable of drawing on appropriate public input for decisions that substantively affect collective patient or public expectations and relationships; and (4) fit between the particular biobank and specific structural elements of governance that are implemented.³¹

A few cases of multi-agency research networks that engage patients or research participants in governance are also available. For instance, the Patient-Centered Outcomes Research Institute (PCORI) in the USA established multiple patient-powered research networks, each focusing on a particular health condition (www.pcori.org). In the UK, the Managing Ethico-social, Technical and Administrative issues in Data ACcess (METADAC) was established as a multi-study governance infrastructure to provide ethics and policy oversight to data and sample access for multiple major population cohort studies. Murtagh and colleagues identify three key structural features: (1) independence and transparency, with an independent governing body that promotes fair, consistent and transparent practices; (2) interdisciplinarity, with the METADAC Access Committee comprising individuals with social, biomedical, ethical, legal and clinical expertise, and individuals with personal experience participating in cohort studies; and (3) patient-centred decision-making, which means respecting study participants’ expectations, involving them in decision-making roles and communicating in a format that is clear and accessible.³²

12.4 Enabling Conditions and Institutional Designs

12.4.1 Enabling Conditions: Power/Resource Imbalances and Representativeness

Fung and Wright propose that an enabling condition to facilitate participatory governance is ‘a rough equality of power, for the purposes of deliberative decision-making, between participants’.³³ Nonetheless, power and resource imbalances are a common problem in many cases of patient and public engagement. Patients and publics bring different forms of knowledge that could be seen as challenging traditional scientific knowledge production and the legitimacy of professional skills and knowledge. Such knowledge could be constructed positively by researchers, but it could also be constructed in ways that question its validity compared to professional/academic knowledge.³⁴ Furthermore, patients and publics may not always be capable of articulating their needs as researchable questions, which limits the uptake of their ideas in research prioritisation, or a perceived mismatch may lead to resistance from researchers to act upon priorities identified by patients and publics.³⁵

Articulating a vision for advancing patient and public engagement in a health research system is important, whether it is at an organisational or broader level.³⁶ We further propose recognition of patients and publics as having legitimate representation as stewards or governors, with a role in articulating vision, identifying research priorities, setting ethical standards, and evaluation. Moreover, we suggest that formal arrangements are required to enable patients and publics in their role as stewards and governors within institutional architecture. A range of innovative mechanisms have been explored and implemented. For instance, ArthritisPower, which is a patient-powered research network within PCORI, established a governance structure in which patients have representation and overlapping membership across the Executive Board, Patient Governor Group and Research Advisory Board. Clear communication of expectations, provision of well-prepared tools for engagement (e.g. work groups organised around particular tasks or topics, online platform for patient governors to connect) and regular assessments of patient governors’ viewpoints are found to be necessary to support and build patients’ capacity within a multi-stakeholder governance structure.³⁷

It should be recognised that members of the public vary in their capacity to participate, deliberate and influence decision-making. Those who are advantaged in terms of education, wealth or membership in dominant racial/ethnic groups often participate more frequently and effectively in deliberative decision-making.³⁸ Power and resource imbalances can result in the problem of co-optation whereby stronger stakeholders are able to generate support for their own agendas. The lack of representation of certain groups – i.e. youth, Indigenous, Black and ethnic minority groups – has been noted in many efforts of patient and public engagement in health research,³⁹ which reflects structural barriers and/or historical discrimination and mistrust due to past ethical violations. This raises challenges of how to promote and support inclusion and equity in decision-making. This also serves as a valuable counterpoint on power dynamics as discussed by Brassington, chapter 9.

There are also concerns that patients may risk becoming less able to represent broader patient perspectives as they become more trained and educated in research and more involved in the governance of research activities. For instance, Epstein documented the use of ‘credibility tactics’, such as the acquisition of the language of biomedical science by HIV/AIDS activists to gain acceptance in the scientific community, and Thompson and colleagues identified the emergence of professionalised lay experts who demonstrated considerable support for dominant scientific paradigms and privileged professional or certified forms of expertise among patients and caregiver participants in cancer research settings in England.⁴⁰ To guard against this, the governance structure of ArthritisPower maintains a mix of veteran and new members by limiting patient governors’ memberships to three years.⁴¹

12.4.2 Institutional Designs: Relationships, Trust and Leadership Support

Fung and Wright outline three institutional design elements that are necessary for participatory governance: (1) devolution of decision-making power to local units that are charged and held accountable with implementing solutions; (2) centralised supervision and coordination to connect the local units, coordinate and distribute resources, reinforce quality of local decision-making, and diffuse learning and innovation; and (3) transformation of formal governance procedures to institutionalise the ongoing participation of laypeople.⁴² At a national level, devolution of power implies that the state solicits local units, such as community organisations and local councils, to devise and implement solutions. Members of the public are engaged at a local level through these organisations as stakeholders who are affected by the targeted problems. Within a health research system, network or organisation, patients and publics may serve on advisory boards and committees as members within a multi-stakeholder governance structure.

In this section, we discuss factors that may facilitate or impede the participation of patients and publics in the governance structures of health research systems, networks or organisations. It is important to consider multilevel engagement strategies for matching participation opportunities to varying interests, capacities and goals of patients and publics.⁴³ These strategies may range from patients and publics having one-time input into a targeted issue, to serving in leadership roles as members of a research team or governing body. Involving patients and publics in governance structures in an ongoing manner requires relationship building over much longer periods of time.

Clarity of roles and purposes of patient and public engagement is needed for relationship building, as well as for developing and maintaining trust. Participatory forms of governance are more feasible when stakeholders have opportunities to identify mutual gains in collaboration. However, pre-existing relationships can discourage stakeholders from seeing the value of collaboration. In health research that spans multiple sites, approaches and willingness to engage patients and publics may differ considerably across the participating sites.⁴⁴ Establishing new relationships with patients as partners may be considered too risky and jeopardising to current relationships by some sites.

Additionally, engagement activities that focus on ‘patients’, ‘citizens’ or ‘members of a community’, may each carry different sets of assumptions. Patients often have a personal connection to the health issue in question, whereas community members are selected to represent a collective experience and perspective. In national biomedical research initiatives, engagement as ‘citizens’ may lead to the exclusion of certain groups, such as advocacy groups and charities, from governing committees to avoid ‘special interests’.⁴⁵ While people may be able to navigate and draw on different aspects of their lives to inform research and policy, further exploration is needed to understand the common and distinctive aspects between different types of roles that people occupy.⁴⁶ In any case, clarity regarding roles and responsibilities, and transparency in the aims of engagement are necessary for relationship and trust building.

Fung and Wright assert that centralised supervision and coordination is needed to stabilise and deepen the practice of participatory governance among local units.⁴⁷ At a national level, centralised coordination is a component of leadership capacity to ensure accountability, distribute resources, and facilitate communication and information sharing across local units. According to Ansell and Gash, facilitative leadership is important for bringing together stakeholders, promoting the representation of disadvantaged groups, and facilitating dialogue and trust-building in the collaborative processes.⁴⁸ Trust-building requires commitment and mutual recognition of interdependence, shared understanding of the problem in question and common values, and face-to-face dialogue. Senior leadership and supportive policy and infrastructure are recognised as building blocks for embedding patient and public engagement in a health research system.⁴⁹

12.5 Conclusion

In this chapter, we have discussed the potentials and challenges of involving patients and publics as stewards or governors of health research, whether within a broad health system, a research network, or a specific organisation. We have also outlined some of the conditions and institutional design elements that may impede or facilitate the engagement of patients and publics in governance structures, focusing on issues of power/resource imbalances, representativeness, relationships, trust and leadership support. Some conditions and institutional design elements are necessary for the implementation of participatory governance, but our discussion is not intended to be comprehensive or prescriptive. In particular, we are not proposing a specific governance structure or body as an ideal. Governance structures can vary in their purposes and constituencies. With rapid scientific advances and potential for unanticipated ethical and social issues, a multi-stakeholder governance structure needs to contain an element of reflexivity and adaptivity to evolve in ways that are respectful of diverse needs and interests while responding to changes. Moreover, the literature on patient and public engagement has documented the need for rigorous evaluation of the impact of engagement on healthcare and health research, especially given the problems of inconsistent terminology and lack of validated frameworks and tools to evaluate patient and public engagement.⁵⁰ Stronger evidence of the impact and outcomes, both intended and unintended, of patient and public engagement may help normalise the role of patients and publics as partners in health research regulation.

13 Risk-Benefit Analysis

Carl H. Coleman

13.1 Introduction

This chapter explores the concept of risk-benefit analysis in health research regulation, as well as ethical and practical questions raised by identifying, quantifying, and weighing risks and benefits. It argues that the pursuit of objectivity in risk-benefit analysis is ultimately futile, as the very concepts of risk and benefit depend on attitudes and preferences about which reasonable people disagree. Building on the work of previous authors, the discussion draws on contemporary examples to show how entities reviewing proposed research can improve the process of risk-benefit assessment by incorporating diverse perspectives into their decision-making and engaging in a systematic analytical approach.

13.2 Identifying Risks

The term ‘risk’ refers to the possibility of experiencing a harm. The concept incorporates two different dimensions: (1) the magnitude or severity of the potential harm; and (2) the likelihood that this harm will occur. The significance of a risk depends on the interaction of these two considerations. Thus, a low chance of a serious harm, such as death, would be considered significant, as would a high chance of a lesser harm, such as temporary pain.

In the context of research, the assessment of risk focuses on the additional risks participants will experience as a result of participating in a study, which will often be less than the total level of risks to which participants are exposed. For example, a study might involve the administration of various standard-of-care procedures, such as biopsies or CT scans. If the participants would have received these same procedures even if they were not participating in the study, the risks of those interventions would not be taken into account in the risk-benefit analysis. As a result, it is possible that a study comparing two interventions that are routinely used in clinical practice could be considered low risk, even if the interventions themselves are associated with a significant potential for harm. This is the case with a significant proportion of research conducted in ‘learning health systems’, which seek to integrate research into the delivery of healthcare. Because many of the research activities in such systems involve the evaluation of interventions patients would be undergoing anyway, the risks of the research are often minimal, even when the risks of the interventions themselves may be high.¹

The risks associated with health-related research are not limited to potential physical injuries. For example, in some studies, participants may be asked to engage in discussions of emotionally sensitive topics, such as a history of previous trauma. Such discussions entail a risk of psychological distress. In other studies, a primary risk is the potential for unauthorised disclosure of sensitive personal information, such as information about criminal activity, or stigmatised conditions such as HIV, or mental disorders. If such disclosures occur, participants could suffer adverse social, legal, or economic consequences.

Research-related risks can extend beyond the individuals participating in a study. For example, studies of novel interventions for preventing or treating infectious diseases could affect the likelihood that participants will transmit the disease to third parties.² Similarly, studies in which psychiatric patients are taken off their medications could increase the risk that participants will engage in violent behaviour.³ Third-party risks are an inherent feature of research on genetic characteristics, given that information about individuals’ genomes necessarily has implications for their blood relatives.⁴ Thus, if a genetic study results in the discovery that a participant is genetically predisposed to a serious disease, other persons who did not consent to participate in the study might be confronted with distressing, and potentially stigmatising, information that they never wanted to know.

In some cases, third-party risks extend beyond individuals to broader social groups. As the Council for International Organizations of Medical Sciences (CIOMS) has recognised, research on particular racial or ethnic groups ‘could indicate – rightly or wrongly – that a group has a higher than average prevalence of alcoholism, mental illness or sexually transmitted disease, or that it is particularly susceptible to certain genetic disorders’,⁵ thereby exposing the group to potential stigma or discrimination. One example was a study in which researchers took blood samples from members of the Havasupai tribe in an effort to identify a genetic link to type 2 diabetes. After the study was completed, the researchers used the blood samples for a variety of unrelated studies without the tribe members’ informed consent, including research related to schizophrenia, inbreeding and migration patterns. Tribe members claimed that the schizophrenia and inbreeding studies were stigmatising, and that they never would have agreed to participate in the migration research because it conflicted with the tribe’s origin story, which maintained that the tribe had originated in the Grand Canyon. The researcher institution reached a settlement with the tribe that included monetary compensation and a formal apology.⁶

Despite the prevalence of third-party risks in research, most ethics codes and regulations do not mention risks to anyone other than research participants. This omission is striking given that some of these same sources explicitly state that benefits to non-participants should be factored into the risk-benefit analysis. A notable exception is the EU Clinical Trials Regulation, which states that the anticipated benefits of the study must be justified by ‘the foreseeable risks and inconveniences’,⁷ without specifying that those risks and inconveniences must be experienced by the participants themselves.

In addition to omitting any reference to third-party risks, the US Federal Regulations on Research With Human Participants state that entities reviewing proposed research ‘should not consider possible long-range effects of applying knowledge gained in the research (e.g. the possible effects of the research on public policy) as among those research risks that fall within the purview of its responsibility’.⁸ This provision is intended ‘to prevent scientifically valuable research from being stifled because of how sensitive or controversial findings might be used at a social level’.⁹

13.3 Identifying Benefits

The primary potential benefit of research is the production of generalisable knowledge – i.e. knowledge that has relevance beyond the specific individuals participating in the study. For example, in a clinical trial of an investigational drug, data sufficient to establish the drug’s safety and efficacy would be a benefit of research. Data showing that an intervention is not safe or effective – or that it is inferior to the existing standard of care – would also count as a benefit of research, as such knowledge can protect future patients from potentially harmful and/or ineffective treatments they might otherwise undergo.

Whether a study has the potential to produce generalisable knowledge depends in part on how it is designed. The randomised controlled clinical trial (RCT) is often described as the ‘gold standard’ of research, as it includes methodological features designed to eliminate bias and control for potential confounding variables.¹⁰ However, in some types of research, conducting an RCT may not be a realistic option. For example, if researchers want to understand the impact of different lifestyle factors on health, it might not be feasible to randomly assign participants to engage in different behaviours, particularly over a long period of time.¹¹ In addition, ethical considerations may sometimes preclude the use of RCTs. For example, researchers investigating the impact of smoking on health could not ethically conduct a study in which non-smokers are asked to take up smoking.¹² In these situations, alternative study designs may be used, such as cohort or case-control studies. These alternative designs can provide valuable scientific information, but the results may be prone to various biases, a factor that should be considered in assessing the potential benefits of the research.¹³

A recent example of ethical challenges to RCTs arose during the Ebola outbreak of 2013–2016, when the international relief organisation Médicins Sans Frontières refused to participate in any RCTs of experimental Ebola treatments. The group argued that it would be unethical to withhold the experimental interventions from persons in a control group when ‘conventional care offers little benefit and mortality is extremely high’.¹⁴ The difficulty with this argument was that, in the context of a rapidly evolving epidemic, the results of studies conducted without concurrent control groups would be difficult to interpret, meaning that an ineffective or even harmful intervention could erroneously be deemed effective. Some deviations from the ‘methodologically ideal approach’, such as the use of adaptive trial designs, could have been justified by the need ‘to accommodate the expectations of participants and to promote community trust’.¹⁵ However, any alternative methodologies would need to offer a reasonable likelihood of producing scientifically valid information, or else it would not have been ethical to expose participants to any risk at all.

The potential benefit of scientific knowledge also depends on the size of a study, as studies with very small sample sizes may lack sufficient statistical power to produce reliable information. Some commentators maintain that underpowered studies lack any potential benefit, making them inherently unethical.¹⁶ Others point out that small studies might be unavoidable in certain situations, such as research on rare diseases, and that their results can still be useful, particularly when they are aggregated using Bayesian techniques.¹⁷

Often, choices about study design can require trade-offs between internal and external validity. While an RCT with tightly controlled inclusion and exclusion requirements is the most reliable way to establish whether an experimental intervention is causally linked to an observable result – thereby producing a high level of internal validity – if the study population does not reflect the diversity of patients in the real world, the results might have little relevance to clinical practice – thereby producing a low level of external validity.¹⁸ In assessing the potential benefits of a study, decision-makers should take both of these considerations into account.

In addition to the potential benefit of generalisable knowledge, some research also offers potential benefits to the individuals participating in the study. Benefits to study participants can be divided into ‘direct’ and ‘indirect’ (or ‘collateral’) benefits.¹⁹ Direct benefits refer to those that result directly from the interventions being studied, such as an improvement in symptoms that results from taking an investigational drug. In some studies, there is no realistic possibility that participants will directly benefit from the study interventions; this would be the case in a Phase I drug study involving healthy volunteers, where the purpose is simply to identify the highest dose humans can tolerate without serious side effects. Indirect benefits include those that result from ancillary features of the study, such as access to free health screenings, as well as the psychological benefits that some participants receive from engaging in altruistic activities. Study participants may also consider any payments or other remuneration they receive in exchange for their participation as a type of research-related benefit.

Most commentators take the position that only potential direct benefits to participants and potential contributions to generalisable knowledge should be factored into the risk-benefit analysis. The concern is that, otherwise, ‘simply increasing payment or adding more unrelated services could make the benefits outweigh even the riskiest research’.²⁰ Other commentators reject this position on the ground that it is not consistent with the ethical imperative to respect participants’ autonomy, and that it could preclude studies that would advance the interests of participants, investigators, and society.²¹ The US Food and Drug Administration has stated that payments to participants should not be considered in the context of risk-benefit assessment,²² but it has not taken a position on consideration of other indirect benefits, such as access to free health screenings.

13.4 Quantifying Risks and Benefits

Once the risks and benefits of a proposed study have been identified, the next step is to quantify them. Doing this is complicated by the fact that the significance of a particular risk or benefit is highly subjective. For example, a common risk in health-related research is the potential for unauthorised disclosure of participants’ medical records. This risk could be very troubling to individuals who place a high degree of value on personal privacy, but for persons who share intimate information freely, the risk of unauthorised disclosure might be a minor concern. In fact, in some studies, the same experience might be perceived by some participants as a harm and by others as a benefit. For example, in a study in which participants are asked to discuss prior traumatic experiences, some participants might experience psychological distress, while others might welcome the opportunity to process past experiences with a sympathetic listener.²³

In addition to differing attitudes about the potential outcomes of research, individuals differ in their perceptions about risk-taking itself. Many people are risk averse, meaning that they would prefer to forego a higher potential benefit if it enables them to reduce the potential for harm. Others are risk neutral, or even risk preferring. Similarly, individuals exhibit different levels of willingness to trade harmful outcomes for good ones.²⁴ For example, some people are willing to tolerate medical treatments with significant side effects, such as chemotherapy, because they place greater value on the potential therapeutic benefits. Others place greater weight on avoiding pain or discomfort and would be disinclined to accept high-risk interventions even when the potential benefits are substantial.

Another challenge in attempting to quantify risks and benefits is that the way that risks and benefits are perceived can be influenced by a variety of cognitive biases. For example, one study asked subjects to imagine that they had lung cancer and had to decide between surgery and radiation. One group was told that 68 per cent of surgical patients survived after one year, while a second group was told that 32 per cent of surgical patients died after one year. Even though the information being conveyed was identical, framing the information in terms of a risk of death increased the number of subjects who chose radiation from 18 per cent to 44 per cent.²⁵ Another common cognitive bias is the ‘availability heuristic’, which leads people to attach greater weight to information that is readily called to mind.²⁶ For example, if a well-known celebrity recently died after being implanted with a pacemaker, the risk of pacemaker-related deaths may be perceived as greater than it actually is.

Individuals’ perceptions of risks and benefits can also be influenced by their level of social trust, which has been defined as ‘the willingness to rely on those who have the responsibility for making decisions and taking actions related to the management of technology, the environment, medicine, or other realms of public health and safety’.²⁷ In particular, research suggests that, when individuals are considering the risks and benefits of new technologies, their level of social trust has ‘a positive influence on perceived benefits and a negative influence on perceived risks’.²⁸ This is not surprising: those who trust that decision-makers will act in their best interests are less likely to be fearful of changes, while those who lack such trust are more likely to be worried about the potential for harm (see Aitken and Cunningham-Burley, Chapter 11, in this volume).

Compounding these subjective variables is the fact that risk-benefit analysis typically takes place against a backdrop of scientific uncertainty. This is true for all risk-benefit assessments, but it is especially pronounced in research, as the very reason research is conducted is to fill an evidentiary gap. While evaluators can sometimes rely on prior research, including animal studies, to identify the potential harms and benefits of proposed studies, most health-related research takes place in highly controlled environments, over short periods of time. As a result, prior research results are unlikely to provide much information about rare safety risks, long-term dangers or harms and benefits that are limited to discrete population subgroups.

13.5 Weighing Risks and Benefits

Those responsible for reviewing proposed research must ultimately weigh the risks and benefits to determine whether the relationship between them is acceptable. This process is complicated by the fact that risks and benefits often cannot be measured on a uniform scale. First, ‘risks and benefits for subjects may affect different domains of health status’,²⁹ as when a risk of physical injury is incurred in an effort to achieve a potential psychological benefit. Second, ‘risks and benefits may affect different people’;³⁰ risks are typically borne by the participants in the research, but most of the benefits will be experienced by patients in the future.

Several approaches have been suggested for systematising the process of risk-benefit analysis in research. The first, and most influential, approach is known as ‘component analysis’. This approach calls on decision-makers to independently assess the risks and potential benefits of each intervention or procedure to be used in a study, distinguishing those that have the potential to provide direct benefits to participants (‘therapeutic’) from those that are administered solely for the purpose of developing generalisable knowledge (‘non-therapeutic’). For therapeutic interventions, there must be genuine uncertainty regarding the relative therapeutic benefits of the intervention as compared to those of the standard of care for treating the participants’ condition or disorder (a standard known as ‘clinical equipoise’³¹). For non-therapeutic interventions, the risks must be minimised to the extent consistent with sound scientific design, and the remaining risks must be reasonable in relation to the knowledge that is expected to result. In addition, when a study involves a vulnerable population, such as children or adults who lack decision-making capacity, the risks posed by nontherapeutic procedures may not exceed a ‘minor increase above minimal risk’.³²

Component analysis has been influential, but it is not universally supported. Some critics maintain that the distinction between therapeutic and non-therapeutic procedures is inherently ambiguous, as ‘all interventions offer at least some very low chance of clinical benefit’.³³ Others argue that the approach’s reliance on clinical equipoise rests on the mistaken assumption that researchers have a duty to promote each participant’s medical best interests, which conflates the ethics of research with those of clinical care.³⁴

One alternative to component analysis is known as the ‘net risk test’, which is based on the principle that the fundamental ethical requirement of research is ‘to protect research participants from being exposed to excessive risks of harm for the benefit of others’.³⁵ The approach has four elements. First, for each procedure involved in a study, the risks to participants should be minimised and the potential clinical benefits to participants enhanced, to the extent doing so is consistent with the study’s scientific design. Second, instead of clinical equipoise, the approach requires that, ‘when compared to the available alternatives, a research procedure must not present an excessive increase in risk, or an excessive decrease in potential benefit, for the participant’.³⁶ Third, to the extent particular procedures involve greater risks than benefits, those net risks ‘must be justified by the expected knowledge gained from using that procedure in the study’.³⁷ Finally, the cumulative net risks of all of the procedures in a study must not be excessive.³⁸

Both component analysis and the net risk test can add structure to the process of risk-benefit analysis by focusing attention on the risks and potential benefits of each intervention in a study. The advantage of this approach is that it reduces the likelihood that potential direct benefits from one intervention will be used as a justification for exposing participants to risks from unrelated interventions that offer no direct benefits. However, neither approach eliminates the need for subjective determinations. Under component analysis, the principle of clinical equipoise offers a benchmark for judging the risks and potential benefits of therapeutic procedures, but for non-therapeutic procedures, the only guidance offered is that the risks must be ‘reasonable’ in relation to the knowledge expected to result. The net benefit test dispenses with clinical equipoise entirely, instead relying on a general principle of avoiding ‘excessive risk’. Whether a particular mix of risks and potential benefits is ‘reasonable’ or ‘excessive’ is ultimately left to the judgment of those charged with reviewing the study.

Most regulations and ethics codes provide little guidance on the process of weighing the risks and potential benefits of research. The primary exception is the CIOMS guidelines, which adopts what it describes as a ‘middle ground’ between component analysis and the net risk test. In most respects, the CIOMS approach reflects component analysis, including its reliance on clinical equipoise as a standard for evaluating interventions or procedures that have the potential to provide direct benefits to participants. However, the guidelines also call for a judgment that ‘the aggregate risks of all research interventions or procedures … must be considered appropriate in light of the potential individual benefits to participants and the scientific social value of the research’,³⁹ a requirement that mirrors the final step of the net risk test.

Neither component analysis nor the net risk test explicitly sets an upper limit on permissible risk, at least in studies involving competent adults. However, one of the developers of component analysis has stated that ‘the notion of excessive net risks, and the underlying ethical principle of non-exploitation, clearly impose a cap on the risks that individuals are allowed to assume for the benefit of others’.⁴⁰ The notion of an upper limit on risk also appears in several ethical guidelines. For example, the CIOMS guidelines state that ‘some risks cannot be justified, even when the research has great social and scientific value and adults who are capable of giving informed consent would give their voluntary, informed consent to participate in the study’.⁴¹ Similarly, the European Commission has suggested that certain ‘threats to human dignity and shared values’ should never be traded against the potential scientific benefits of research, including ‘commonly shared values like privacy or free movement … certain perceptions of the integrity of a person (e.g. cloning, technological modifications) … [and] widely shared view[s] of our place in the world (e.g. inhumane treatment of animals or threat to biodiversity)’.⁴²

In light of the inherent ambiguities involved in weighing the risks and benefits of research, the results of risk-benefit assessments can be heavily influenced by the type of decision-making process used. The next section looks at these procedural issues more closely.

13.6 Procedural Issues in Risk-Benefit Analysis

In most health-related research, the process of risk-benefit assessment is undertaken by interdisciplinary bodies known as research ethics committees (RECs), research ethics boards (REBs), or institutional review boards (IRBs). These committees make judgments based on predictions about the preferences and attitudes of typical research participants, which do not necessarily reflect how the actual participants would react to particular risk-benefit trade-offs.⁴³ In addition, because few committees rely on formal methods of risk-benefit analysis, decisions are likely to be influenced by individual members’ personal attitudes and cognitive biases.⁴⁴ For this reason, it is not surprising that different committees’ assessments of the risks and potential benefits of identical situations exhibit widespread variation.⁴⁵

Some commentators have proposed techniques to promote greater consistency in risk-benefit assessments. For example, it has been suggested that committees issue written assessments that could be entered into searchable databases.⁴⁶ Others have called on committees to engage in a formal process of ‘evidence-based research ethics review’, in which judgments about risks and potential benefits would be informed by a systematic retrieval and critical appraisal of the best available evidence.⁴⁷

Outside of research ethics, a variety of techniques have been developed to systematise the process of risk-benefit analysis. For example, several quantitative approaches to risk-benefit assessment exist, such as the Quality-Adjusted Time Without Symptoms and Toxicity (Q-TWIST) test, which ‘compares therapies in terms of achieved survival and quality-of-life outcomes’,⁴⁸ or the ‘standard gamble’, which assigns utility values to health outcomes based on individuals’ stated choice between hypothetical health risks.⁴⁹ Committees reviewing proposed studies can draw on these quantitative analyses when relevant ones exist.

In some cases, formal consultation with the community from which participants will be drawn can be an important component of assessing risks and benefits. For example, in the study of Havasupai tribe members discussed above, prior consultation with the community could have alerted researchers to the fact that research on migration patterns was threatening to the tribe’s cultural beliefs. In cancer research, consultation with patient advocacy groups may help identify concerns about potential adverse effects that might not have been sufficiently considered by the researchers.⁵⁰ Further lessons might be learned from the the analysis by Chuong and O'Doherty, Chapter 12, this volume.

13.7 Conclusion

Risk-benefit analysis is a critical part of the process of evaluating the ethical acceptability of health-related research. The primary challenge in risk-benefit assessment arises from the fact that perceptions about risks and potential benefits are inherently subjective. Those charged with assessing the ethical acceptability of research should make efforts to incorporate as many different perspectives into the process as possible, to ensure that their decisions do not simply reflect their own idiosyncratic views.

14 The Regulatory Role of Patents in Innovative Health Research and Its Translation from the Laboratory to the Clinic

Dianne Nicol and Jane Nielsen

14.1 Introduction

Regulators must ensure that innovative health research is safe and undertaken in accordance with laws, ethical norms and social values, and that it is translated into clinical outcomes that are safe, effective and ethically appropriate. But they must also ensure that innovative health research and translation (IHRT) is directed towards the most important health needs of society. Through the patent system, regulators provide an incentive-based architecture for this to occur by granting a temporary zone of exclusivity around patented products and processes. Patents thus have the effect of devolving control over IHRT pathways to patentees and to those to whom patentees choose to license their patent rights.

The sage words of Stephen Hilgartner set the backdrop for this chapter: ‘Patents do not just allocate economic benefits; they also allocate leverage in negotiations that shape the technological and social orders that govern our lives’.¹ Patents have been granted for many – if not all – of the major recent innovations in health research, from the earliest breakthroughs like recombinant DNA technology, the polymerase chain reaction, the Harvard Oncomouse and the BRCA gene sequences, through to a whole variety of viruses, monoclonal antibodies, receptors and vectors, thousands of DNA sequences, embryonic stem cell technology, intron sequence analysis, genome editing technologies and many more.² These innovations have laid the foundations for whole new health research pathways, from basic research, through applied research, to diagnostic and therapeutic end points.³ Broad patent rights over these fundamental innovations give patentees the freedom to choose how these research pathways will be progressed. Essentially then, the patent grant puts patentees in a position to assert significant private regulatory control over IHRT.

The first part of this chapter outlines this regulatory role of patents in IHRT. The chapter then considers the ways in which patentees choose to use their patent rights in IHRT, and the scope for government intervention. The chapter then explores recent actions by patentees that indicate a willingness to moderate the use of their patent rights by engaging in self-regulation and other forms of collaborative regulation. Finally, the chapter concludes with a call for greater government oversight of patent use in IHRT. Although self-regulation has merit in the absence of clear governmental direction, it is argued that private organisations should not have absolute discretion in deciding how to employ their patents in areas such as health, but that they must be held to account in exercising their state-sanctioned monopoly rights.

14.2 Patents as a Form of Private Regulation

In many markets, the regulation of market entry, prices, product availability and development is left to the market to varying degrees, there being at least some general consensus that competitive decision-making is a hallmark of market efficiency.⁴ At the same time, granting patent rights removes an element of competition from a market in order to induce innovation and disclosure.⁵ While it is unclear how much innovation is optimal, it has been suggested that there is unlikely to ever be too much from an economic welfare perspective.⁶

Although primary innovators are arguably best placed to organise and control follow-on innovation,⁷ vesting decision-making power in a single private entity has the potential to scuttle efficiency in much the same way as absolute government control. Nonetheless, conferring this power on individual entities through the grant of patents – and accompanying Intellectual property (IP) rights – is generally justified on efficiency grounds.⁸ However, non-efficiency goals such as distributive fairness may also be important drivers of private regulatory arrangements and may be incorporated either consciously or unconsciously in regulatory schemes.⁹

Granting a patent gives a property right in an invention. As Mark Lemley observes, IP constitutes both a form of government regulation and a property right around which parties can contract,¹⁰ and its confused identity partly explains why policy makers have grappled with exactly how to manage the delicate innovation balance. Studies have provided mixed evidence as to the necessity to grant IP rights: in some technology areas, patents are viewed as necessary in order to recoup research and development investment, but this is by no means universal.¹¹

The value of patents in IHRT has not been unequivocally established, although there is some evidence to suggest they are crucial for signalling purposes.¹² Patent law can be said to form a ‘corrective’ function in the health context, particularly in relation to pharmaceuticals and biotechnology, where the development of clinical products is subject to substantial regulation.¹³ Without patents, it is argued that researchers would not commit the considerable investment required to conduct research with the ultimate aim of a clinical outcome.

14.3 Use of Patent Rights in Innovative Health Research and Translation

Patentees can limit who enters a field by choosing who, if anyone, they will authorise to use their patents. This can create problems for broad breakthrough technologies, where insistence on exclusivity gives patentees and their licensees control over whole research pathways, allowing them to dictate how those pathways develop. Patentees and their licensees could choose to block others completely from using the technology, or restrict access, or charge excessive prices for use. Conversely, they could allow their patented technology to be used widely for minimal costs. The tragedy of the anticommons posited by Michael Heller and Rebecca Eisenberg, adds further complexity, speculating that a proliferation of patents in particular areas of technology exacerbates the problem because no one party has an effective privilege of use.¹⁴ Rather, agreement with multiple patentees would be required in order to utilise a particular resource.

Fortunately, empirical studies have revealed little evidence of blocking or anticommons effects in IHRT,¹⁵ suggesting that, on the whole, working solutions employed by researchers have allowed them to work around ‘problematic’ patents so that research and development may progress. ‘Working solutions’ mean strategies such as entering into licence agreements or other collaborative arrangements; inventing around problematic patents; relying on research exemptions; or challenging the validity of patents.¹⁶ These working solutions can be viewed as facets of the regulatory scheme that encompasses the grant of patent rights. However, solutions that involve entering into a licence agreement or other collaborative arrangement also involve a degree of conformity on the part of a patentee. It may be fruitless to approach a patentee unless they are willing to negotiate, which takes time and effort on their part, as well as on the part of the licensee. Unless these processes can be streamlined, the incentive to license is low.

14.4 Scope for Government Intervention

Arguably, the fruits of all health-related research should be distributed openly, because of its vital social function of improving healthcare. However, this is hardly a realistic option for aspects such as drug development, where the enormous cost of satisfying regulatory requirements for marketing approval must be recoverable. For other aspects of IHRT, however, the case for more open access is compelling, particularly since it generally originates in public research laboratories, funded by governments from the public purse.¹⁷ Yet the ensuing patents may ultimately be controlled by private parties, whether spin-offs or more established firms. This phenomenon has been referred to by Jorge Contreras and Jacob Sherkow as ‘surrogate licensing’.¹⁸

Given the public contribution made to IHRT, the argument for open access, at least for research purposes, is appealing. Public funders are within their rights to insist on some form of open dissemination in such circumstances.¹⁹ But what are the options when patentees or their licensees insist on exclusivity, even for the most fundamental research tools? If governments see patents as providing a broader social function beyond giving monopoly rights to patentees – albeit temporary in nature – they must ensure that, along with incentives to innovate, the patent system provides appropriate incentives to disseminate innovative outputs, or other regulatory mechanisms to compel the provision of access where needed.²⁰ Patents provide patentees with significant freedom to decide who can enter a particular field of research, and what they can do. Some jurisdictions do have legislative provisions allowing government or private providers to step in should patentees fail to work the invention.²¹ Most countries exempt from infringement the steps needed for regulatory approval of generic pharmaceuticals and other chemicals.²² Some also exempt use of the patent for experimental purposes, although the scope of protected experimental use remains unclear.²³ However, the reality is that the role of governments in regulating patent use is limited.

14.5 Emergent Self-Regulatory Models for Use of Patent Rights in Innovative Health Research and Translation

Recognising these limitations on government control of patent use, some promising developments are emerging in IHRT that indicate that patentees and their licensees are willing to consider a range of self-regulatory models in ensuring optimal patent utilisation. Some of the more prominent examples are discussed below.

14.5.1 Non-exclusive Research Tool Licensing

Because foundational research tools are just that – foundational to whole new areas of research – best practice dictates they should be licensed non-exclusively. US funding agencies and universities agree; for example, the US National Institutes of Health released guidance to this effect in 1999 and 2005.²⁴ In 2007, the Association of University Technology Managers, recognising that ‘universities share certain core values that can and should be maintained to the fullest extent possible in all technology transfer agreements’, provided nine key points to consider in licensing university patents. Point 5 recommends ‘a blend of field-exclusive and non-exclusive licenses’.²⁵

Yet non-exclusive licensing is not cost-free. The problem that it presents to users is that it imposes a fee in return for not being sued for infringement, with little or no additional benefit for the user.²⁶ Inclusion of reach through rights to future uses adds to the burden on follow-on researchers.²⁷ If governments were really concerned about the toll of research tool patent claims on IHRT they could choose to exclude them, or to require them to be exchanged through some form of statutory licensing scheme, with minimal or no licensing fees and no other restrictive terms. For now, however, governments seem content to leave such decisions to patentees.

We are witnessing some interesting developments in this area, illustrating that government intervention may not yet be necessary. Companies like Addgene and the Biobricks Foundation have been established as intermediaries to facilitate no-cost, non-exclusive patent licensing and sharing of research materials for genome editing and synthetic biology research, respectively.²⁸ There are also other examples of these types of intermediary arrangements, or ‘clearinghouses’ as they are sometimes called, in IHRT. Such arrangements appear to provide a valuable social function provided that fees are not excessive and that technology that is of real value to IHRT is included, so that the clearinghouse does not become a ‘market for lemons’.²⁹

14.5.2 Mixed Licensing Models

Realistically, a more nuanced approach over the simple choice of exclusive or non-exclusive licensing is needed, involving a mix of licensing strategies for a single patented technology. Licensing of the clustered regularly interspersed palindromic repeats (CRISPR) patents illustrates this point. CRISPR, as explained in Chapter 34, is a genome editing technology that has captivated the research world because of its ease of use and enhanced safety, owing to reduced incidence of off-target effects.³⁰

Already, we are witnessing the adoption of nuanced approaches for licensing CRISPR patents. For example, the Broad Institute, one of the giants of CRISPR technology, non-exclusively licences CRISPR constructs freely for public sector research through Addgene, and charges a fee for use in more commercially-oriented research. Broad exclusively licences to its own spin-off company, Editas, for therapeutic product development. Broad describes this as an ‘inclusive innovation model’.³¹ However, this model has been criticised by Oliver Feeney and colleagues on the basis that the decision whether to allow other uses for therapeutic purposes is left to Editas.³² They see this as a ‘significant moral hazard’, because of the potential restrictions it imposes on therapeutic development. While Feeney and colleagues propose government-imposed time limitations on exclusivity as a means of addressing such hazards, it is doubtful, given past history, that governments would be persuaded to incorporate this level of post-grant regulatory intervention within the patent system.

Knut Egelie and colleagues, equally concerned about CRISPR patent licensing, argue that public research organisations should commit more fully to a self-regulatory model that balances social responsibilities with commercial activity.³³ Their ‘transparent licensing model’ would minimise fees and other restrictions for uses of patented subject matter as research tools, and narrow field-of-use exclusive licences for commercial development. They suggest government intervention as an alternative to this self-regulatory model, referring to some of the recently emerging contractual funding strategies in Europe. However, they themselves criticise both options, the former for lacking public control and the latter for over-regulation and unnecessary bureaucracy. More cooperative and collaborative strategies, involving both public sector and private sector organisations, might provide alternative models.

14.5.3 Collaborative Licensing

A greater commitment to social responsibility might be achieved by patentees and their licensees through entry into collaborative IP arrangements.³⁴ Patent pools have been used in some high technology areas – particularly information technology – to overcome patent thickets and cluttered patent landscapes.³⁵ In IHRT, however, complex arrangements such as patent pools have gained limited traction,³⁶ primarily because of the lack of need to date. Simpler strategies such as non-exclusive licensing and clearinghouses appear to be adequate at the present time, and predicted anticommons effects have not yet emerged.³⁷

Where patentees are reluctant to engage in collaborative strategies, there is some scope to mandate engagement. Patent pools, for example, have in some instances – especially in the USA – been established by government regulators in order to ease innovative burdens and address competition law concerns.³⁸ Mandatory arrangements are rarely optimistically embraced, and prospects for the sustainability of collaborative arrangements is probably significantly greater where they are voluntary. Patent pools are complex structures and involve many legal considerations. Although there has been some success in establishing patent pooling-type arrangements in public health emergencies like HIV/AIDS and other epidemics,³⁹ it is difficult to see what would motivate patentees to come together to create such complex structures in IHRT at the present time, particularly given the rapid pace of technological development and change.

Patent aggregation is another increasingly popular strategy, referring to the process of collecting suites of IP required to conduct research and development within a particular field of use. The process of patent aggregation has brought with it some negative press, because of concerns that aggregators could be ‘patent trolls’, whose sole motivation is extracting licensing revenue.⁴⁰ However, not all aggregators have this trolling motivation, but rather license out entire bundles of patents on a non-exclusive basis. To this extent, their role in advancing the research agendas in IHRT can be seen as broadly facilitative.⁴¹

14.5.4 Ethical Licensing

Aside from the social good associated with self-regulatory models of patent use discussed above, there are other ethical and social considerations that could be addressed through more public-focused approaches to licensing. For example, even where public sector organisations exclusively license to private partners, whether spin-offs or established firms, it is common practice for the license terms to reserve rights for the organisation’s researchers to continue to conduct research using patented subject matter.⁴² Reservation of the right to engage in broader sharing of patented subject matter for non-commercial research purposes might also be included in such agreements, effectively circumventing the lack of a statutory or common law research exemption in some jurisdictions.

Patent pledges and non-assertion covenants can be used to serve essentially the same purpose.⁴³ The role of reservation of rights could also extend to humanitarian uses, which has been mooted specifically in the context of agricultural biotechnology. As Alan Bennett notes, these voluntary measures can serve the purpose of meeting the humanitarian and commercial needs of developing countries in the absence of national policies to this effect.⁴⁴ Such measures could be equally effective in the context of humanitarian uses of innovative health technologies, an area which likewise suffers from a lack of clear government policy direction.

There has been recent discussion on the efficacy of introducing ethical terms into patent licences for the new genome editing technologies, particularly CRISPR. The emergence of this technology triggered a range of ethical debates in relation to its applications in agriculture, the natural environment – for example, in pest eradication through a combination of CRISPR and gene drives – and humans – for example, in genetic enhancement, germline genome modification and gene editing research using human embryos.⁴⁵

The Broad Institute, through Editas, and other public research organisations and their licensees, are already using licences that exclude these types of ethically questionable uses, whether in human or non-human contexts. As Christi Guerrini and colleagues note, there are some obvious advantages with this approach, including that: licence terms are enforceable; they can be tailored; and they are negotiated, leading to better buy in.⁴⁶ Given that the regulation of genome editing varies widely across jurisdictions,⁴⁷ the introduction of ethical licensing terms also has the advantage of creating enforceable obligations across the jurisdictions where the patent has been granted and where the licence applies. Potentially, then, ethical licences could impose global standards on uses of CRISPR technology, which is otherwise considerably conjectural if relying on agreement between countries.

Despite the apparent attractiveness of ethical licensing, however, there is likely to be some unease with the notion of devolving decisions about what is or is not ethical to patentees.⁴⁸ In areas such as this, which are highly contentious, community consensus would usually be a precursor to government regulation. Is regulatory failure in this area significant enough to justify private action? Is this a step too far when it amounts to ceding regulation to private entities?

14.6 Conclusion

Patents play a key role in the progress of IHRT. By granting patents, governments devolve to patentees considerable decision-making power about who can enter particular fields of IHRT and what they can do. This chapter has shown that patentees can and do choose to exercise this power wisely, by engaging in open and collaborative models for patent use. However, not all choose do so, and governments currently have limited regulatory tools with which to compel such engagement.

Patentees can decide to work collaboratively with other interested parties, or not. They can decide whether to share broadly, or not. They can even decide what types of uses are ethical or unethical. This is a significant set of delegated powers. Regulators have at their disposal various policy levers that could provide them with broad discretion to specify criteria for patent eligibility, periods of exclusivity and access.⁴⁹ Regulatory control can be asserted by governments both pre-grant, influencing the ways in which patents are granted, and post-grant, on the ways in which patents are used. Governments can use these regulatory tools to impose limits on these delegated powers, but these are not being fully utilised at present.

The current situation is that non-enforceable guidelines have been issued in some jurisdictions to assist patentees in deciding how to exercise their powers, but not in others. Internationally, although the OECD has issued licensing guidelines,⁵⁰ for the most part there is no jurisdictional consensus on how best to set limits on the exercise of patent rights. This is not surprising in view of the diversity of technologies and actors involved and given jurisdictional discrepancies. More research is needed to assist governments in finding optimal ways to support, guide and regulate public research organisations and private companies in their use of the patent system in IHRT.

15 Benefit Sharing From Compensation to Collaboration

Kadri Simm

15.1 Introduction

Benefit sharing pertains to the distribution of benefits and burdens arising from research. More specifically, it concerns what, if anything, is owed to individuals, communities or even populations that participate in research (benefits to investors, to other populations or the social value of research more generally understood are not the focus of benefit sharing).

Traditionally, health research has been concerned with compensating those participants who have been more or less directly involved. The practice of benefit sharing, especially in agriculture, introduced a perspective that recognised the contributions of communities and populations in safeguarding biological resources.¹ The issue is further complicated in human genetics as genetic information is by nature shared, and thus implicates individuals and communities who might not have participated in research in the traditional sense. At the same time, contemporary global research activities have increasingly been associated with for-profit companies. Some of their practices – ‘helicopter research’, ethics dumping – have given credence to broader political and social worries that have now been harnessed to the concept of benefit sharing, which was initially used within more limited research settings.

Framing benefit-sharing debates are several central concepts – the duty to avoid exploitation, the rights and interests of all research stakeholders, the requirements of fairness and compensation, and the various principles of distributive justice. In many ways, benefit sharing as an ethics and governance framework attempts to deal with most of those concerns and anxieties. Thus, responses to the question, ‘why is benefit sharing a duty?’ vary. In practical terms, benefit sharing is a thoroughly context-sensitive topic. It matters which risks and harms are involved in research (if any), who the investigators and funders are (for-profit, local, NGOs etc), where research takes place (developed or low- or middle-income countries), who is involved (e.g. vulnerable groups), what local needs are, and whether research is successful.

In what follows, I will give a brief overview of the ethical arguments and historical dynamics behind benefit-sharing practices, then outline major governance frameworks and discuss the potential problems around applying this concept in health research. The overall aim of this chapter is to highlight the complexity of benefit sharing and argue that success hinges on the careful balancing of universal research ethics duties with the particularities of concrete research projects taking place in distinct locations. Benefit sharing is no panacea for solving the inequalities of access and opportunities associated with global health research. Yet it can be a profoundly empowering tool, especially as the framework is shifting from compensation to collaboration.

15.2 History and Rationale of Benefit Sharing

Looking back, the rationale behind access and benefit-sharing justifications has been dynamic. It was originally employed in the context of agriculture and non-human biological resources (plants, animals, microorganisms). The 1992 UN Convention on Biological Diversity (CBD) acknowledged national sovereignty in all genetic resources and requested ‘fair and equitable sharing of the benefits arising out of the utilization of genetic resources’.² As the majority of the world’s biological diversity is found in developing countries, benefit sharing was seen as a necessary instrument in guaranteeing these countries’ continuing interest in safeguarding this heritage and curbing biopiracy (when indigenous knowledge and resources are patented or otherwise exploited by third parties with no permission or compensation for the locals). The supplementary Nagoya Protocol on Access and Benefit-sharing (2010) is a legal framework that supports the implementation of the objectives of CBD.³

Since the 1990s, benefit sharing emerged as an important component of health research and made its appearance in various international documents (in the rest of the chapter, I will focus on benefit sharing in health research only, excluding research on non-human materials and populations). The Human Genome Organisation (HUGO) Ethics Committee Statement on benefit sharing formulates:

A benefit is a good that contributes to the well-being of an individual and/or a given community (e.g. by region, tribe, disease-group …). Benefits transcend avoidance of harm (non-maleficence) in so far as they promote the welfare of an individual and/or of a community. Thus, a benefit is not identical with profit in the monetary or economic sense. Determining a benefit depends on needs, values, priorities and cultural expectations.⁴

Benefits put forward by scientists, as well as the pharmaceutical industry, patients, investors and public health officials, span a wide array of potential valued ‘goods’, from improved health and science to financial gains and wider social benefits.⁵ A fixed definition of what would constitute a benefit would be quite useless, or worse, unfair (an informative list of possible benefits regarding non-human research is available from the annex of the Nagoya Protocol). Potential benefits and harms arising from clinical trials would be rather different from those associated with population biobanks, for example. Benefits can be related to healthcare, but they could also encompass other socially important goals, such as support for infrastructure, development of local research capacities and build-up of community resilience. The kind and scope of potential benefits has few limits, although the minimum threshold for satisfying the ‘reasonable availability’ should surpass the simple licensing of drugs or interventions with market prices.⁶

When is an appropriate time for benefit sharing? These issues deserve consideration from the very earliest phases of research design. It is necessary to find out the characteristics and needs of the potential research sites to ensure that the planned investigations, as well as potential benefits, respond to those needs. Equally, benefit sharing could involve long-term follow up of participants or training and employment of community members that continues for years after research has ended.

The HUGO statement on benefit sharing mapped the following justifications for the concept in human genetic research:

1. 1. Descriptive argument: There is an ‘emerging international consensus’⁷ that benefits should be shared with participants.

2. 2. Common heritage argument – we all share (in one sense) the same genome, so there is a shared interest in genetic heritage of humankind; thus, the Human Genome Project should benefit all humanity.

3. 3. Justice-based arguments – compensatory (compensation in return for contribution), procedural (procedural justice should be adhered to in benefit-sharing) and distributive (equitable allocation and access to resources and goods) justice as important aspects to consider.

4. 4. Solidarity argument on two levels: first, as a potential basis for benefit sharing among a group of research participants (communities, host populations); second, to foster health for wider communities and eventually the whole of humanity, thus benefits should not be limited strictly to those participating in research.⁸

Of these various justifications, the overall concern fuelling benefit-sharing debates has been justice, and the concept itself has been likened to a device in the toolbox of justice.⁹ Yet, justice is notoriously difficult to pin down given that the principles of justice vary – one can refer to equality as fundamental, or point at the importance of merit, and in healthcare contexts the principle of need has often served as central. Decisions about what justice requires (i.e. what principles are important in a particular context) can result in divergent benefit-sharing patterns and practices – how benefits are defined and by whom, as well as with whom the sharing is foreseen.¹⁰ Certain justifications necessarily exclude or include specific groups or communities. For example, the compensatory logic associated with the principles of merit and desert would benefit those directly involved but could leave out those who did not directly participate but are nevertheless part of the community. Focus on a shared human heritage of genetic resources tends to disregard the needs and deserts of particular communities where research is undertaken. This is why, for example, in the agricultural and plant genetics context, the early employment of the global heritage model was quickly replaced by the nationalisation and property model of genetic resources.¹¹ The patenting practices through which the ‘shared free resources’ were turned into private profits and property were eventually rejected and the nationalisation of biological resources took over as the dominant framework.

To conclude, benefit-sharing negotiations always entail choices between some publics over others and upholding of certain principles before others. The above considerations about what justice requires have historically played a role in benefit-sharing discussions and none of them may be discounted as irrational or irrelevant. So how have these justice-related concerns been framed, operationalised, and translated into regulation and governance?

15.3 Regulation and Governance Frameworks

Ethically sound and respectful research practices do not only benefit researchers, participants and science but also support public trust towards research in general.¹² All approaches to benefit sharing assume the baseline of the usual ethics requirements for research (thus benefit sharing does not substitute some or all ethics principles but is to be considered an additional one). In 1993, the Council for International Organisations of Medical Sciences (CIOMS) argued that ‘any product developed will be made reasonably available to inhabitants of the underdeveloped community in which the research was carried out’.¹³ In the latest updated Guidelines from 2016, exploitative research was defined as the kind of research that did not respond to the health needs of the community where it took place or who would later not be able to access or afford the resulting product.¹⁴

The prominence of benefit sharing as an ethics requirement in global health research is exemplified by the existence of many national¹⁵ and international documents, statements and opinions. Both national and international health research organisations, policy think tanks and research funders have thought it important to discuss and state their views on the matter. Most discuss benefit sharing in the context of research in developing countries: the European Group on Ethics in Science and New Technologies to the European Commission’s Opinion on Ethical Aspects of Clinical Research in Developing Countries (2003), the Nuffield Council on Bioethics’ The Ethics of Research Related to Healthcare in Developing Countries (first paper in 2002), the US National Bioethics Advisory Commission’s Ethical and Policy Issues in International Research (2001), and the Wellcome Trust’s Statement on Research Involving People Living in Developing Countries: Position Statement and Guidance Notes for Applicants.¹⁶ Even general health research frameworks have included references to benefit sharing in their more recent drafts – for example the WHO’s Good Clinical Practice, the World Medical Association’s Declaration of Helsinki (2013), and the UNESCO Universal Declaration on Bioethics and Human Rights (2005).¹⁷

All of the above documents constitute what may be called soft law (i.e. non-binding instruments), yet a number of them have been influential in regulating health research practices (especially the WHO, CIOMS and funders’ guidelines). When applied routinely, such ethics regulations could be considered customary international law,¹⁸ but there have also been calls to formulate dedicated legal instruments to provide stronger support for benefit-sharing negotiations.¹⁹ The latest attempt to ensure that benefit sharing constitutes an important normative aspect of research is the Global Code of Conduct for Research in Resource-Poor Settings (2018), which the European Commission endorsed as a reference document for its research funding programme Horizon 2020.²⁰

While declarations and guidelines can highlight important principles and values for research, their interpretation and implementation are less straightforward. Over time, the developments in health research practices and the pressures from various stakeholders have resulted in a repeated re-framing of benefit sharing as various competing accounts have been promoted.

The earliest versions advanced a duty to benefit the particular people participating in research or a somewhat wider circle of beneficiaries (communities or populations in the case of Low and Middle Income Countries (LMICs)). This is the ‘reasonable availability model’ espoused by CIOMS, which has traditionally tied the benefits to products or interventions resulting from a particular research project. An ethical prerequisite here is that research should respond to the health needs of the community and therefore any positive results of research are directly relevant to those needs.

A somewhat overlapping concept of post-trial obligations has also been argued for and applied in the context of health research, especially clinical trials. The language of post-trial obligations has its roots in the 2000 edition of the Declaration of Helsinki (§30: ‘At the conclusion of the study, every patient entered into the study should be assured of access to the best proven prophylactic, diagnostic and therapeutic methods identified by the study.’).²¹ Later versions of the Declaration specify this duty further. Post-trial obligations are often formulated as prior agreements that are signed between stakeholders before research is begun and there exist a number of successful examples of post-trial access agreements globally.²²

The reasonable availability model has been roundly criticised for a variety of reasons.²³ Most importantly, it is said that the focus on types of benefits arising from particular research projects does not adequately remove the dangers of exploitation and it unnecessarily limits the scope of potential benefits. Thus, the alternative ‘fair benefits’ model was proposed, widening the scope of potential benefits as well as beneficiaries.²⁴ Benefits should not be limited to the results of particular research projects, and the distribution of benefits could take place both during as well as after research. Yet, while the increased flexibility in benefit-sharing discussions is a pragmatically useful development, it might also involve adverse side-effects. For example, a community might agree to participate in research that will not target their health needs at all, but will provide other benefits that they need.²⁵ This means that some of the fundamental ethical premises of research in LMICs have been effectively replaced. Perhaps this is acceptable – after all, such flexibility can be construed as less paternalistic and respectful of local needs. But it could also hint at the problematic infiltration of commercial bargaining rules into health research, which I discuss further below.

The latest re-framing, driven largely by funders, construes benefit sharing as a comprehensive cooperative tool for capacity-building that is justified via the larger framework of global health research and justice concerns.²⁶ In 2002, the Nuffield Council on Bioethics suggested that healthcare-related research in developing countries should proceed through genuine partnerships that provide transfer of knowledge and technology to strengthen the expertise of local partners. More recently, a group of influential research funders (NIH, Wellcome and the African Society of Human Genetics) have launched an H3Africa benefit-sharing vision where the more established avenues of ‘reasonable availability’ and ‘fair benefits’ have been replaced by straightforward requests for capacity building as the objective of collaborative research.²⁷ Such activities thus no longer constitute simply one of the options in the extensive list of potential benefits that parties to the benefit-sharing arrangement should consult and pick from. Benefit sharing is here no longer a positive side-effect or even an intended externality to a successful research project. Rather, it has been moved to the very core – it is one of the most important reasons the research collaboration should take place at all. In many ways, this is a welcome development, as benefit sharing has often been misunderstood as disbursement of tangible research ‘results’.

15.4. What, When and How: The Practicalities of Benefit Sharing

Much of the rationale for benefit sharing is articulated in the language of principles and values. Somewhat less guidance is given on the procedural aspects – how these principles and values are to be negotiated, prioritised and enforced. In most cases, a variety of potential benefits and beneficiaries can realistically be considered based on diverse justificatory reasons and local needs. Obviously, the host population needs to be the judge of the value of benefits to itself.²⁸ An answer to a practical question of whom does one talk to when negotiating with communities should look for engagement with those who might bear burdens for research, but are not given a voice (this concerns especially the voice of women in LMICs – their meaningful participation in all phases of benefit-sharing negotiations should be required²⁹). At the same time, one needs to be conscious – and transparent – of the fact that defining and refining participant categories or negotiation partners is already a highly selective, political act.³⁰

While community involvement is a crucial part of the benefit-sharing process, the mere fact of participation and consent does not necessarily guarantee the fairness of the agreement.³¹ To ensure transparency and that involved communities and populations do have a fair chance to make up their minds about research participation, an influential statement recommended that publicly accessible repositories of previous benefit-sharing agreements be created.³² This would provide a chance for stakeholders to assess the fairness of what they are offered and would support the procedural side of benefit sharing. Critics, however, have claimed that the principles and structures of transparency and fairness that the fair benefits approach supports might turn out to be an ‘ethical Trojan horse’.³³ The proposed auction-like model could make host communities compete against each other in offering services to global research contract organisations, turning benefit-sharing negotiations into ‘a race to the bottom’.³⁴ While the funders of non-profit research or even public–private partnerships could be held accountable for checking the fairness of the reached deals, much of for-profit research lacks such oversight structures.

15.5 Worries and Future Challenges

While benefit sharing is by now a relatively standard and well-established requirement regarding ethical research practices (especially in LMICs), I would like to draw attention to several critical points that problematise the appropriateness and scope of benefit sharing in research settings.

Some of the most discussed worries associated with sharing benefits with research participants concern the dangers of therapeutic misconception and undue inducement. Research has traditionally been about serving future generations and producing generalisable knowledge. Focus on benefiting research participants introduces the risk that they might volunteer because they expect research to benefit them directly. While research participants are often well cared for, this should not be mistaken for therapy.

Undue inducement concerns instances where benefit-sharing negotiations result in overly generous and disproportionate advantages to participants such that their ability to rationally weigh the benefits and harms of participation might be jeopardised. In the LMIC context, the local public health infrastructure might be minimal or lacking; clinical trials and other types of research often offer services that are not otherwise available. Access to medical services might motivate research participation and raise the potential of undue inducement. In these situations, a proper balance between potential risks and benefits is crucial to ensure fairness and to distinguish undue inducement from fair compensation.

A different kind of unease about the extensive employment of benefit-sharing language and practices in health research was voiced already decades ago. Debates then revolved around benefit sharing as a side-effect of unwelcome commercialisation of health research. Often focused on the patenting of the human genome,³⁵ the arguments ranged from the consequentialist (threats to scientific progress as it changes the altruistic motivation for scientific research) to the deontological (metaphysical dangers to the ‘ethical self-understanding of the species’³⁶). The worry was that benefit sharing as a conceptual framework had opened health research up to the vagaries of global commercial markets and had turned it into a shameless profit-driven activity, where the services of the participants were nothing but tradable commodities.

Over the past decade, we have grown used to the increasing prominence of for-profit health research. The noble idea of volunteering for research to support the project of science that may benefit humankind is no longer easily applicable nor ethically acceptable in the context of global biomedical research where powerful for-profit companies choose to do their research among possibly vulnerable populations in LMICs. While altruistic volunteering and even a gift-relationship dynamic might still be possible for health research within affluent and more sheltered communities, it would be distinctly unfair to insist on this rationale for other contexts. Even in developed countries, fierce battles regarding patenting and access to screening tests have taken place between those who contributed to research and those who were granted a patent (e.g. the Canavan disease controversy in the USA).

A different kind of worry is that if benefit sharing is motivated by the wider concerns of global justice (‘an effective way of helping people in LMIC’³⁷), then benefit-sharing practices and procedures are not well-equipped to deal with these much larger and complex challenges arising from global (and local) political, social and economic inequalities. Indeed, numerous funders have explicitly stated that too wide a scope for post-trial or benefit-sharing obligations (bordering on aid) is not to be required of investigators; some of the funders are, in fact, prohibited from funding healthcare provision. Furthermore, while it is clear that in many cases research is undertaken by for-profit companies who may go on to earn substantial benefits, there are also numerous trials and projects that do not translate into profits and may prove unsuccessful. Yet even such research constitutes valuable knowledge that is crucial to guide further research. The framework of benefit sharing as capacity-building gets around that challenge because it no longer focuses nor depends on the tangible results but on the cooperative aspects of research where ‘negative’ results are also valuable for involved local researchers.

Benefit sharing is an attempt to offer the vulnerable and the burdened communities a fair and well-earned chance to improve their situation. This means that benefit sharing can sometimes rightfully be associated with the tendencies to commodify relations and objects that, in a different world, would perhaps be guided by other, more altruistic and less monetised motives. Yet, from the perspective of LMICs, the dynamic of benefit-sharing logic over the past decades has enabled those countries themselves to increasingly have a say in steering benefit sharing. It should no longer be constrained by a particular research project or be seen as contributing towards the local scarcities in a haphazard way of plugging the holes in responding to the most desperate needs. Rather, benefit sharing is increasingly construed as a systematic tool within the wider project of collaboration, of taking control of one’s resources and setting one’s own research and health policies and priorities. In short, it is coming to be seen as crafting a space for a ‘lab of their own’.³⁸

Such an interpretation of benefit sharing frames it as part of a more general tendency of rethinking the function and practice of research and science in society. This has been visible, for example, in the European Commission’s funding guidelines. The requirement of transparency in setting research priorities, the democratising of science through involvement of various stakeholder groups (e.g. patients) in the early stages of research, and the rhetoric of responsible research and innovation are all instances of opening up research as a social practice, shifting away from a view of research as a boxed-up end-product. Perhaps some benefit-sharing partnerships might already be viewed as examples of such ‘power sharing’,³⁹ although one should remain cautious in terms of the concept’s ability to revolutionise health research around the globe.

Benefit sharing is not immune to the many changes happening in health research: learning healthcare systems are doing away with the once central distinctions between clinical and research ethics; multi-site research makes it difficult to assess the contributions of distinct locations and partners; and it is unclear what the relationship will be between benefit sharing and data sharing in the context of open data and the increased role of health-related data in health research. Certain flexibility that has always been necessary for a successful implementation of benefit-sharing frameworks – the integration of universal ethical principles with the particular research partnerships – needs to continue to ensure that, at least as long as we live in an imperfect world of great inequalities, benefit sharing can successfully be integrated into the evolving practices of health research. Yet we need to be cautious about pinning too many hopes on that one framework.

15.6 Conclusion

Benefit sharing in health research is by now a well-established ethical requirement. There are a plethora of documents and established best practices to guide the researchers, funders and regulators, as well as communities and other stakeholders. The rationale for benefit sharing has evolved and continues to do so. Starting from the idea that individuals and communities taking certain risks and accepting potential harms deserve compensation and should not be exploited, we have now reached frameworks that view capacity-building and development support as one of the primary goals of research cooperation.

Benefit sharing is an activity that is grounded in potentially conflicting sets of justifications. While that might seem philosophically problematic (leading to e.g. various inconsistencies, potentially contradictory duties), in pragmatic terms, detailed global agreements are not necessary. It is best to regard benefit sharing as a mandatory ethics frame(work) that is to be applied to all international research collaborations as it highlights certain moral concerns and provides conceptual and governance resources for dealing with those. But the actual agreements need to be contracted by particular stakeholders and the details of the planned research and the distinct context will determine which sets of concerns are paramount, which justifications make sense, what benefits are realistic, and who should be involved. There is a danger of potential relativism involved in such a governance framework, but only combining universal research norms with unique contextual components provides the sensitivity and flexibility that is needed for ethical health research as a collaborative enterprise.

16 Taking Failure Seriously Health Research Regulation for Medical Devices, Technological Risk and Preventing Future Harm

Mark L. Flear

^*

16.1 Introduction

Failure in health research regulation is nothing new. Indeed, the regulation of clinical trials was developed in response to the Thalidomide scandal, which occurred some fifty years ago.¹ Yet, health research regulation is at the centre of recent failures.² Metal-on-metal hip replacements,³ and, more recently, mesh implants for urinary incontinence and pelvic organ prolapse in women – often referred to as ‘vaginal mesh’ – have been the subject of intense controversy.⁴ Some have even called the latter controversy ‘the new Thalidomide’.⁵ In these cases, previously licensed medical devices were used to demonstrate the safety of supposedly analogous new medical devices, and obviate the need for health research involving humans.⁶

In this chapter, I use health research regulation for medical devices to look at the regulatory framing of harm through the language of technological risk, i.e. relating to safety. My overall argument is that reliance on this narrow discourse of technological risk in the regulatory framing of harm may marginalise stakeholder knowledges of harm to produce a limited knowledge base. The latter may underlie harm, and in turn lead to the construction of failure.

I understand failure itself in terms of this framing of harm.⁷ Failure is taken to be ontologically and normatively distinct from harm, and as implicating the design and functioning of the system or regime itself. Failure is understood as arising when harm is deemed to thwart expectations of safety built into technological framings of regulation. This usually occurs from stakeholder perspectives. Stakeholders include research participants, patients and other interested parties. However, the new force of failure in public discourse and regulation,⁸ apparent in the way it ‘now saturates public life’,⁹ ensures that the language of failure provides a means to integrate stakeholder knowledges of harm with scientific-technical knowledges.

In the next section, I use health research relating to medical devices to reflect on the role of expectations and harm in constructing failure. This sets the scene for the third section, where I outline the roots of failure in the knowledge base for regulation. Subsequently, I explain how the normative power of failure may be used to impel the integration of expert and stakeholder knowledges, improving the knowledge base and, in turn, providing a better basis on which to anticipate and prevent future failures. The chapter thus appreciates how failure can amount to a ‘failure of foresight’, which may mean it is possible to ‘organise’ failure and the harm it describes out of existence.¹⁰

16.2 Expectations and Failure in Health Research

Failure has long been understood, principally though not exclusively, in Kurunmäki and Miller’s words, ‘as arising from risk rather than sin’.¹¹ Put differently, failure can be understood in principally consequentialist, rather than deontological, terms.¹² This understanding does not exclude legal conceptualisations of failure in tort law and criminal law, in which the conventional idea of liability is one premised on ‘sin’ or causal contribution.¹³ However, within contemporary society and regulation, such deontological understandings are often overlaid with a consequentialist view of failure.¹⁴

This is apparent in recent work by Carroll and co-authors. Through their study of material objects and failure, they describe failure as ‘a situation or thing as [sic] not being in accord with expectation’.¹⁵ According to van Lente and Rip, expectations amount to ‘prospective structures’ that inform ‘statements, brief stories and scenarios’.¹⁶ It is expectation, rather than anticipation or hope, then, that is central to failure. Unlike expectation, anticipation and hope do not provide a sense of how things ought to be, so much as how they could be or an individual or group would like them to be.¹⁷ Indeed, as Bryant and Knight explain: ‘We expect because of what the past has taught us to expect … [Expectation] awakens a sense of how things ought to be, given particular conditions’.¹⁸

This normative dimension distinguishes expectation from other future-oriented concepts and furnishes ‘a standard for evaluation’, for whether a situation is ‘good or bad, desirable or undesirable’,¹⁹ and, relatedly, a failure. Indeed, for Appadurai ‘[t]he most important thing about failure is that it is not a fact but a judgment’.²⁰ Expectations rely on the past to inform a normative view of some future situation or thing, such as that it will be safe. When, through the application of calculative techniques that determine compliance with the standard for evaluation, this comes to be seen as thwarted, there is a judgment of failure.²¹ Expectations, and hence a key ground for establishing failure, are built into regulatory framings²² and the targets of regulation.²³

These insights can be applied and developed through the example of health research regulation for medical devices. In this instance, technological risk, i.e. safety, provides the framing for medical devices within the applicable legislation and engenders an expectation of safety.²⁴ However, in respect of metal-on-metal hips and vaginal mesh, harm occurred, and the expectation of safety was thwarted downstream once these medical devices were in use.

Harm was consequent, seemingly in large part, on the classification of metal-on-metal hips and vaginal mesh as Class IIb devices. IIb devices are medium to high-risk devices, which are usually devices installed within the body for thirty days or longer. This meant that it was possible for manufacturers to rely on substantial equivalence to existing products to demonstrate conformity with general safety and performance requirements. These requirements set expectations for manufacturers and regulators to demonstrate safety, both for the device and the person within which it was implanted. Substantial equivalence obviates the need for health research involving humans via a clinical investigation.

It is noted in one BMJ editorial that this route ‘failed to protect patients from substantial harm’.²⁵ Heneghan et al. point out that in respect of approvals by the Food and Drug Administration in the USA, which are largely mirrored in the European Union (EU): ‘Transvaginal mesh products for pelvic organ prolapse have been approved on the basis of weak evidence over the last 20 years’.²⁶ This study traced the origins of sixty-one surgical mesh implants to just two original devices approved in the USA in 1985 and 1996. The reliance on substantial equivalence meant that safety and performance data came from implants that were already on the market, sometimes for decades, and that were no longer an accurate predicate. In other words, on the basis of past experience – specifically, of ‘substantially equivalent’ medical devices – there was an unrealistic expectation that safety would be ensured through this route, and that further research involving human participants was unnecessary.

Stakeholders reported adverse events including: ‘Pain, impaired mobility, recurrent infections, incontinence/urinary frequency, prolapse, fistula formation, sexual and relationship difficulties, depression, social withdrawal or exclusion/loneliness and lethargy’.²⁷ On this basis, stakeholders, including patient groups, demanded regulatory change. Within the EU, new legislation was introduced, largely in response to these events. The specific legislation applicable to the examples considered in this chapter, the Medical Devices Regulation (MDR),²⁸ came into force on 26 May 2020 (Article 123(2) MDR).

In respect of metal-on-metal hips and vaginal mesh, the legislation reclassifies them as Class III. Class III devices are high risk and invasive long-term devices. Future manufacturers of these devices will, in general, have to carry out clinical investigations to demonstrate conformity with regulatory requirements (Recital 63 MDR). The EU’s new legislation takes up a whole chapter on clinical investigations and thus safety. The legislation is deemed to provide a ‘fundamental revision’ to ‘establish a robust, transparent, predictable and sustainable regulatory framework for medical devices which ensures a high level of safety and health whilst supporting innovation’ (Recital 1 MDR). One interpretation of the legislation is that it is a direct response to problems in health research for medical devices, and intended to provide ‘a better guarantee for the safety of medical devices, and to restore the loss of confidence that followed high profile scandals around widely used hip, breast, and vaginal mesh devices’.²⁹

As regards metal-on-metal hips and vaginal mesh, however, there has been little or no suggestion of failure by those formally responsible, and who might be held accountable if there were – perhaps especially if it could be said there were any plausible causal contribution by them towards harm. Instead, the example of medical devices demonstrates how the construction of failure does not necessarily hinge on official accounts of harm as amounting to ‘failure’. This is apparent in the various quotations from non-regulators noted above. As Hutter and Lloyd-Bostock put it, these are ‘terms in which events are construed or described in the media or in political discourse or by those involved in the event’. As they continue, what matters is an ‘event’s construction, interpretation and categorisation’.³⁰

Failure is an interpretation and judgment of harm. Put differently, ‘failure’ arises through an assessment of harm undertaken through calculative techniques and judgments. Harm becomes refracted through these. At a certain point, the expectations of safety built into framing are understood by stakeholders as thwarted, and the harm becomes understood as a failure.³¹ Official discourses are significant, not least because they help to set expectations of safety. But these discourses do not necessarily control stakeholder interpretations and knowledge of harm, or how they thwart expectations of safety, and lead to the construction of failure³²

In what follows, I shift attention to the lacunae and blind spots in the knowledge base for the regulation of medical devices, which are made apparent by the harm and failure just described. I outline these missing elements before turning to discuss the significance of failure for improving health research regulation.

16.3 Using Failure to Address the Systemic Causes of Harm

Failure, at its root, emerges from the limited knowledge base for health research regulation: for medical devices, and other areas framed by technological risk, it is derived from an archive of past experience and scientific-technical knowledge. The focus on performance (i.e. the device performs as designed and intended, in line with a predicate) marginalised attention to effectiveness (i.e. producing a therapeutic benefit) and patient knowledge on this issue. Moreover, in relation to vaginal mesh implants, female knowledges and lived experiences of the devices implanted within them have tended to be sidelined or even overlooked. The centrality of the male body within research and models of pain, and gender-based presumptions about pain,³³ help to explain the time taken to recognise a safety problem in respect of medical devices, and the gaping hole in research and knowledge.

Another part of the explanation for the latter problem is that there was a lengthy delay in embodied knowledge and experiences of pain being reported and recognised – effectively sidelining and ignoring those experiences. New guidance on vaginal mesh in the United Kingdom (UK) has faced criticism on gender-based lines. Safety concerns are cited and it is recommended that vaginal mesh should not be used to treat vaginal prolapse. However, as the UK Parliament’s All Party Parliamentary Group on Surgical Mesh Implants said, the guidelines: ‘disregard mesh-injured women’s experiences by stating that there is no long-term evidence of adverse effects’.³⁴

The latter may amount to epistemic injustice, what Fricker describes as a ‘wrong done to someone specifically in their capacity as a knower’.³⁵ More than a harm in itself, epistemic injustice may limit stakeholder ability to contribute towards regulation, leading to other kinds of harm and failure. This is especially true in the case of health research regulation, where stakeholders may be directly or indirectly harmed by practices and decisions that are grounded on a limited knowledge base. Moreover, even in respect of the EU’s new legislation on medical devices, doubts remain whether these will prevent future harms and thus failures similar to those mentioned above. Indeed, the only medical devices that are required to evidence therapeutic benefit or efficacy in controlled conditions before marketing are those that incorporate medicinal products.³⁶

A deeper explanation for the marginalisation of stakeholder knowledges of harm, and a key underpinning for failure, lies in the organisation of knowledge production. Hurlbut describes how: ‘Framed as epistemic matters – that is, as problems of properly assessing the risks of novel technological constructions – problems of governance become questions for experts’.³⁷ This framing constructs a hierarchy of knowledge that privileges credentialised knowledge and expertise, while marginalising those deemed inexpert or ‘lay’. Bioethics plays a key role here. As a field, bioethics tends to focus on technological development within biomedicine and principles of individual ethical conduct or so-called ‘quandary ethics’, rather than systemic issues related to epistemic – or social – justice. Consequently, bioethics often privileges and bolsters scientific–technical knowledge, erases social context and renders ‘social’ elements as little more than ‘epiphenomena’.³⁸ In this setting, stakeholder knowledges and forms of expertise relating to harm are, as Foucault explained, ‘disqualified … [as] naïve knowledges, hierarchically inferior knowledges, knowledges that are below the required level of erudition or scientificity’.³⁹

The specific contemporary cultural resonance of the language of failure means that it can be used as a prompt to overcome this marginalisation and improve the knowledge base for regulation. Specifically, the language of failure can be used to generate a risk to organisational standing and reputation. Adverse public perceptions may cast failure as regulatory failure, effectively framing regulators as ‘part of the cause of disasters and crises’.⁴⁰ A perception of regulatory failure thus has key implications for the accountability and legitimacy of regulation and regulators – and such perception is therefore to be avoided by them. Relatedly, regulators want to avoid the shaming and blaming that often accompany talk of failure. Blaming can even amplify⁴¹ or extend the duration of an institutional risk to standing and reputation. This may produce a crisis for regulation, including for its legitimacy, quite apart from any interpretation and judgment of failure or regulatory failure.

The risk posed by failure to standing and reputation may prompt the integration of stakeholder knowledges with the scientific–technical knowledges that currently underpin regulation. The potential to use failure in this way is already apparent in the examples above, and perhaps especially vaginal mesh. Stakeholders have been largely successful in presenting their knowledges of harm, placing a spotlight on health research regulation and demanding change to prevent future failure.

Despite the limitations within much bioethics scholarship, there is a growing plethora of approaches to injustice, most recently and notably vulnerability, within which embodied risk and experiential knowledge are central.⁴² These approaches are buttressed by a developing scientific understanding of the significance of environmental factors to genetic predisposition to vulnerability and embodied risk.⁴³ Further, within such approaches, the centrality of the human body and experience is foregrounded precisely to recast the objects of bioethical concern. The goal: to prompt a response from the state to fulfil its responsibilities in respect of rights.⁴⁴ In the context of health research, this research can be leveraged to counter the lack of alertness and communicative failures for which institutions and powerful people must take responsibility,⁴⁵ and expand the knowledges that count in regulation.

There are mechanisms to facilitate the integration of stakeholder with scientific–technical knowledges and improve health research for medical devices. Further attention to effectiveness could yield important additional data (i.e. on producing a therapeutic benefit) on top of performance (i.e. the device performs as designed and intended). Similar to clinical trials for medicines, which produce data to demonstrate safety, quality and efficacy, this would require far more involvement and data from device recipients. Recipient involvement and data could come pre- or post-marketing – or both. Involvement pre-marketing seems both desirable and possible:

The manufacturers’ argument that [randomised controlled trials] are often infeasible and do not represent the gold standard for [medical device] research is clearly refuted. As high-quality evidence is increasingly common for pre-market studies, it is obviously worthwhile to secure these standards through the [Medical Devices Regulation] in Europe and similar regulations in other countries.⁴⁶

One proposed model for long-term implantable devices, such as those discussed in this chapter, involves providing limited access to them through temporary licences that restrict use to within clinical evaluations, with long follow-up at a minimum of five years. Wider access could be provided once safety, performance and efficacy have been adequately demonstrated. In addition, wider public access to medical device patient registries, including the EU’s Eudamed database, could be provided so as to ensure transparency, open up public discourse around safety and tackle epistemic injustice.⁴⁷

16.4 Conclusion

In this chapter, I described how failure is constructed and becomes recognised through processes that determine whether harm has thwarted the expectation of safety built into technological framings of regulation. Laurie is one of the few scholars to illuminate, not only how health research regulation transforms its participants into instruments, but how this may underlie failure:

if we fail to see involvement in health research as an essentially transformative experience, then we blind ourselves to many of the human dimensions of health research. More worryingly, we run the risk of overlooking deeper explanations about why some projects fail and why the entire enterprise continues to operate sub-optimally.⁴⁸

By looking at the organisation of knowledge that supports regulatory framings of medical devices, it becomes clear how the marginalisation of stakeholder knowledge may provide a deeper explanation for harm and failure. Failure can be used to prompt the take-up of stakeholder knowledges of harm in regulation, by recasting regulation or using its mechanisms differently in light of those knowledges, so as to better anticipate and prevent future harm and failure, and enable success. See further on users’ experiences, Harmon, Chapter 39, this volume.

Why, then, has more not been done to ensure epistemic integration as a way to enhance regulatory capacities to anticipate and prevent failure? Epistemic integration would involve bringing stakeholders within regulation via their knowledges. As such, epistemic integration would seem to undermine the dominant position of those deemed expert within extant processes. Knowledge of harm becomes re-problematised: what knowledges from across society are required by regulation in order to ensure its practices are ethical and legitimate? Integration of diverse knowledges might reveal to society at large the limits of current regulation to deal with risk and uncertainty. More deeply, epistemic integration would challenge modernist values on the import of empirically derived knowledge, and the efficacy of society’s technological ‘fixes’ in addressing its problems. However, scientific–technical knowledge and expertise would still be necessary in order to discipline ‘lay’ knowledges and ensure their integration within the epistemic foundations of decision-making. To resist epistemic integration is, therefore, essentially to bolster extant power relations. As the analysis in this chapter suggests, these relations are actually antithetical to addressing failure and maintaining the protections that are central to ethical and legitimate health research and regulation more generally.

17 Rules, Principles and the Added Value of Best Practice in Health Research Regulation

Nayha Sethi

17.1 Introduction

In this chapter I consider some important implications of adopting rules, principles and supplementary guidance-based approaches to the regulation and governance of health research. This is a topic that has not yet received sufficient attention given how impactful different regulatory approaches can be on health research. I suggest that each approach has strengths and limitations to be factored-in when considering how we shape health research practices. I argue that while principles-based approaches can be well suited to typically complex health research landscapes, additional guidance is often required. I explore why this is so, highlighting in particular the added value of best practice and noting that incorporating additional guidance within regulatory approaches demands its own important considerations, which are laid out in the final section.

17.2 The Significance of Regulatory Approaches

Determining which regulatory/governance approach (RGA)¹ to adopt is a recurring predicament spanning the diverse spectrum of health research activities. For example, a key challenge concerning emerging technologies is regulatory lapse – law’s inability to keep up with the fast pace of technological development and adoption.² Novel practices/technologies may be subsumed under pre-existing frameworks through processes of commensuration such as legislative analogy. Alternatively, it may be determined that entirely new frameworks are required.³ Pre-existing frameworks may be too rigid and restrictive, or conversely, overly flexible and permissive.⁴ Content included within RGAs may deviate substantially from what takes place ‘on the ground’, raising problems for those charged with interpreting regulation, leading to theory-practice gaps.⁵ Similarly, current approaches may fail to reflect embodied experiences of the subjects affected by regulation.⁶

Universally, then, important questions arise relating to what form RGAs should take. Should they manifest as specific prescriptive norms, which often appear in the form of rules? Or would high-level and more abstract norms, such as those typically communicated through principles be more effective? Is additional guidance needed alongside rules and principles? If so, what form should this take? Each approach can have repercussions for the patients, researchers, regulators, developers, manufacturers, technologies and other key actors, subjects and objects constituting health research ecosystems. It is imperative that prior to adopting a particular RGA, the respective benefits and limitations of different potential approaches are granted due regard.

Many spheres of health research are widely populated by rules, principles and supplementary guidance. These manifest in diverse forms including: international instruments, primary and secondary legislation, ethical frameworks, professional guidance, codes of conduct, best practice instantiations, recommendations and standards. Consider, for instance, use of patient health data for research purposes. UK-based researchers wishing to access such data must consider the requirements laid out within (among others) the General Data Protection Regulation (Regulation (EU) 2016/679) (GDPR), the UK Data Protection Act 2018 and the NHS Act 2006. Additionally, they must consider guidance from the Information Commissioner’s Office, and adhere to the Caldicott Principles and applicable professional guidance. Technical standards such as those set out by the International Organization for Standardization (ISO) must also be observed. Researchers may be required to obtain research ethics committee approval, and demonstrate due consideration and mitigation of the risks and privacy impacts of data uses and whether or not such uses carry social value and are in the public interest. Many additional spheres of health research can prove similarly labyrinthine.

Navigating such complex regulatory frameworks and interpreting provisions included within them is challenging. A balance must be sought between offering clear articulations of what is required, permitted and prohibited, while retaining sufficient flexibility to guarantee applicability across a wide array of contexts. This tension between specificity and flexibility is a recurring dilemma for regulation. Regardless of the technologies/activities under consideration, an additional balance must be sought of providing adequate coverage of the range of pre-existing activities associated with a specific type of research and simultaneously avoiding the risk of becoming obsolete when new applications of, or progressions in, those research practices/technologies appear. For example, one of the driving factors for the introduction of the GDPR was the drastic transformations in how data are used today as compared to when its predecessor, the European Directive 95/46/EC, was drafted. Given the challenges of navigating health research regulation, we ought also to consider how best to communicate norms, while supporting decision-makers in the inevitable exercise of discretion. These concerns lead us to engage with two dominant RGA approaches in health research: rules-based and principles-based approaches. The next section considers these, placing an emphasis on principles-based approaches, which, I argue, can be especially helpful for complex regulatory landscapes.

17.3 Rules and Principles-Based Approaches

To understand what rules-based and principle-based approaches are, we should briefly define rules and principles. It may be more meaningful to talk of ‘rule and principle-type features’ than attempt to provide hard and fast definitions of rules and principles. These mean different things to different people in different contexts⁷ and grey areas exist where differentiation based on any sole ‘typical’ characteristic is unhelpful. For instance, reliance upon high specificity of language as the identifying feature of rules is problematic because rules can be articulated in general terms. Conversely, principles are not always communicated through abstract language, despite frequently being described as such. Consider the Nuremberg Code 1947, the norms included within it, referred to as principles, are articulated through prescriptive language: ‘The experiment should be conducted only by scientifically qualified persons. The highest degree of skill and care should be required through all stages of the experiment of those who conduct or engage in the experiment’.⁸ Another example is the CIOMS International Ethical Guidelines for Health-related Research Involving Humans, its content being collectively referred to as ‘rules and principles’. Upon closer inspection, it is unclear which of the guidelines are rules and which principles.⁹ Given these definitional challenges, reference is made here to typical but not unequivocal features of rule and principle-like norms.

Rules are typically specific, prescriptive and fixed iterations of what to do.¹⁰ They may be conceptualised in terms of rigidity, enforceability and whether they carry legal obligations. They can be characterised according to their pedigree or the manner in which they were adopted or developed.¹¹ Examples include rules contained within the GDPR and the UK Data Protection Act 2018. According to legal theorist Alexy, ‘rules are norms which are always either fulfilled or not. If a rule validly applies, then the requirement is to do exactly what it says, neither more nor less. In this way rules contain fixed points in the field of the factually and legally possible’.¹² Rules can be considered as norms that are applicable in an all-or-nothing fashion, i.e. barring an exception to a rule, they either apply to a scenario or they do not.¹³ A rule-based approach (RBA) to regulation is dominated by such rule-like norms.

In contrast, principles are frequently characterised as high-level, general and abstract norms.¹⁴ These may be ethical and/or legal, conceptualised as broad iterations of individual or sets of ethical values, such as those included within Beauchamp and Childress’ Four Principles (Principlism).¹⁵ Accordingly, respect for beneficence and non-maleficence implies that health research should aim to provide benefit and to minimise foreseeable harm. The Four Principles are considered prima facie in nature, implying that they must be satisfied barring conflict between the principles. Within legal theory, it has been suggested that principles are optimisation requirements, i.e. norms that can be satisfied to varying degrees¹⁶ as opposed to the ‘valid or not’ quality of rule-like norms. Principles can be articulated in more general and less legally enforceable terms but equally, breach of principles can lead to legal repercussions. For example, infringement of any of the seven principles included within the GDPR renders organisations subject to fines of up to €20 million or 4 per cent total worldwide annual turnover. Regardless of whether enshrined within legislative provisions or guidance documents, principles have the potential to shape behaviour within the health research setting, given the various commitments – legal, moral, political and other – to which they can give rise. A principle-based approach (PBA) is dominated by such principle-like norms.

Choosing to adopt a PBA dominant path in preference to RBA or vice versa carries important repercussions for health research landscapes and actors navigating them. The specific question of RBA v PBA received attention within the context of financial market regulation during the shift from RBA to PBA in the 1990s.¹⁷ Different categories of PBA were identified, including full, polycentric,¹⁸ formal and substantive.¹⁹ Their commonality lies in a preference towards broad principle-like standards over detailed prescriptive and specific rules for setting standards of behaviour.²⁰ In contrast, discussions within bioethics have centred on: (1) specific content of particular rules/principles; (2) how principles ought to be balanced against each other when conflict arises – including whether certain principles ought to take priority over others; (3) which particular rules/principles ought to be in/excluded from ethical frameworks; and (4) how to extract action-guiding content from abstract principles.²¹

Within health research regulation more specifically, consideration of PBA in contrast to RBA has been more limited. Some contributions exist in the contexts of regulating the use of stem cells²² and health data²³ in health research. In those arenas, PBA has been preferred over RBA in recognition of the value of principles and limitations of rules but without concluding that approaches dominated by principles obviate the need for rules. Rules play pivotal roles in delineating ‘boundaries beyond which research ought not to stray and therefore over which society requires closer regulatory oversight’.²⁴ Rule-like norms may provide certainty to decision-makers given their typically detailed and prescriptive nature. The value of hard and fast rules, particularly manifested via legislation is not under dispute. Rather, as will become apparent, I suggest that their rigidity can leave RBA-dominated frameworks ill-suited to the demands of complex regulatory landscapes, especially rapidly-evolving health research terrains. In contrast, PBA affords the flexibility fast-paced technological change often necessitates. Principles can create and leave space for interpretation and the exercise of discretion, essential when dealing with difficult decisions and ensuring applicability across a variety of contexts.²⁵ The remainder of this section therefore focuses on PBA and, through exploration of several key functions principles can perform (often in contrast to rules), explains how they may be especially useful to health research regulation.

Principles can protect against over/under-inclusiveness of activities or subjects of regulation, in contrast to rules. Where specific and prescriptive rules are employed, there is a risk that by virtue of their rigidity, rules either fail to capture relevant activities within them, or are applied to activities that ought not to fall under their purview. It is impossible to legislate for every eventuality, particularly at the cutting edge of scientific research. Consider the proliferation of data-driven technologies that revealed the European Data Protection Directive 95/46/EC was no longer fit for purpose. Its replacement, the GDPR, seeks to better reflect the status quo viz potential data use and applications, albeit that ongoing and rapid developments in Artificial Intelligence (AI), computing and analytics are generating new regulatory concerns.²⁶ The GDPR is, however, underpinned by seven high-level ‘principles’ to be factored-in to all interpretations of activities falling within its scope. These may have more longevity and reach than prescriptive rules because principles are less likely to be as detailed and technology-specific as rules tend to be. Of course, principles may also necessitate revision, for example to reflect changes in consensus around what the overarching principles ought to be, but they are more likely to outlast the technological changes that can frequently make prescriptively drafted rules obsolete.

A further strength of principles is their interpretive/guiding function, in communicating the spirit with which more specific norms – including rules – ought to be applied, especially where tensions exist within law, e.g. simultaneously restricting, banning and promoting behaviour. This function can be observed in the approach of the UK Human Fertilisation and Embryology Authority (HFEA). It includes within its Code of Practice a series of regulatory principles to be adhered to when licensed activities are carried out under the Human Fertilisation and Embryology Act 1990, as per S8(1(ca). For example, the first principle states licensed centres must ‘treat prospective and current patients and donors fairly, and ensure that all licensed activities are conducted in a non-discriminatory way’.²⁷ Such overarching principles can guide and assist decision-makers in all of their related activities.

The paramountcy of stakeholder engagement is a strong theme within this volume.²⁸ High-level principles can provide an effective dialogical tool for engagement with different stakeholders, enabling ongoing moral debate, and identifying interests and values at stake. As I argue elsewhere, PBA are more conducive to fostering meaningful dialogue because they avoid prescribing specifically (as rules often do), what ought to be done. Further, they ‘promote reflection precisely on this point … and in particular, they offer us the opportunity to lay out the core values which matter to us in the specific context. Rules, in contrast, can do the opposite, they can either prohibit something that might not be problematic or … grant licence where there is little’.²⁹ The UK care.data debacle illustrates the danger of overreliance on rules and failure to effectively engage in discussion of core principles of concern to stakeholders.³⁰

Relatedly, the legitimacy of RGAs that are not co-produced alongside individuals/groups affected by them is problematic. For example, dominant policy framings can tend to portray innovation per se in a positive light,³¹ but some innovation is high risk. Appropriate frameworks must be developed as simultaneously responsive to potential value and dangers of innovation.³² Key to this is explicit acknowledgement from the outset of the imperative to shape the trajectory of research and innovation alongside and for society. Lipworth and Rexler call for development of a bioethics of innovation, which necessitates dialogue and engagement with stakeholders. The framework for Responsible Research and Innovation³³ advanced by Stilgoe and colleagues contains four ‘dimensions’ (anticipation, reflexivity, inclusion and responsiveness) that are akin to high-level principles and can serve as a helpful framing device through which to engage in dialogue. Indeed, as Devaney notes, PBA may have the capacity and potential to reflect, encompass and be facilitative of the process of innovation itself.³⁴

In this section, I have laid out some key strengths of PBA, illustrating why they may be better suited to complex health research landscapes than RBA. The discussion now advances to consider an equally important aspect of developing appropriate RGAs: the need for additional tools to support decision-making.

17.4 Rules and Principles: Necessary but Not Sufficient

PBA and RBA have limitations and additional tools are necessary to guide decision-makers. For instance, resolving conflict between principles is challenging. Balancing, whereby each principle is assigned a weight, is a methodology through which it is suggested competing principles may be prioritised. However, balancing implies commensurability (that each principle can be assigned a weight),³⁵ which is obviously problematic at a practical level. Further, balancing can give rise to subjectivism, decisionism and intuitionism where decision-makers justify weighting according to preconceived prejudices.³⁶ For instance, Principlism has been criticised for prioritising respect for autonomy over other principles. Balancing is also a challenge for RBA: inter-rule conflict may equally arise as ‘[r]ules look more certain when they stand alone; uncertainty is crafted in the juxtaposition with other rules’.³⁷ Conflict between competing norms may be inevitable and balancing requires both judgement and justification. Opting for one interpretation/resolution to a decision is only legitimate if it comes with well-reasoned and justifiable bases. Nonetheless, decision-makers require support in determining how to approach balancing. Likewise, concrete examples are required to elucidate how conflict between principles and rules ought to be addressed in practice.

Another criticism of high-level norms is that their abstract nature leaves too much interpretative space; extracting meaningful, action-guiding content becomes challenging. For instance, the Declaration of Helsinki states: ‘Groups that are underrepresented in medical research should be provided appropriate access to participation in research’.³⁸ This does not suggest what ‘appropriate access’ entails. Even rules, – particularly when articulated broadly – are open to challenges of interpretative uncertainty, given the ‘open texture’ of language.³⁹ Content included within rules/principles may be interpreted overly-cautiously in fear of potential regulatory repercussions, stifling important research which may actually be legally and ethically permissible, as has been the case in some data sharing contexts.⁴⁰ Alternatively, interpretative latitude can leave room for creative compliance and exploitation of excessively abstract norms. Further, any potential certainty derived by RBA or PBA articulated in prescriptive language still necessitates shared understandings of the content – especially key terminology – of rules/principles and the overall objectives to be pursued, again suggesting the need for supplementary guidance to aid interpretation.

Theory-practice gaps and the need for context-sensitivity are also significant. Failure to adequately reflect the practical realities of conducting research ‘on the ground’ risks rendering norms ineffective. For instance, health research activities during global health emergencies have revealed disparities between what regulations demand and what is practically feasible or context-appropriate. Requirements to obtain timely ethical approval and adhere to randomised control trial protocols are not always possible/appropriate in time-sensitive settings and where proven therapeutics are lacking. Traditional distinctions between medical care/practice/treatment and research/innovation activities are blurred.⁴¹ Discerning between practices primarily seeking to benefit the individual patient (treatment) and those aimed towards generating generalisable knowledge (research) is difficult, particularly regarding ‘innovative’ practices.

It is apparent from this discussion that rules and principles, while indispensable as regulatory tools, possess weaknesses that can limit their effectiveness in decision-making. As considered next, more is often required to support decision-makers, particularly in interpreting relevant norms, offering context-sensitivity and reflecting practical realities.

17.5 Supplementary Guidance and the Added Value of Best Practice

Additional supplementary guidance alongside RBA and PBA exists across many health research domains. This appears in myriad forms including: standards; guidelines; codes of practice; good practice; and, as will receive special attention below, best practice.

Clinical guidance in the form of guidelines proliferated from the 1970s onwards in the UK to achieve technical, procedural and administrative standardisation in medical practice and to maintain professional autonomy.⁴² Good Clinical Practice Guidelines and evidence-based guidance from the National Institute for Health and Care Excellence are UK-based examples. Internationally, the World Health Organization (WHO) continually issues and updates guidelines on a variety of health topics, each designed to ensure the appropriate use of evidence in health policies and interventions, in accordance with the standards set out for guideline development.⁴³

The role of guidance and its role in shaping health research practices also necessitates attention. Numerous international guidance documents exist, including the CIOMS Guidelines, which are supported by a ‘commentary’. Likewise, the fourteen guidelines included within WHO Guidance for Managing Ethical Issues in Infectious Disease Outbreaks are each accompanied by questions illustrating the scope of ethical issues and ‘a more detailed discussion that articulates the rights and obligations of relevant stakeholders’.⁴⁴

Given that regulatory landscapes are frequently complex and pre-saturated with rules and principles (in their various forms), it is paramount that the introduction of supplementary guidance is approached with caution. Arguably, more guidance alone could suffer from the same criticisms as PBR and RBR; simply more norms requiring interpretation. Important considerations must be factored-in to the design and implementation of guidance to ensure its effectiveness. For example, the legitimacy of guidance is interlinked with the sources from which it has been generated and we must ask who this gives power to. As mentioned above, guidance may represent a means for actors to preserve autonomy and freedom from external interference/control. This raises questions of fairness, justice and transparency. Consider legal and ethical issues associated with current/anticipated uses of data and AI within health research. Concerns have been raised of technology firms developing their own guidance, facilitating creative compliance and self-regulation.⁴⁵ Even where guidance is drafted by independent committees, diverse interests must be balanced. For example, in the contexts of Big Data and AI, trade-offs are apparent in the UK between protection of privacy rights of individual citizens and national ambitions of economic growth, international competitiveness and participation in the fourth industrial revolution.

Additionally, the form guidance takes carries important ramifications for legitimacy and uptake. Distinct categories of guidance exist, at times operating at different levels. Guidance comprises a broad category ranging from anecdotal to evidence-based guidelines. At the time of writing, COVID-19 is causing a global pandemic. As health systems around the world struggle to treat patients, a plethora of new guidance documents are emerging from multiple sources, based on varying degrees of evidence. These include guidelines to support decision-making around public health responses to containment and guidance to support frontline health workers in resource allocation. Where guidance diverges across different sources, challenges arise as to which guidance to follow and why. Due regard must be given to how guidelines interact with pre-existing regulation. Another important consideration is what the repercussions, if any, might be for non-observation of, or derogation from guidance.

Relatedly, this generates considerations around how compliance with guidance is to be measured, incentivised or even enforced. It also follows that fundamental questions arise around the processes involved in drafting, endorsing, disseminating and implementing guidance. Central to these, is the question of where public voices are in each of these processes, as considered elsewhere in this volume.⁴⁶

I have argued previously that best practice (BP), a form of supplementary guidance, can be particularly helpful for decision-makers within health research⁴⁷ and in ways that are distinct from other forms of guidance. For example, BP, as co-produced through inclusive and consultative processes, can provide a platform for inclusion of public(s) and additional stakeholder perspectives.⁴⁸

One example of such an approach can be viewed again in the context of AI. In recognition of the widespread development and adoption of AI applications, the European Commission has developed Guidelines for Trustworthy AI. The Commission has taken a phased approach to piloting these, including wide consultation with various stakeholders. It is notable that alongside the guidelines, explanatory notes are offered and the Commission has established a Community of Best Practices for Trustworthy AI.⁴⁹ Through the European AI Alliance, registered participants could share their own best practices on achieving trustworthy AI.

Further, a recent report on the regulation and governance of health research lamented the ‘disconnect between those making high-level decisions on how regulations should be applied and those implementing them on the ground’.⁵⁰ BP instantiations also offer concrete examples of principles or rules ‘in action’, based on lessons learned from those experienced in interpreting and applying the relevant norms. BP can thus serve an important function of helping to bridge such problematic policy-practice divides.

BP instantiations can also support decision-makers in interpreting relevant legislative provisions and/or ethical frameworks and related obligations. They provide more detailed explanatory notes on the legislative or normative intent behind overarching principles/rules. They provide a mechanism through which to make explicit to intended users of guidance what the status of such guidance is, and how it relates to pre-existing rules, principles and additionally relevant guidance. BP also guides decision-makers in approaching resolution of conflicting principles or rules, which I identified earlier as a key challenge for PBA and RBA.

Finally, I noted earlier that rules can close down conversations. BP also carries such risks; reference to BP may decontextualise and thwart discussion/use of other practices. Arguably, use of the term ‘best practice’ or ‘good practice’ suggests a superlative and that derogations from BP interpretations are suboptimal. But best practices (as construed here) are subject to constant review and revision, thus by definition, always seeking what is ‘best’ in a given context. In turn, in order to remain fit for purpose, best practices require us to constantly revisit the underlying rules/principles to which they correspond. In this regard they drive a symbiotic relationship between all of the norms in play towards an optimal system of regulatory and governance approaches.

17.6 Conclusion

In this chapter I have outlined key considerations in adopting rules, principles and supplementary guidance-based approaches to health research regulation. In particular, I have laid out the suitability of principles for guiding decision-makers across complex regulatory landscapes. I suggested that the introduction of supplementary guidance could tend to limitations of PBA and RBA. But, in turn, I stressed that generating new guidance must be approached with caution and due regard to additional concerns. Finally, I highlighted the added value that best practice – to be distinguished from other forms of supplementary guidance – can bring to complex regulatory landscapes.

18 Research Ethics Review

Edward Dove

18.1 Introduction

Across most jurisdictions today, researchers who propose to involve humans, their tissue and/or their data in a health research project must first submit an application form, which includes the research protocol and attendant documents (e.g. information sheets and consent forms), to one or several committees of experts and lay persons, who then assess the ethics of the proposed research. In some jurisdictions, this review, known as research ethics review, is mandated by law. In these cases, the law may be general¹ or it may apply to specific kinds of health research, such as clinical trials of an investigational medicinal product² or health research involving adults lacking capacity.³ In other jurisdictions, and depending on the type of research project, research ethics review may be required or expected by ‘softer’ forms of regulation, such as guidelines, policy or custom, with the processes for the review consequently less standardised – and more flexible – than in a rules-based regime.

The principal aim of these research ethics committees (RECs), also known as institutional review boards (IRBs) and research ethics boards (REBs),⁴ is to protect the welfare and interests of prospective (and current) participants and to minimise risk of harm to them. Another aim is to promote ethical and socially valuable research. This phenomenon of evaluating the ethics of proposed health research and determining whether the research may proceed – and on what grounds – has been in existence largely since the 1960s.⁵ Originally designed for review of clinical research involving healthy human volunteers, research ethics review has since expanded to cover all fields of health research, including social science-driven health research such as qualitative studies investigating patient experiences with a disease or treatments that they receive. Given their central role in determining the bounds of ethical research, it is unsurprising to learn that RECs have been subject to sustained scrutiny; in many quarters, this has resulted in criticism within the health research and academic community that, among other things, the process of research ethics review is not fit for purpose. The cumulative charge is that research ethics review by committees promotes a wicked combination of inexpert review, inconsistent opinions, duplicative work, mission creep and heavy-handed regulation of health research.

This chapter places this charge at the focal point. In what follows, I chart the process of research ethics review with a view towards arguing that RECs have become regulatory entities in their own right and very much are a form of social control of science. As I detail, while RECs are far from perfect in terms of regulatory design and performance, they do perform, at least in principle, a valuable role in helping to steward research projects towards an ethical endpoint. In what follows, I analyse the nature and aims of research ethics review and the body of academic research regarding research ethics review. In so doing, this chapter also offers a critique of existing work and suggests some future directions for both the regulatory design of research ethics review and also researching the field itself.

18.2 Research Ethics Review as a Regulatory Process

Many scholars have long viewed the notion of evaluation of the ethics issues of a proposed research project by a committee of people qualified in some way to assess the project’s ethics as necessary, but not necessarily sufficient, for the successful functioning of, and securing of public trust in, health research. RECs, it is said, reflect a pragmatic system of ‘social control’ by researchers’ academic and community peers. As William May opined in 1975: ‘The primary guarantee of protection of subjects against needless risk and abuse is in the review before the work is undertaken. […] [I]t is the only stage at which the subject can be protected against needless risk of injury, discomfort, or inconvenience’.⁶ John Robertson similarly concluded in 1979: ‘The [REC] is an important structural innovation in the social control of science, and similar forms are likely to be developed for other such controversial areas’.⁷ By influencing research in an event-licensing capacity – that is, by offering opinion on and approval (or rejection) of a research project before it commences – RECs are seen to mitigate risks to researchers, participants and society. To this extent, research ethics review can be cast as a regulatory process.

As RECs have become more entrenched in the regulatory apparatus of health research over the past half-century, they have come to hold tremendous power over how research is shaped – and thus, influence over what knowledge is produced – as well as how the relationship between a researcher and a research participant is circumscribed. As Laura Stark observes, ethics committees ‘are empowered to turn a hypothetical situation (this project may be acceptable) into shared reality (this project is acceptable). […] [T]hey change what is knowable’.⁸

But it remains unclear what exactly constitutes research ethics review. Indeed, we might ask whether RECs engage in ethics deliberation at all – and, just as critically, whether this matters to fulfilling their putative regulatory role of assessing the relevant ethics issues in a project. Perhaps the challenge lies with the term ‘research ethics review’. This suggests less of a focus on formulaic, bureaucratic – arguably synonymous with ‘regulatory’ – answers to questions (e.g. ‘Is there informed consent?’; ‘Have they used our consent form template?’) and more of a focus on seeking deeper, more philosophically engaged answers to penetrating questions, such as: ‘Do we really need informed consent here?’; ‘What sort of alternative and preferable safeguards might there be and why?’; ‘Is this research in the public interest?’; or ‘What public good might come from this research and is the financial and social cost commensurate?’.

What is reasonably clear is that a REC provides a favourable opinion only if it is assured that the ethics issues in the proposed research are appropriately addressed by the researcher – and sponsor – before the project proceeds. As the issues will vary depending on the research in question, REC members receive training and guidance about the issues they should consider, both in general and in particular cases. For example, according to the Governance Arrangements for Research Ethics Committees (GAfREC), which is a formal governance document for National Health Service (NHS) RECs in the UK: ‘The training and guidance reflect recognised standards for ethical research, such as the Declaration of Helsinki, and take account of applicable legal requirements’.⁹ If REC members learn about what research ethics is supposed to entail according to ‘recognised standards’ and take account of ‘applicable legal requirements’, we might reasonably ask whether the REC meetings themselves reflect a kind of instantiated deliberative decision-making ethics – that is, ethics as input, process, and outcome – where members individually and collectively evaluate and come to decide on the ethical acceptability of research proposals by invoking and deliberating on standards and requirements more than (ethical) norms or principles. If this is so, the REC, as a form of a decision-making body, need not necessarily ‘do ethics’ at all.

Some evidence of this comes when we shift our gaze from theory to practice. As Mary Dixon-Woods and colleagues have found in their empirical investigation of REC opinion letters to researchers:

Though clearly RECs are making firm recommendations to researchers in these [previously discussed] examples of both inconsistent and consistent advice, the source of ethical authority for the REC in coming to their conclusions is rarely explicit in the letters. GAfREC – which provides the framework within which RECs are expected to work – is not referred to in any of the letters in our sample. Specific ethical principles or even guidelines are rarely invoked explicitly, and when they are, it is to authenticate or legitimise the decisions of the committee […].¹⁰

If the REC opinion letter is a reasonably accurate reflection of the contents of a REC meeting’s discussion, then there is some doubt as to whether ethical rules, norms or principles are openly discussed. Other empirical research has affirmed this doubt.¹¹

Yet, the names bestowed upon these bodies by many jurisdictions (‘ethics committees’ or ‘ethics boards’), and the related expectation that they should engage in research ethics review – and related criticism that they do not do enough of this – may, in fact, be somewhat misplaced. I have suggested through my own empirical research that as RECs become institutionalised and professionalised, acting as multi-faceted and multidisciplinary micro-regulators of health research, and as further national and international regulations come into force that impact health research, RECs might be expected to act more as risk-assessing ‘health research regulatory committees’ writ large.¹² Somewhat similarly, based on her own recent empirical research, Sarah Babb makes the case that IRBs in the USA have transformed from academic committees to ‘compliance bureaucracies’, where specialised administrative staff members define and apply federal regulations.¹³ Even if RECs do not engage in something approaching truly substantive ethics deliberation, and this is (partly) accepted as an outcome of practical constraints (e.g. limited resources and pressed time), might they still be able to fulfil their aim of targeting areas of health research that pose moral concern, and might they still be able to mitigate the manifestation of those concerns?

Indeed, I would argue that it is not necessarily problematic to acknowledge that RECs rarely engage in deep ethics deliberation. RECs are a valuable regulator in health research, and so if there are criticisms of them, we should look to those criticisms that speak to their regulatory functions – procedures, performance and so on – more than the absence or presence of ethics deliberation per se. By focusing here, we may come to see that concerns about efficiency, effectiveness, proportionality, reduced burden and so on, must be addressed more directly. Acknowledging this is not to say that RECs cannot spot and deal with thorny ethics issues when or if they arise, but it does allow us to be arguably more accurate and honest to cast them for what they are: regulators with a gatekeeping and promotional role about getting safe and good science done.

Let us, then, look at some of the persistent criticism of the research ethics review process that speak to the regulatory functions of RECs.

18.3 REC Criticisms: Poor Design and Performance and the Fetishisation of Consent

For as long as they have existed, RECs have been subject to opprobrium from the research community and academic commentators, mainly because they are seen as under-, over- or simply mis-regulated bureaucratic bulwarks against otherwise ethical, minimally risky or non-risky research. For years, research into RECs has revealed a high level of variation of decision-making processes in RECs¹⁴ and dissatisfaction from various stakeholders.¹⁵ These criticisms can be grouped into concerns about (a) design and performance and (b) the fetishisation of consent.

18.3.1 Poor Design and Performance

Many of the problems scholars have identified with research ethics review have been due both to weak regulation – which contributes to procedural and substantive inconsistency of decision-making – and also over-regulation – which contributes to duplicative review and cumbersome and complex thickets of disproportionate regulation for research that presents minimal risk.¹⁶ In their review of US IRBs, Emanuel and colleagues identified fifteen ‘problems’ and grouped them into three broad categories: (1) structural problems deriving from the organisation of the system as established by the US federal regulations, (2) procedural problems stemming from the ways in which individual IRBs operate, and (3) performance assessment problems resulting from the absence of systemic assessment of current protections.¹⁷ Arguably, many of these structural, procedural and performance assessment problems also could be identified in RECs in other jurisdictions.

Indeed, the main design and performance concerns with the research ethics review process commentators have identified over the years include:

• inconsistency in procedures and substantive decisions within and across committees;

• delays or impediments to research due to slow-moving RECs that have no built-in efficiency incentive;

• cumbersome bureaucratisation and standardisation of application forms that are ill-suited to different types of research, that slow and muddy the process of ethics review and that lead to heavy administrative burdens for researchers;

• distortion of research methods imposed by RECs who may not be trained in research methods and are not qualified (or expected) to judge the scientific merit of applications;

• over- or exclusive reliance on prior (ex ante) review that inadequately assures that the actual conduct of research is in accordance with ethical standards;

• imposition of inappropriate consent requirements in certain types of research projects (e.g. surveys, behavioural intervention studies) that can lead to potential selection bias in participation and responses; and

• increased risk of unethical research, in part due to the ever-growing length of information sheets that participants do not bother to read, and also in part due to lengthy application forms that researchers and REC members alike either may not adequately read or quickly complete – in other words, the insidious growth of a ‘tick-box mentality’.¹⁸

The cumulative account of these concerns suggests that better regulation is needed to improve the efficiency and effectiveness of research ethics review by RECs, and this may entail, among other things, streamlining existing regulation, enacting robust standard operating procedures (SOPs), designing templates tailored to the specific type of research project, and embedding in regulation and policy the emerging notion of stewardship. But before I address these ways forward, I now turn to a second persistent criticism of RECs, namely their fetishisation of consent.

18.3.2 The Fetishisation of Consent

Another major criticism of RECs centres on their putatively over-bearing emphasis on consent forms and information sheets, and minute wordsmithing of both, that leads to inevitable elongation of the documents and thereby increased risk of non- or miscomprehension by participants, which ironically may lead to other harms not related to the research, such as stigmatisation or disrespect. Since at least the 1960s,¹⁹ commentators have argued that consent cannot and should not act as a stand-alone rampart to prevent unethical research. Yet many consider that RECs disproportionately fixate on consent as a locus for determining and setting researchers’ ethical behaviour, demonstrating ‘the acme of self-defeating ritual compliance’.²⁰ Perhaps it is because ‘these [consent] documents constitute one of the few aspects of researcher interactions with subjects – a very downstream process – that committees feel they can control.’²¹

This bureaucratic addiction to procedure and process, coupled no doubt with an uptick in legal – albeit siloed – regulation of health research, has led to a legalisation in the workings of RECs, which is to say: a fetishisation for more forms, longer forms and ongoing insistence on boiler-plate language tacked on to information sheets and consent forms so that RECs and institutions protect themselves and others from liability. Consent is treated as a panacea for all ethical concerns,²² a kind of Pollyanna-ish hope that, ‘If only we can inject all possible risks and relevant information into the form, then participants can truly exercise their autonomy’. This is not the ‘good kind’ of REC legalisation William Curran envisioned in 1969, replete with a common law-like generalisable body of precedents and principles of procedure and substance that allow the process of deliberation to flourish.²³ Instead, it is the troubling kind: rigid and overly standardised, treating ethics as a tick-box, form-ridden, technocratically structured event. Once again, this militates against ethics committees actually ‘doing ethics’ in the genuine sense that is understood of that discipline.

Given the groundswell of criticisms over the years, what, then, might be the future directions for research ethics review as a core process in health research regulation, and what might be the future directions for researching research ethics review to assess what is working well and not so well?

18.4 Future Directions for Research Ethics Review

While many support the underlying idea of ex ante ethics review by a competent committee as a means to protect and promote the rights, interests and welfare of participants, as this chapter has observed, many also have expressed dissatisfaction with the structure and function of the ethics review system and the individual processes of RECs. Multiple regulatory techniques and instruments have been employed over the years in the hopes of remedying the myriad problems attributed to RECs, foremost the concerns of inefficiency and ineffectiveness.

Scholars have proposed a number of changes to the regulatory design of research ethics review. For the purposes of this chapter, I want to focus on three that have gained attention recently and may be among the most promising: streamlining, standardisation and stewardship.

18.4.1 Future Directions for Regulatory Design

A number of jurisdictions are now streamlining the process of research ethics review in at least two ways. First, they have introduced proportionate review systems, whereby a research project that is deemed by assessors to present no (or limited) material ethics issues undergoes a lighter-touch review. In the UK, for example, under the Health Research Authority’s (HRA) Proportionate Review Service, such projects are reviewed via email correspondence, teleconference or at a face-to-face meeting by a sub-committee – comprising experienced expert and lay members – rather than at a full meeting of a REC.²⁴ The final decision is notified to the applicant by email within twenty-one calendar days of receipt of a valid application, which is a faster turn-around time than an application that goes to a full-committee review. Second, a group of efforts are underway internationally to streamline multiple REC review of multi-site research projects, which is seen as duplicative and disproportionate.²⁵ Since 2004, the UK requires only one NHS REC opinion per research project, even if the project involves multiple sites in the country. In the USA, since 2020, a revised rule in the Federal Policy for the Protection of Human Subjects – better known as the ‘Common Rule’ – generally requires US-based institutions that receive federal funding and are engaged in cooperative research projects (i.e. projects covered by the Common Rule that involve more than one institution in the USA) to use a single IRB for that portion of the research that takes place within the USA if certain requirements are met.²⁶ This ‘sIRB rule’ reflects a growing effort by regulators and policymakers in countries around the world – including Uganda,²⁷ Canada²⁸ and Australia²⁹ – to reduce the procedural inefficiencies, redundancies, delays and research costs that have become synonymous with the absence of research ethics review mechanisms designed for multi-site health research projects.³⁰

A number of jurisdictions are also working on standardisation of the processes involved in ethics reviews, with the aim of achieving more consistent outcomes in review and fairness to applicants. The Care Act 2014 in the UK, for example, requires the HRA to co-operate with several other regulatory authorities in the exercise of their respective functions relating to health or social care research, ‘with a view to co-ordinating and standardising practice relating to the regulation of such research’.³¹ Standardisation is accomplished through various means, including the introduction and maintenance of:

• SOPs to ensure procedural consistency across RECs;

• template research application forms – including information sheets, consent forms and research protocols – for researchers to devise more thorough and ethically robust applications;

• template review forms for REC members to complete when reviewing applications; and

• systems of accreditation, qualification or certification of RECs to encourage mutual trust in each REC’s processes of review.

It must be said, though, that while many commentators support standardisation as a way to drive consistency and fairness in ethics review, others blame standardisation for the growth of an undesirable ‘tick-box’ approach that many see as defining REC work today. This, however, might be a product of the continued confusion about whether we see RECs as philosophically attuned ethics deliberation entities rather than as regulatory assessors situated within a wider health research ecosystem. I have argued above that the latter view is more accurate.

Third, the emerging concept of regulatory stewardship may have resonance in reforming the regulatory design of research ethics review to better account for the network of actors involved in bringing an application through the various regulatory thresholds in the research lifecycle. A key finding from recent empirical investigation³² is the ability of actors within the health research regulatory space to serve as ‘regulatory stewards’. Research suggests that regulatory stewardship involves different actors – including RECs and others involved in the regulation of health research – helping researchers and sponsors navigate complex regulatory pathways and work through the thresholds of regulatory approvals. Collective responsibility, as a component of regulatory stewardship, requires relevant actors to work together to design and conduct research that is ethical and socially and scientifically valuable and that ultimately aims to improve human health. This can only be accomplished if a framework delineates how and when regulators and regulatees should communicate with one another and makes clear who has what responsibility and role to be played (if any) at each stage in the research lifecycle.

The regulatory environment for research ethics review could be designed to provide clearer channels for RECs – and members within them who may have closer contact with researchers and sponsors – and their own managing regulators (e.g. institutions, ministries, regulatory authorities) to engage with researchers and sponsors in improving the quality of research protocols and applications, and in working through law, regulation, and regulatory approvals. These communicative channels may include online toolkits and more personalised support via email, telephone, or digital meetings.

All of this would have the added advantage of engaging multiple actors in earlier stages of the research design process, including on the actual ethics issues (or not) that arise. Where these are considerable, the further downstream ethics review will still have a role to play; however, where these are minimal or negligible, they might be addressed sooner in the regulatory pathway, leaving the REC to undertake its regulatory role more efficiently and effectively.

18.4.2 Future Directions for Researching Research Ethics Review

Further empirical evidence is needed to investigate questions about extant research ethics review processes and to test new models that seek to improve REC efficiency and effectiveness. There have been few in-depth qualitative studies of RECs focusing on assessment of regulatory design. This undermines effective regulation, as policymakers and regulators – through state actors or otherwise – increasingly seek to develop regulation through intricately documented evidence of problems and the effects of regulation on society. There is a need for qualitative research that explores how and why RECs make the decisions they do, and how the nested dynamics of RECs and central ‘managing’ regulators play into decisions.³³

Documented problems of RECs have largely relied on evidence and anecdote proffered by researchers. While there is a welcome growing corpus of empirical literature on RECs,³⁴ more evidence is needed from regulatory scholars who can go inside RECs to test new models via pilot studies or randomised controlled trials; or who can examine how RECs, both as individual members and as a body, see themselves and their committee in a changing regulatory environment, and can go inside regulatory bodies to gather the regulators’ perspective on the roles of a REC within the health research regulatory space. Research ethics review thus remains an area ripe for investigation.

18.5 Conclusion

In this chapter, I have argued that RECs have become regulatory entities in their own right, governed by – depending on the jurisdiction – institutions, central regulatory agencies, administrative staff and offices, standardised forms and communications, and lengthy governance arrangements and SOPs. Just as some legal scholars speak of ‘juridification’,³⁵ which is an encroachment of law into ever more aspects of our society, so too might we speak of ethics review increasingly ‘colonising’ the health research regulatory space, structured according to the logic of its codes and customs. When RECs were first coming into being in the 1960s, Harvard Law Professor Louis Jaffe opined that ‘[a] general statutory requirement requiring institutional committees in any “experiment” would raise monstrous problems of interpretation, would unduly complicate medical practice, and would add unnecessary steps to experiments where the risks to the subject or patient are trivial.’³⁶

Yet this is where we stand today, with REC review required formally by law or informally by policy for an array of health research, from the trivial to the complex and risky, albeit with more proportionate review processes than occurred previously. Over time, like all of health research, the regulatory space in which RECs are situated has expanded, along with the paperwork and resources researchers must dedicate in order to pass over the ‘ethics hurdle’.

At the same time, scholars remind us that: ‘The role of the Research Ethics Committee is to advise. It does not itself authorise research. This is the responsibility of [another] body under whose auspices the research will take place’.³⁷ While technically accurate – at least in many jurisdictions – this fails to appreciate the power of a REC to control what knowledge can be produced and how that knowledge is shaped. RECs, as noted previously, are a form of social control of science. The ‘advisory’ role of a REC masks its profound ability to impact health research, which is precisely why RECs have faced such criticism and undergone reform. They are not minor actors in the health research regulatory space; on the contrary, they may be among the most important. And, as I have stressed, the obligations imposed on RECs have only increased over time as myriad regulation is brought to bear on them. Ethics and regulation must go hand-in-hand – indeed, one might say that the process of research ethics review must be co-produced with regulation, and regulation and ethical judgement are co-dependent. It is crucial that we appreciate the respective roles of each when it comes to entities such as the REC. This chapter has sought to reveal how we can better understand and deliver these dual roles.

19 Data Access Governance

Mahsa Shabani , Adrian Thorogood and Madeleine Murtagh

19.1 Introduction

Enabling researchers’ access to large volumes of health data collected in both research and healthcare settings can accelerate improvements in clinical practice and public health. Because the source and subject of those data are people, data access governance has been of concern to scientists, ethics and regulatory scholars, policymakers and citizens worldwide. While researchers have long provided colleagues access to data in an ad hoc fashion, many research funders – e.g. US National Institutes of Health, Wellcome, Bill and Melinda Gates Foundation, United Kingdom Research and Innovation, European Research Council – journals,¹ professional societies and associations,² and regulators now systematically promote the deposit of research data in repositories that aim to provide responsible and timely access to data. Data sharing aims to enable meta-analyses and creative (re)uses, reduce duplicative effort in data generation, and improve reproducibility through validation studies, so as to support data-intensive research and thereby improve human health. In many countries, routinely collected healthcare data is also increasingly being made available to researchers. In both research and healthcare contexts, technical and governance strategies for promoting responsible data sharing and access continue to evolve.

The broad sharing of health research data promises many benefits, but it can also involve risks. Health research data can reveal sensitive information about individuals (in legal terms, data subjects) and their relatives, posing risks to privacy and of discrimination and stigmatisation. Broad sharing of health research data can also raise professional concerns for the researchers or organisations who produce data in terms of receiving adequate credit and recognition for their efforts in collecting, curating and analysing data.³ Likewise, commercial research companies may be concerned their data will be appropriated or misused by competitors. Data access governance aims to promote organisational, scientific and societal interests in data re-use, while protecting the rights and interests of the range of stakeholders with an interest in data. Data access governance manages who has access to data, for what purposes, and under what conditions. Governance mechanisms include policies, due diligence processes, data access agreements and monitoring. Data access governance is closely linked to the concept of data stewardship, where organisations aim to ensure data are shared widely in the interest of science and society, while also mitigating associated ethical, societal and privacy risks.⁴

In contemporary data-driven science, data access governance often involves Data Access Committees (DACs) as the key institutional setting in which access decisions are made. DACs are diverse and may be composed of individuals with a range of relevant expertise, including familiarity with the scientific area, privacy and security, and research ethics.⁵ As Lowrance notes, ‘…[s]ome DACs are formally constituted and appointed, while some are more casual. Some publish their criteria, decisions and decision rationales, but most don’t. Some directly advise the data custodians, who then make the yes/no (or revise-and-reapply) access decisions. But many DACs make binding decisions’.⁶

Against this backdrop, this chapter examines the topic of data access governance. We discuss the underlying values and goals of data access governance, focusing in particular on the scientific and social implications for open access and data sharing, on the rights and interests of data subjects as well as those of data producers, and on the ethical conduct of data sharing. We contrast the general structural and normative components of open and controlled data access. We then present existing data access arrangements of organisations and repositories that exemplify varying modes of good practice. We argue these models exemplify the tension between promoting open access to databases on the one hand, and, on the other, protecting the rights and interests of the parties involved, including data subjects, researchers, funding organizations and commercial entities. We suggest that principles of transparency, fairness and proportionality in consideration of all stakeholders’ interests and values is key to achieving this balance. We conclude by discussing existing challenges in data access governance, including potential conflicts between various stakeholders’ views and interests, resource issues, (mis)coordination between oversight bodies, and the need for better harmonisation of access policies and procedures.

19.2 Goals of Data Access Governance

Key goals of data access governance aim to strike a balance between protecting data subjects and data producers’ rights and interests, while also promoting broad access to data to advance scientific research in the public interest.

19.2.1 Protecting Data Subject Rights and Interests and Promoting Research Integrity and Ethics

Data access governance supports research ethics principles for research involving human subjects. Minimising privacy risks to participants, respecting participant autonomy, and holding researchers accountable for the scientific validity and ethical conduct of research through research ethics committee (REC) approval and oversight, are key goals of governance of data access.⁷ These goals are increasingly furthered by engaging communities in the design of governance.

Privacy and security: Data access governance can protect participant privacy in several ways. Data access agreements, which are signed by data custodians and data users, typically include requirements regarding protecting privacy and security. Privacy safeguards include restrictions on unauthorised individual-level linkage of datasets, which may increase the re-identifiability of data, or prohibitions on attempting to re-identify participants. The greater the combinations of individual-level data for any given individual, the more likely re-identification becomes. Privacy rules in access processes are therefore often designed to control the level of individual-level data linkage. Security safeguards may include general or specific requirements to adopt physical, organisational, and technical protections, as well as data breach reporting obligations.

Respect for the provisions of ethical approvals: Data access governance models often aim to ensure users respect high standards of scientific integrity, and meet the ethical requirements related to compatibility of downstream use of data with the original consent obtained from the participants at the time of enrolment to a study and data collection. Where researchers have stated that data will only be used for certain kinds of research – e.g. disease-specific – this condition will inform the review of an access proposal by the relevant oversight bodies, notably DACs. Data access review may be informed by the following questions:⁸ Does the application violate – or potentially violate – any of the ethical permissions granted to the study or any of the consent forms signed by the study participants or their guardians? Does the application run a significant risk of upsetting or alienating study participants or thereby reducing their willingness to remain as active participants in the research? Does the application run a significant risk of bringing disrepute to study, repository or steward and thereby reducing participant trust and willingness to remain as active participants in the research?

Respect for communities and relevant stakeholders: Responding to relevant stakeholders including communities’ concerns and seeking to strike a balance between the views of different groups is fundamental to respecting these communities. This may mean championing the rights of less powerful groups and taking steps to seek out their views and actively responding to those views. In the context of data access, stakeholders include study participants and communities who provide the data, study managers and the researchers who develop the data and related resources, researchers who wish to access those data, the funders who support the studies which produce the data and the public who are the ultimate funders as well as beneficiaries of research. Each of these groups has a legitimate and vested interest in the responsible and respectful uses of data and provide a unique perspective on how such governance can be achieved. For example, study participants and community representatives sitting on oversight committees such as DACs can provide a unique insight into what other study participants may view as acceptable uses of data.

19.2.2 Data Producer Rights and Interests

One goal of access controls is to protect the rights and interests of the researchers or institutions generating data. Academically, researchers compete for high-impact publications and, in turn, for academic positions and promotions. Commercially, researchers and research institutions may compete to develop commercial applications from research findings. These considerations are often addressed through publication and commercialisation clauses in data access agreements.

Data access governance may include publication policies that seek to ensure that data producers are appropriately recognised for their contribution to science. Given that publication remains the major currency in academia, there may be a tendency for data producers to request co-authorship as a condition of access. This is discouraged for reasons of scientific freedom and accountability. Having independent DAC members adjudicating access is one remedy to the potential conflicts of interest in such practices. A compromise position is sometimes used whereby the data producer has a right to review manuscripts before publication, or to at least to be informed in advance of forthcoming publications based on (re)analysis of shared datasets. Commercialisation policies aim to ensure that the data producer benefits from, or at least does not have its competitive position harmed by, downstream use of data.

Finally, responsible data access governance requires transparency, fairness and proportionality towards participants and other stakeholders. Transparency can be improved by the publishing of policies and procedures, as well as publication of approved data recipients and plain language summaries or abstracts of approved uses. Moreover, ensuring timely and consistent access review without imposing unnecessary constraints on data access are of salient importance with regard to fairness. Where data governance seeks to achieve competing goals of openness and privacy protection, as well as meeting social and participant expectations of data use, a proportionate balance needs to be struck. Proportionality may call for different types of access controls to be applied to different types of data. Increasingly, there is emphasis that the balance between public benefit and individual risks be evidence-based.⁹

19.3 Data Access Governance: Policies, Processes, Agreements and Oversight

The values and goals of data access governance are operationalised through the policies and practices of DACs and various models of data access.

19.3.1 Controlled Versus Open Access Data

The nature of data – and the associated ethical, policy and legal issues – largely determines the access model, which can range from open to controlled to closed. Open access models generally make data available to any user, anywhere, over the internet, without financial or technical constraints. The Human Genome Project, for example, which sequenced the entire human genome, shared the sequence data openly. Subsequent publicly-funded projects sequenced more individuals and combined these data with richer social, demographic and clinical data, prompting concerns about the privacy of data subjects. Controlled access models emerged to ensure data could still be shared broadly with qualified and trusted researchers, while also protecting the privacy of data subjects and sometimes also the interests of researchers producing data. In controlled access, access is managed by a REC or increasingly by a specialised DAC, which reviews requests for data access. In this regard, DACs often carry out a due diligence review of access requests and may hold deliberations over the scientific, feasibility and ethical aspects of the request. This is in line with the recommendations issued by the Organisation for Economic Co-operation and Development’s (OECD) Council on Health Data Governance that review and approval processes should involve an evidence-based assessment and adhere to principles of transparency, objectivity and fairness. In addition, the OECD’s recommendations underline the importance of independent multi-disciplinary review with an ultimate aim of risk mitigation for individuals and society.¹⁰

19.3.2 Data Access Agreements

One component of both controlled and open access models is the data agreement (termed ‘data transfer’, ‘data access’ or ‘data use’ agreement), which establishes the conditions governing the accessing researcher’s use of the data. The terms of data access agreements typically address data subject protections, including prohibition on unauthorised linkage of individual-level data and attempts to re-identify participants, respect for consent-based use conditions and ensuring appropriate security safeguards are in place. The terms may also include protections for the rights and interests of the researchers producing data, such as publication embargoes to allow data producers the first attempt at publication or intellectual property clauses governing ownership of downstream commercialisation. Benefit-sharing clauses are important in countries with emergent research infrastructures. Other clauses may serve multiple stakeholders, such as obligations to only use data for specified purposes. Still other clauses may address the interests of science and society, such as requirements for open access publication, or to share analysis code or derived datasets. While data access agreements are legally binding if designed properly, their practical enforceability, especially across borders, is largely untested and remains a concern.¹¹ Especially where terms are associated with open access data, they are typically meant more as a means of communicating community norms to users.

19.3.3 Monitoring of Data Use

DACs may additionally develop tools and mechanisms to maintain ongoing oversight of downstream data uses. For instance, data users may be required to provide periodic reports regarding the projects in which data are being used. In addition, data users may be asked to report to the DAC the publications resulting from the data use, or issues arising from special conditions of access, e.g. risk management strategies for sensitive or potentially ‘sensational’ research, or return of incidental findings. Such oversight may enable the DACs to check compliance of the data uses, but implementation requires infrastructure and human resources that may be burdensome for DACs that do not have dedicated funding. There may also be important burdens – e.g. reporting or transparency obligations – placed on data users that discourage frivolous use. Research teams releasing data or DACs may have little ability to monitor data users or to directly sanction them for misuse, except by withdrawing or refusing access in the future. Some level of accountability is available via community reporting and norms. Research institutions, funders, journals and databases themselves may have mechanisms to hold researchers accountable for respecting their commitments.¹²

19.3.4 Maintaining Transparency

The constitution of DACs shape how policies and governance mechanisms are implemented in practice. DACs are the site around which tensions between the competing interests of stakeholders may play out and therefore, examining how they do or do not maintain transparency allows scrutiny of those governance processes. DAC members may be part of the scientific team that generated the data, though the independence of members is often advocated in order to avoid conflicts of interest. Real or perceived conflicts of interest may arise where the researcher who collected the data restricts access to potential competitors, described as data ‘hugging’ or hoarding by those advocating data sharing.¹³ And yet, data producers have important expertise: they know the affordances and limits of the data as well as its provenance. In some DACs, this expertise is recognised by including members of the study team in an advisory role.¹⁴ Furthermore, all stakeholders should have some representation in governance of data access including as decision-making members of DACs. Stakeholder engagement may also comprise forms of transparency, for example through publication of high-quality plain language summaries to communicate how study data are, or will be, used.

19.4 Best Practice Examples

Depending on the organisation or its specific needs, data access governance can emphasise different governance-related values and goals.

19.4.1 Multi-study Access: European Genome-Phenome Archive (EGA)

An example of the local access management model is the collection of study DACs under the framework of the EGA. EGA is a database of all types of ‘sequence and genotype experiments, including case-control, population, and family studies, hosted at the European Bioinformatics Institute’.¹⁵ According to the EGA website: ‘The EGA will serve as a permanent archive that will archive several levels of data including the raw data (which could, for example, be re-analysed in the future by other algorithms) as well as the genotype calls provided by the submitters.’¹⁶ Data submitters via EGA maintain control over the downstream uses of datasets via DACs located in the original study or consortium. An advantage of local data access review is that data generators who are familiar with the dataset can stay involved in the process of review and inform the access review procedure. The disadvantage of this model is that the access control is entirely left to the local committees, making it hard if not impossible to track/audit whether all data access requests are being handled in a timely manner.

19.4.2 Centralised Access: Database of Genotypes and Phenotypes (dbGaP)

In contrast, dbGaP exemplifies a centralised approach to managing data access requests. The dbGaP is designed by the National Institutes of Health (NIH) to archive and distribute the results of studies that have investigated the interaction of genotype and phenotype. Within this database, sixteen DACs ‘review requests for consistency with any data use limitations and approve, disapprove or return requests for revision’, except for large studies in which a local DAC leads access review.¹⁷ The centralised access model seems advantageous for smaller research groups who lack resources to establish their own data access review infrastructure. However, the handling of data access requests centrally may lead to latency in data access, due to complex administrative arrangements.

19.4.3 Tiered Access: International Cancer Genome Consortium/25K Initiative

The International Cancer Genome Consortium (now called the 25K Initiative) was a large-scale genomics research initiative aiming to generate and share 25,000 whole genome sequences from fifteen jurisdictions to better understand the genetic changes occurring in different forms of cancer.¹⁸ The International Cancer Genome Consortium (ICGC) adopted a tiered access approach, with open access for data unlikely to be linked to other data that could re-identify individual participants, and controlled access for more sensitive data such as raw sequence and genotype files – though the exact data types in these two categories evolved over time.¹⁹ These more sensitive data can only be accessed through the Data Access Committee Office (DACO) to protect the privacy and reasonable expectations of study participants, uphold scientific community norms of attribution and publication priority, and ensure the impartiality of access decisions. The DACO reviews the purpose and relevance of research proposals, and the trustworthiness of applicants to protect participant privacy and data security. The ICGC adopted a plain language access agreement restricting users from establishing parasitic intellectual property on primary data or attempting to re-identify individual participants, with signatures from the principal investigator and institutional signing official. Recognising that requirements for ethics review vary from country to country, the DACO asks applicants to indicate if their study of ICGC data requires local ethics approval.

19.4.4 Independent, Interdisciplinary Access Involving Stakeholder Participation in Decisions: METADAC (Managing Ethical, Socio-Technical and Administrative Issues in Data Access)²⁰

METADAC provides data access governance for only the most sensitive data and data combinations (as well as sample access). While separating access in this way produces a complex data governance setting for researchers, the devolvement to different degrees of scrutiny for differently risky data allows resources for human-mediated decision making, where this is necessary and allows administrative or algorithm-based decisions for low risk data types. The human-mediated decisions made by METADAC include a proportionate review process for routine-but-sensitive data access applications and full committee decision-making for the remaining sensitive data access applications. The METADAC committee comprises a highly multidisciplinary committee, including study-facing members (currently drawn from the participants of longitudinal studies not regulated by METADAC), with non-voting representation from the studies (including their technical teams) and the funders of these studies. Data access under METADAC does not require additional ethical approval as data sharing is based on tissue bank approval under the Human Tissue Act 2004,²¹ study ethical approval and/or explicit participant consent to sharing. METADAC’s key criteria for access follow precisely the questions outlined in ‘Respect for the provisions of ethical agreements’ above. The METADAC committee does not review the scientific merit of data access applications except in the case of finite resources (i.e. samples).

19.4.5 Data Producers’ Rights and Interests: ClinicalStudyDataRequest

ClinicalStudyDataRequest.com is a portal facilitating access to patient-level data from clinical studies carried out by pharmaceutical companies and academic researchers.²² The portal involves independent review of proposals as well as protections for participant privacy and confidentiality. A major differentiator of this access model from the publicly funded genomic research context is protection of commercial interests. For pharmaceutical company-sponsored trials, the data sharing agreement requires users to keep all information provided confidential, in part to protect commercially sensitive information.²³ The user must also agree to give the sponsor an exclusive licence to any new intellectual property generated from the study. The agreement also requires users to publish or otherwise publicly disclose their results, which helps to ensure research is pursued for verification rather than commercial purposes.

19.4.6 Transparency and Reflexive Governance: UK Biobank (Ethics and Governance Framework)

In the late 2000s, in what would be an example of reflexive data access governance,²⁴ the UK Biobank revised its Ethics and Governance Framework (to address challenges that were current at the time). More specifically, the UK Biobank had originally committed to destroy the data of participants who chose to withdraw from the biobank. However, it soon realised that it could not uphold this commitment due to technical issues.²⁵ These issues included the establishment of IT systems that made it impossible to destroy data completely in order ‘to protect the integrity and security of those people who have taken part’.²⁶ One year after identifying these issues, the UK Biobank discussed and agreed with its Ethics and Governance Council to amend the scope of its commitment: rather than destroying participant data, the biobank would commit to ensure these data would be made completely unusable. UK Biobank subsequently revised both the participant information materials and governance frameworks not only to reflect this change, but to also describe the underlying reasons. In effect, such transparency and reflexiveness could increase participant trust, and ultimately, participation in biobanks.

19.5 Challenges and Future Directions

19.5.1 Resources, Effectiveness and Efficiency of Data Access Governance

Not all research teams or repositories have the guidance, resources or expertise to establish responsible data access governance. Adequate support from funding agencies and institutions is key. This support may include establishing community data repositories to store and manage access on behalf of researchers.

Concerns regarding the workload of DACs in manually reviewing data access requests are the basis for emerging innovations around automation of at least some parts of the data access review.²⁷ One example of such efforts has been to automate the review of the conformity of the proposed data use with any use restrictions attached to the dataset – e.g. a consent agreement restricting use to non-commercial or disease specific research. In this regard, a recent initiative supported by the Global Alliance for Genomics and Health (GA4GH) developed a matrix for machine-readable consent forms. While these technical approaches will support the work substantially, there will likely always be a need for human review of the most sensitive or disclosive data access requests.

19.5.2 Coordination between Oversight Bodies

Oversight of access to biomedical databases would benefit considerably from further coordination between the relevant oversight bodies, such as DACs and RECs.²⁸ A single data-intensive research project may require access to multiple resources governed by multiple DACs, meaning multiple forms, reviews and delays. Multi-study DACs, such as METADAC, address the problem of repeated and time-consuming access processes. Requirements for multiple approvals from both ethics committees and DACs are dealt with in different ways. In the UK, for example, ethics review under the Human Tissue Act 2004 provides for broad approval for data sharing at the biobank level if relevant consents and other ethical safeguards are in place; permission for specific data access requests then only needs approval from the relevant DAC. Where national legislation is not in place, local or consortia arrangements are possible. The ICGC have disentangled ethics review from data access request review. Indeed, the ICGC’s DACO consistently maintains that its DAC is not an ethics review committee and that it should not evaluate the consent forms of users or their research protocols, relying instead ‘on the local ethics processes of the data users without imposing another layer of ethics review requirements on them’.²⁹

19.5.3 Harmonisation of Access Policies and Processes

Interoperability of data access governance supports an important goal of data science, which is to combine similar datasets together to increase statistical power and thereby produce greater scientific insight. Access arrangements are currently fragmented, differing across countries, institutions and databases. These fragmented access arrangements have the potential to undermine usability of databases and produce data silos as users battle to conform to a variety of – sometimes contradictory – access requirements and conditions. Undertaking multiple roughly similar access processes to access different databases is not only burdensome, it also does not necessarily improve participant/data subject protections. Different aspects of access review can be streamlined so that they do not have to be repeated every time a researcher seeks access. Interoperability and predictability can be improved where different data stewards adopt standard access criteria. Central access portals could accept single requests to multiple data resources. This may be possible even where there are differences between the access conditions applying to the datasets. A step further would be to delegate certain aspects of access review. A common authentication body, for example, could be responsible for establishing the identity and affiliation of researchers, who could then present a single set of credentials to different access bodies.³⁰

19.6 Conclusion

Data access governance has an ultimate goal of taking into account and maintaining balance between the rights and interests of various stakeholders involved in data sharing. A central aim of data access governance, of course, is to promote broad access to data to advance knowledge and improve human health. In doing so, it is essential to have a comprehensive overview of the rights and interests of the involved parties that might be in contrast with each other when establishing rules for data access reviews and approvals.

In view of increasing data sharing among researchers, it is crucial to ensure the DACs and RECs have sufficient resources to achieve the ultimate goals of access review, namely transparency, fairness and proportionality. In doing so, adopting a number of already proposed approaches would be advantageous, including – partly – automating the process of access review and introducing light-touch forms of review when sharing non-sensitive data.

Technological advancements could lead to heightened risks of re-identification of individuals when sharing sensitive health related data. Therefore, it is important to ensure the adopted governance mechanisms include adequate safeguards when sharing data. In addition, in establishing governance mechanisms, attention should be paid to the social values underpinning data sharing. Thereby, the focus of data governance should not be limited to only protecting the individual rights and interests of the involved parties, but also to fostering social values that can arise from promoting responsible data sharing.

20 Is the Red Queen Sitting on the Throne? Current Trends and Future Developments in Human Health Research Regulation

Stuart G. Nicholls

20.1 Introduction

Human health research is a vast enterprise; worldwide, hundreds of billions of dollars are spent annually on health research involving millions of research participants.¹ This research is guided by multiple regulations and guidance documents that commonly reflect several core principles: the protection of the rights and welfare of individual research participants; the promotion of justice in the practice and outcomes of research; and that human health research should be socially valuable (see van Delden and van der Graff , Chapter 4, in this volume). In addition, oft-cited goals of health research regulation include the development of a culture of ethical concern among researchers and institutions, and the maintenance of public trust in the research enterprise.²

However, these generally accepted principles belie an ongoing tension between the protection of individual participants through appropriate regulation, and the facilitation of health research.³ Authors have, for example, written regarding the amount of waste in research, including inefficient research regulation and management.⁴ Others have pointed to the variation in decisions⁵ and time taken – with associated costs⁶ – of obtaining ethics approval. This criticism has led to increased focus on the efficiency and effectiveness of human research regulation.⁷

In this chapter, I highlight areas that have and, I suggest, will continue to stretch health research regulation, requiring the regulatory infrastructure to adapt and evolve in order to be both effective and efficient. In doing so, I point to changes in risk assessment considerations, underlying trends toward harmonisation and streamlining of research regulation, and alternative approaches to consent. However, I also highlight countertrends that may serve to undermine these changes. Thus, like the red queen in Lewis Carroll’s Through the Looking-Glass, I propose that the health research regulatory system runs and runs as fast as it can, only to remain in the same place.

20.2 Adaptation to the Environment: The Changing Research Landscape

The evolution of the research regulatory landscape is shaped by new technologies and research approaches. In this next section, I consider several areas that have pushed, and I suggest will continue to push, the boundaries of human research regulation, namely: increasingly diverse and multijurisdictional research; Big Data and artificial intelligence in health research; the learning healthcare system, and; emergency/disaster research.

While multisite studies are by no means novel,⁸ the scale of research has exploded, with a proliferation of data repositories, biobanks and other sources of data (see Shabani et al., Chapter 19, this volume). It is now common for research to routinely cross jurisdictional boundaries, with consequent variability in experiences with regulations that exist.⁹ This has prompted discussion of ways to facilitate and improve the regulation and oversight of multijurisdictional research,¹⁰ including the development of data sharing structures.¹¹

However, the size of data is just one element. Research is now generating new types of data that stretch existing regulations. Social media, smartphones and wearable technology are being used as sources of data. Changes in hardware have been accompanied with rapid developments in analytic methods with a variety of Artificial Intelligence (AI) and other computationally heavy approaches (see Ho, Chapter 28, in this volume). These Big Data approaches have raised questions about access to data and its use,¹² as well the status of information generated by wearable devices¹³ and whether this constitutes health data that would be protected under privacy regulations.

Finally, the contexts of learning activities have raised fundamental questions about how different activities ought to be regulated and what oversight they should be subject to. Two examples of this are the learning healthcare system (LHS), in which research is incorporated into routine clinical practice and research within humanitarian crises,¹⁴ which has pushed regulatory and oversight processes due to the emergent and time-sensitive nature of the research (see Ganguli-Mitra and Hunt, Chapter 32 in this volume). While both of these activities blur research and practice, research conducted within disasters or humanitarian crises raise additional concerns regarding the inclusion of participants who are in a vulnerable position, the changing nature of risks in a time of crisis, and logistical issues in forming ethics review committees and conducting review in this context.¹⁵

20.3 Supporting Socially Valuable Research and the Role of Risk

In order to adapt to these changing environmental stressors, as well as the noted pressure from researchers to improve the efficiency of regulation and oversight, human health research regulations continue to evolve. Indeed, the vast increases in stored data and biological materials were an explicit driver for revisions to both the 2016 Council for International Organizations of Medical Sciences (CIOMS) International Ethical Guidelines for Health-related Research Involving Humans and the US Code of Federal Regulations, Title 45, Part 46 (45 CFR 46; herein referred to as the Common Rule),¹⁶ while research in humanitarian crises were also an addition to CIOMS 2016, prompted by experiences with Ebola and other humanitarian crises.

One aspect to these developments has been an evolution regarding how risk is assessed and responded to. In part, this is driven by greater attention to the perceived need to balance risk against the potential social value of research with historically under-researched populations, for example, pregnant women, children, or patients with co-morbidities,¹⁷ as discussed in the recent CIOMS revisions. The regulatory response to managing risks generated by research with traditionally under-researched or vulnerable populations – such as with disaster research – while supporting socially valuable research remains a live topic of debate.¹⁸

Further, the concept of minimal risk is increasingly raised in the regulatory context; it is a precondition in the US regulations – both Common Rule and forthcoming US Food and Drug Administration (FDA) regulations – with respect to approvals for alteration of consent, and particularly a waiver of consent, as well as featuring in EU Clinical Trials Regulation.¹⁹ This will be particularly relevant for comparative effectiveness research within the LHS and where many interventions being assessed may be argued to be usual care and minimal risk.

Moreover, the LHS and disaster research blur the line between research and practice and several examples now exist where controversy has erupted over whether an ascribed activity would constitute research or not.²⁰ Indeed, the LHS, disaster research, and secondary research use of data collected for routine clinical care problematise traditional frameworks for governance given the close approximation of activities to both research and practice.²¹ The blurring of the traditional distinction between research and care will continue to press regulations regarding the types of research that require regulatory approvals or ethics review, as well as the type and level of oversight they need.

20.4 Responding to Criticism and Supporting Research: Simplifying and Harmonising Processes

A feature of many recent regulatory changes is the effort to reduce the burden from the research ethics review process in order to meet the efficiency demands of researchers. Changes in the USA reflect an ongoing trend of streamlining and harmonisation of ethics review structures and processes, which in some countries – such as the UK – have been ongoing over the last two decades.²² Other regulatory reform, such as the EU Directive 2001/20/EC on clinical trials, have sought to further harmonise approaches between jurisdictions.²³

These changes attempt to both simplify approaches by exempting certain types of research from the need for ethics review – such as happened with changes to the US Common Rule – and reduce the need for multiple review. Examples of such streamlining include changes in the Common Rule to implement a single Institutional Review Board (sIRB) as the board of record. Indeed, this is a common theme in many initiatives.²⁴ In Canada, one such as example is the Ontario Cancer Research Ethics Board (OCREB), which serves as the single board of record for multicentre oncology trials for twenty-six of twenty-seven cancer research sites in the province.²⁵

A second trend in recent regulatory changes has been to simplify consent requirements. Indeed, it is a longstanding point of discussion that written consent forms do not necessarily facilitate informed decision-making on the part of potential research participants²⁶ due to their length and complexity of language.²⁷ Recent changes to the US Common Rule have served to try and improve these consent processes²⁸ and the FDA has sought to open up the possibility of waivers of consent to align with the Common Rule.²⁹ Moreover, alternate forms of consent were also part of the recent CIOMS revisions, which provide guidance regarding consent approaches and alternatives to prospective individual written consent.³⁰ Indeed, some authors have recently argued that in the context of standard-of-care comparative effectiveness trials – such as those envisioned within a LHS – consent could (or should) be waived.³¹

This trend of alternate consent models will, I propose, continue due to developments in disaster and emergency research and the LHS, where the need for timely and expedient consent approaches – or even waivers of consent – will continue the push for the simplification of, or alternative approaches to, consent processes. Meanwhile, Big Data applications and the secondary use of existing data will further press regulations regarding the types of research that require explicit consent from participants and how this is managed.

20.5 Countervailing Trends

Alongside these trends in the regulation of health research, there are emerging examples of regulatory or oversight processes which may run counter to those listed above and may serve to nullify the potential impact of streamlining initiatives or attempts to create a more efficient regulatory environment.

One such example has been the implementation of ‘permission to contact’ or ‘consent to contact’ policies. While such policies vary, they represent a broad consent approach to being contacted about future research and are generally asked of all patients upon arrival or intake at hospital, with the goal of expediting recruitment and easing the burden on clinical investigators by allowing researchers to contact potential patient participants. Despite the intuitive appeal, this may, in fact, increase administrative burden and create uncertainty and inefficiencies in research. For example, studies have suggested that patients may customise what aspects of their data they could be approached about or could limit elements, meaning further review of records would be needed.³² Others have suggested that while uptake to specific research may increase among those giving permission to contact (compared to those recruited through traditional physician-mediated contacts), there may be significant differences in the age and gender of those who agreed compared to those who declined the permission to contact form.³³

A further consideration is whether a permission to contact agreement constitutes a valid consent. Studies indicate that there may be confusion among patients as to whether they were immediately signing up to research,³⁴ and this may be especially true when patients are recruited upon arrival to the hospital prior to diagnosis.³⁵ Consequently, permission to contact approaches not only have the potential to increase administrative burden, thus potentially negating efficiencies created elsewhere, but also raise important questions as to the status of such permission and how it should be considered within the regulatory process.

Other inefficiencies may be introduced through additional regulatory committees.³⁶ Friesen and colleagues, for example, note a multitude of committees that now commonly exist in the USA in addition to Institutional Review Board (IRB) review, with inconsistency in the policies and processes of many of these additional bodies.³⁷ A recent study indicated that the requirement to complete multiple local governance reviews had the effect that overall times to the start-up of research were no less than if multiple local IRB review – as opposed to sIRB review – had been used.³⁸ As such, the introduction of additional governance and oversight structures appear to negate potential benefits generated through streamlining initiatives for research ethics review procedures.

20.6 Conclusion: A Parting Call for the Evaluation of Regulatory Changes

To conclude, we appear to be moving toward an era in which research requires regulation to not only protect participants, but also to be responsive, adaptive and supportive of research; responsive in its ability to facilitate research that can address emerging and emergent topics; adaptive in so far as it is malleable and flexible enough to cope with the ever-changing technologies and data needs of health research; and supportive insofar as the regulatory responses facilitate socially valuable research with no more bureaucracy than is necessary. I have proposed several areas where research regulation will continue to evolve in the near future – streamlined review processes, a focus on the management of risk, and regulatory changes to facilitate and support alternatives to traditional written consent forms.

Yet we also need to be mindful of trends that may run counter to these and which may impede progress. The noted proliferation of additional committees seems to serve only to introduce more variation and regulatory hurdles. Viewed alongside attempts to streamline ethics review, this may be seen as giving with one hand while taking away with the other. Experience in the UK with the rise of research governance following the streamlining of ethics review should serve as a historic warning.³⁹ Similarly, the introduction of permission to contact mechanisms may, in theory, lead to streamlined research by allowing researchers to directly contact patients – as opposed to mediated contact via physicians – who have opted into research, yet questions remain as to the status of such a consent as well as the practicalities of conducting a system which may well create the need for additional checks that could serve to increase administrative burden rather than decrease it. Based on the above, we appear doomed to repeat the failures of the past and despite running as fast as we can to make changes, we may ultimately stay in the same place.

This leads to my final parting call: the need for good evaluative research of changes to health research regulations and oversight systems. Despite all the discussion regarding changes to the regulatory landscape, this is largely an evidence-free space. Indeed, while the Notice of Proposed Rule Making issued in advance of changes to the Common Rule ran to 131 pages, there was a distinct lack of evidence regarding how the proposed changes addressed purported deficiencies in the oversight of research.⁴⁰ Moreover, there is a paucity of evidence regarding whether changes to sRB approaches have had the sought-after effect on review processes. Klitzman, for example, notes that the introduction of policies regarding the use of sIRBs has been done ‘in the absence of the systematic collection of data’,⁴¹ while Rahimzadeh and colleagues note that, as yet, one cannot point to rigorous evidence that the sIRB model offers great advantages over current practice.⁴²

Without empirical evaluation, we will not know whether changes in regulation or oversight are having the desired effect, or whether they increase inefficiencies and convey no benefits to research participants. This potential is borne out by the handful of studies that do exist and which show that, despite a good theoretical basis for an intervention, some may not bring about the desired improvement.⁴³ Thus, there is a need to develop strong theory-based approaches, research designs that bring rigour to the evaluation of regulatory change and a culture change that views evidence-based approaches not with scepticism but with openness.⁴⁴ Without the collection of data and open sharing of results, we cannot develop the necessary learning regulatory environment as envisaged and laid out at the outset to this collection.

21 Regulatory Authorities and Decision-Making in Health Research The Institutional Dimension

Aisling M. McMahon

21.1 Introduction

Institutional theories examine the way in which policies and decisions are structurally determined by institutions. ‘Institutions’ traditionally included state institutions such as the legislature and executive, but can also refer to embedded systems of rules, branches of law, etc. evident within particular organisational contexts. Institutional contexts or systems of rules and practice provide the foundation upon which decision-making within that context takes place. It is also within such institutional contexts that decision-making actors – including decision-makers within regulatory entities such as the Human Tissue Authority, Health Research Authority (HRA), Medicines and Healthcare Products Regulatory Agency and NHS trusts in the United Kingdom – operate. Institutional theories broadly suggest that institutions – and specific institutional contexts within which decision-making actors operate – influence and at times constrain decision-making actors in their decisions. Yet, while there is a body of research on organisational contexts and institutions within sociological studies and political science,¹ the discussion of the effects and influences of institutional contexts on downstream decision-making outcomes and actors is limited within mainstream legal literature. Instead, such questions of institutional effects are often confined to branches of legal theory,² and such institutional influences consequently remain largely under-explored in discussions of decision-making within medical law, including the health research regulation (HRR) context.

This chapter seeks to fill this gap. It argues that institutional influences give rise to engrained institutional predispositions within any decision-making context, which can significantly influence decision-making actors and hence the application of, inter alia, legal rules and professional guidance in practice. The chapter argues that the effect of institutional frameworks is particularly acute where discretion on the application or scope of a legal provision, guidance or rule is left to the decision-maker. Accordingly, the chapter argues that institutional factors and contexts should be very carefully scrutinised when adopting policy/legal changes, and particularly when drafting new provisions or guidance to be applied in any legal/regulatory context, including in the HRR context. Moreover, it will be argued that to achieve effective change within any context – ‘effective’ defined here as a change that fulfils the outcomes intended – it is not be sufficient merely to consider the text of a new provision to indicate how it is likely to be interpreted in practice. Instead, one must also consider the institutionalised context within which that provision or rule will operate.³ It is only by considering a provision within its institutionalised setting that one will gain a more holistic picture of how the rule/provision is likely to be developed and applied by that decision-maker. To this one might argue that some rules leave no scope for discretion and hence must be applied in a regimented manner. This may be the case in some contexts; however, no context is static, and none should be viewed as such. This is because social or technological change often requires decision-makers to apply a rule in a situation for which the rule was not designed. Thus, discretion can emerge within contexts over time. Furthermore, the HRR context, by virtue of the constant developments within medicine and science, can be significantly affected by both social and technological change. Thus, the potential for institutional effects are vital to consider in HRR. Moreover, arguably laws/guidance must contain sufficient tolerance – in the sense of providing a space for the expansion of rules to new social/technological contexts – so that they can evolve to meet new circumstances over time.⁴

Furthermore, arguably, the effect of institutional factors cannot necessarily be accommodated or altered by training (of decision-making actors) from a top-down level. Guidance or training may help to move these decision-makers in a particular way in a particular context, but there is no guarantee that this guidance/training will be assimilated within an institutional framework to be used in other contexts. Instead, if the arguments are borne out, those designing and seeking to implement (legal/policy) change must be mindful of both the change suggested and also how this may be assimilated and interpreted within the institutional framework where decision-making bodies – responsible for the interpretation/implementation of such changes – are situated.

In making these arguments, the chapter is structured as follows: Section 21.2 sets out the nature of how institutions and institutional contexts are defined; Section 21.3 then draws on institutional theories to set out, in brief, a template of the main institutional influences in any decision-making framework – dividing these into key constraining and predictive influences. In doing so, it does not aim to provide an exhaustive list of all potential institutional influences in any context; rather, it sets out a template of key influences that are likely to form the core institutional scaffold of any context. Hence, such influences should be carefully considered when adopting decisions/changes in HRR. To demonstrate the practical significance of these arguments, examples of institutional influences within the HRR context are highlighted throughout this section. Section 21.4 concludes arguing that it is vital to take account of institutional contexts and their influence on decision-makers in HRR if we are to achieve the desired outcomes of policy and legal changes.

21.2 Background: Defining Institutions and Institutional Contexts

Defining what is meant by the term ‘institution’ in any given theory can be a complex task⁵ because several ‘institutions’ may be identified in any process, depending on the level of decision-making and influences investigated. As Weinberger argues, institutions are so varied it is ‘impossible to set down a unified class of attributes to define all of them’.⁶ Indeed, ‘[t]here is … no commonly accepted view of what kinds of institutions exist, or what a typology of institutions ought to look like.’⁷ Despite this diversity, there is a ‘general agreement on a broad conception of institutions as systems of rules that provide frameworks for social action within larger rule-governed settings’.⁸ Institutions are seen as incorporating formal procedures or norms within an organisational structure, and can also include informal aspects, such as aspects of culture within broader society or even within formal organisations’.⁹ As North states, institutions are:

… the rules of the game in society or, more formally, are the humanly devised constraints that shape human interaction. In consequence they structure incentives in human exchange, whether political, social, or economic … Conceptually, what must be clearly differentiated are the rules from the players. The purpose of the rules is to define the way the game is played. But the objective of the team within that set of rules is to win the game …¹⁰

There is disagreement among institutional theorists about whether organisations – including e.g. international organisations, regulatory entities – are institutions.¹¹ Nonetheless, Hodgson argues that:

Organizations involve structures or networks, and these cannot function without rules of communication, membership, or sovereignty. The unavoidable existence of rules within organizations means that, even by North’s own definition, organizations must be regarded as a type of institution …¹²

Moreover, for the purposes of this chapter, the question of a distinction between organisations and institutions is not crucial, because although a formal organisation per se may not be perceived as a specific institution under a given theory, the crux of these theories is still applicable. This is because, if an institution other than a formal organisation is the main site of investigation within a given theory, it is arguably merely taking a different level of analysis or emphasis as its starting point. For example, the EU could be viewed as the overarching amalgamation of the different institutions (of the type described) which, taken together, form the overarching framework for decision-making, within which a decision-making body such as the European Medicines Agency sits. Similarly, a hospital (clinical) ethics committee will sit within a broader hospital management system, which then sits under an overarching NHS context (in the UK). Under such conceptions, the primary argument remains the same, i.e. that the overarching organisation/institution – which may have sub-organisations or entities – comprises a framework peculiar to that entity and within which decision-making actors are situated.

Bearing in mind the foregoing definitional points, the chapter argues that such institutional frameworks offer embedded influences on decision-maker(s) within that body who apply/interpret (legal/ethical/professional guidance) provisions. Therefore, the role of institutional contexts should be carefully scrutinised in the development of effective systems of HRR.

21.3 Institutional Factors and Decision-Making: A Template of Influences

As Immergut states: ‘institutions … act as filters that selectively favour particular interpretations either of the goals toward which political actors strive or of the best means to achieve these ends’.¹³ Institutional frameworks provide the scaffold within which decisions are taken; yet, these factors are often ignored or viewed as neutral within the decision-making context. This section challenges this view, arguing that such institutional influences can be highly significant. It argues that two main types of institutional influences can be identified, namely: (1) prescriptive (constraining) influences that legally constrain the scope of a decision-maker’s actions, e.g. the legal competences of an adjudicative body and (2) predictive influences – e.g. political influences on a decision-making body; such influences, although not legally constraining, can be used to predict and/or explain the way in which decision-makers may act, particularly in relation to controversial issues.¹⁴ Under the categories of prescriptive and predictive influences, four main institutional influences can be discerned (described below) and can be applied to any decision-making framework. Such influences, depending on the legal context applicable, can oscillate between being merely predictive to prescriptive. The template highlights two primarily prescriptive influences, namely: the central objectives of an institution and the path dependencies (this factor may be either constraining or predictive in nature depending on the context); and two primarily predictive influences, namely, the composition, decision-making structure of an institution and the inter-institutional influences (again, this factor may be constraining or predictive in nature).

21.3.1 Central Objectives of the Decision-Making Body

A central factor of influence for any decision-maker is the main objective of the overarching institution within which it sits, and, if relevant, the core objective of the legal provision/guidance it is responsible for applying. MacCormick’s account of institutions of law highlighted the importance of having a grasp of the function or main point of an institution.¹⁵ He stated: ‘an explanation of any institution requires an account of the relevant rules set out in light of its point’.¹⁶

According to MacCormick, this does not mean that institutions cannot be used for a variety of purposes. However, if used for other purposes, then ‘it is the institution that normally functions towards a given broadly-stated end – its ‘final cause’ – that is so adapted’.¹⁷ As a corollary, arguably, if the institution is not adapted, tensions can arise between the new purpose and how this is carried out, and whether it aligns with the core purpose. If a new purpose moves too far from the core broadly stated aim of the institution, that new purpose unlikely will be achieved without institutional change/adaptation. Put simply, the institutional context is not fit for purpose in such contexts.

Several factors can be used to assess the final cause/objective within an institutional context, including the mission statements, self-descriptions of the overarching institution and the text/articles of founding treaties/legislation that the decision-making body is responsible for applying.

An example of this in HRR could include a scenario whereby the HRA adopted policies that went beyond its initial – albeit broadly defined – remit. These policies may be watered down in practice, as their interpretation will likely be interpreted in a way that aligns with HRA’s core purpose. Similarly, if the HRA is perceived by the public/stakeholders as over-stepping its remit in adopting a new policy, the HRA’s actions could be challenged/criticised and knowledge of this may also influence policy change within HRA or other bodies, particularly in areas where its remit is not entirely clear. A ‘conservative’ approach to policy change may be adopted to maintain its ‘acceptability’.

Thus, in setting out the remit of a body in HRR, steps should be taken to ensure that its defined remit provides scope to take actions where needed or can be amended if uncertainty arises related to a body’s remit. Moreover, because institutional theories suggest that the actions of decision-makers will be influenced and applied in furtherance of their central objectives, when legal change or adaptation is sought, one must consider how such change is likely to align with the overarching aims of the framework within which decision-makers are situated.

21.3.2 Institutional Structure, Role and Composition of the Decision-Making Body and Overarching Institution

Second, a key predictive influence is the institutional structure, role and composition of the decision-making body within which that decision-maker sits. This influence draws particularly on March and Olsen’s work, who have argued that institutions ‘are constitutive rules and practices prescribing appropriate behaviour for specific actors in specific situations’.¹⁸ They emphasise how the internal rules of operation or structural elements within decision-making bodies impact upon decision-making outcomes.

The institutional structure is significant as it facilitates access to, and participation in, the decision-making process. This influences aspects such as the level of external opinion in the decision-making process and the types of actors involved or consulted by decision-making actors, thereby shaping the contours of decisions. Furthermore, the avenues in which decisions are structurally made, and shaped (and by whom), can help predict the types of issues likely to be considered by decision-makers. Key factors include (a) the decision-making structure within each institution, the composition of decision-makers, the levels of appeal (if any), and (b) the level – and avenues – of consultation, including mechanisms for public/external participation in the decision-making process.

Furthermore, the role and composition of the decision-making actors themselves may prove influential as it feeds into what Stanley Fish¹⁹ termed an ‘interpretative community’.²⁰ This is a community ‘working with a shared set of assumptions, understandings, conventions and values that settles issues and problems of interpretation’²¹ within the given system. In short, the decision-making community operates with shared understandings – depending on the composition of that decision-making body – and uses such understandings to interpret rules/provisions applicable by them in HRR and other contexts.

Support for this is gleaned by Powell and DiMaggio’s work, which draws on the concept of isomorphism: a mimicking or homogenisation arising across organisations in a similar field. The authors conceptualise isomorphism as ‘a constraining process … [which] forces one unit in a population to resemble other units that face the same set of environmental constraints’.²² One category of isomorphism is normative isomorphism whereby organisations within a field become composed of groups of similar professionals, creating:

… a pool of almost interchangeable individuals who occupy similar positions across a range of organisations and possess a similarity of orientation and disposition that may override variations in tradition and control that may otherwise shape organizations.²³

Accordingly, similar ways of thinking may develop within an institutional context such that decision-making actors act in ways they are familiar with, reinforced or validated by similar groups working in parallel organisations with similar thinking styles. This can lead to similar patterns of action or outcomes across decision-making bodies in the same field – even when faced with different environmental factors such that the resulting outcome is unsuitable to their environment.

Considering the composition of decision-making actors can provide valuable lessons, as in some cases, the greater the similarity of the backgrounds of individuals on a decision-making body, the greater the likelihood for institutional pre-dispositions in favour of certain outcomes. Engrained patterns of thinking may develop and become difficult to shift.²⁴ Arguably, this provides a strong justification for ensuring, where possible, a multi-disciplinary approach within HRR decision-making contexts. Such an approach is crucial to help ensure conversations do not become engrained within aligned or shared understandings to the extent that broader considerations are missed by decision-makers.

This argument is not a criticism of the need for, or role of expertise, within the HRR context or elsewhere. Indeed, expertise is vital. However, arguably, within any regulatory context, if decision-making actors are drawn from similar backgrounds, we must be conscious that this can lead to ‘stickiness’,²⁵ or a failure to question traditional lines of thinking and engrained institutional predispositions towards certain decision-making avenues/outcomes.

Two main related challenges arise from the foregoing from an HRR perspective:

1. 1. Issues of accountability and risk aversion: As each level of decision-making is accountable to the next level within the overarching institutional structure, this can lead decision-makers to adopt risk averse behaviours with positive and negative effects. On the one hand, within HRR, adopting risk-averse strategies can mean that health research bodies insist on strict compliance with research policies, such as requiring strict adherence with data protection guidance for participants. This can assist in maintaining high standards within health research. However, it can also mean that decision-makers asked to, for example, share a health dataset for research in a context where promotion of research is not explicitly part of that organisation’s mission statement, may be reluctant to do so. Such decision-makers may fear reprisal if they allow data use and if there is a lack of explicit institutional policies around how data should be shared to promote health research. The ‘safest’ option for a decision-maker in such a context in terms of protecting itself from potential challenge/liability, is to maintain the status quo and refuse to share datasets. This may be entirely contrary to the public interest but could be seen by some institutions as a safer self-preservation strategy.

2. 2. Public v Private Duties: Relatedly, there is an inherent tension within HRR between how to protect individual rights in health research and maintaining the public interest in conducting scientifically sound and ethically robust research which effectively promotes human health. This duty to promote both individual and public interests is enshrined within UK law,²⁶ but the two goals can be in tension within HRR.²⁷ Where tension arises, arguably the HRA will lean back on a default of protecting individual interests over public interests because institutional incentives may be stacked in this way, given that: (1) it is more likely for challenges to arise from a failure to protect named/groups of individuals, than a duty to promote the ‘public interest’, which often has limited teeth in practice and (2) historically, the focus within health research has been to protect individuals to avoid many of the scandals of the past. Hence, there may be engrained institutional preferences for individual interests over broader public interests. This chapter recognises the intentions behind such approaches as laudable, but we must be aware of its shortcoming in terms of promoting public interests.

In short, understanding the role and composition of the decision-making bodies within any institutional context is vital to discerning the interpretative community evident. This allows consideration of normative biases that may develop, how these may affect the provision being interpreted and ways to mitigate this where deemed necessary.

21.3.3 Path Dependencies and Historical Influences

Path dependency is generally understood within institutional theories as the influence of historical actions on present acts.²⁸ At the most general level, it implies that ‘what happened at an earlier point in time will affect the possible outcomes of a sequence of events occurring at a later point in time’.²⁹ It implies that the way a particular issue – or analogous issue – has been dealt with in the past by the institution will be influential, but not necessarily determinative, of present action(s). Within institutional theories, individual decision-makers are seen as influenced not just by past actions of the institution within which they are situated, but also by their own past actions and experiences.

In a legal/regulatory context, this influence can be either prescriptive or predictive in nature. Prescriptive or legally constraining influences can include, for instance, the principle of stare decisis, whereby in a common law context ‘reliance upon binding precedents leads courts to begin every case with an examination of the past’.³⁰ If one is looking to how legal changes might be interpreted by courts, an examination of the relevant judicial/quasi-judicial context requires an investigation of how the case law on similar provisions developed in the past, whether any pattern can be discerned, analogies likely to be drawn and mapping a trajectory of influences based on what precedents are likely to bind decision-makers. Similarly, HRR decision-making bodies would likely look to their past findings/decisions for future decisions to ensure consistency in decision-making (e.g. to protect itself from review/appeal).

For example, the HRA’s Confidentiality Advisory Group (CAG),³¹ in developing its own approach to advising on whether to approve requests for researchers to use patient-level data for ‘secondary’ health research purposes within the public interest, made public its decisions in this context as precedents for its future work, and to guide future applicants.

Furthermore, past actions of a body can be relevant as suggestive of predictive influences. For instance, the past experiences and criteria for appointment (related to Section 21.3.2) of decision-makers may be instructive to the type of experiences actors in that decision-making body have. Arguably, if decision-makers are unaccustomed to taking decisions on ethical issues, they may be more reluctant to exercise discretion in such areas.³² Furthermore, past failures in regulatory contexts may affect future decisions as decision-makers strive to ensure this does not happen again. However, as noted, this can have unwarranted effects. For example, the UK’s Alder Hey organ retention scandal in the 1990s related to the storing of organs of deceased adults/children without families’ knowledge or consent in many cases, and led to the adoption of strict legal reform and extensive guidance on organ donation/retention under the Human Tissue Act 2004. Such change was undoubtedly warranted, but placing issues of transplantation under this same Act arguably conflated the purposes and potentially could be viewed as having led to overly restrictive policies in some contexts – such as deferring to families in practice even if the deceased consented to donation.³³

21.3.4 Inter-institutional Influences

Finally, inter-institutional influences and relationships/agreements between over-arching institutions and external institutions can be highly significant on the application of rules/provisions. Sociological institutionalism highlights that institutions can affect each other, akin to a form of institutional peer pressure amounting to a potential for diffusion or homogenisation of norms/policies across institutions operating in the same area.³⁴ The urge for homogenisation may be driven by types of isomorphism (discussed earlier) at play, namely:

1. 1. Coercive isomorphism, whereby coercive forces exert pressure on decision-makers, such as legally or politically mandated requirements from external sources on that institution;³⁵

2. 2. Mimetic isomorphism, whereby in cases of uncertainty ‘when the goals are ambiguous or when the environment creates symbolic uncertainty, organizations may model themselves on other organizations’.³⁶ This may be particularly acute within the HRR context as both societal/technological changes cause uncertainty for regulators;

3. 3. Normative isomorphism whereby similar experiences/backgrounds of decision-makers lead to homogenisation given similarities in thought processes.

Understanding homogenisation across institutions is important to consider for HRR, as homogenisation may not deliver the best interests of those regulated, or public interest(s), objectively defined. Inter-institutional effects can be either constraining or predictive in nature, depending on the hierarchical legal relationships – if any – between the institutions in question. For example, a regulatory body may have legal obligations under EU law or under the European Convention on Human Rights (ECHR) system. Such influences will permeate how that body carries out its decision-making processes, particularly if it fears legal sanction for failing to abide by law. However, if the body perceives itself as having discretion in how it can apply rules to conform with EU/ECHR obligations, or if such obligations have not been strictly monitored in the past, this may limit the effect of the overarching institution’s influence.

Relationships between institutions may also be highly persuasive, such as where institutions are not in a hierarchical relationship per se, but may seek to align functions for broader political reasons. For example, in the HRR context, section 111(4) of the Care Act 2014 imposes a legal obligation on multiple regulatory authorities to co-operate with each other to achieve a co-ordination and standardisation of UK HRR practices.

21.4 Conclusion

Institutional influences act as scaffolds for decision-making, and in so doing, shape and influence outcomes. Yet, these influences are often ignored and overlooked. This chapter has argued that institutional influences require much deeper consideration within HRR and in other health contexts. Their effect is particularly acute where decision-makers have discretion on rules/provisions and can lead to engrained pre-dispositions in favour/against certain changes. Moreover, as the health context is one where social/technological change is constant, discretion and/or uncertainty on the application of provisions/rules is always evolving. Hence, we must take more seriously the effect of institutional influences on decision-making. Such influences must be actively considered and accounted for in legal/policy change in HRR and other contexts.

22 The Once and Future Role of Policy Advice for Health Regulation by Experts and Advisory Committees

Eric M. Meslin

There is nothing a government hates more than to be well informed, for it makes the process of arriving at decisions much more complicated and difficult.

—John Maynard Keynes

22.1 Introduction

In many countries, principally those with an established research infrastructure and a national commitment to science and technology policy, there is a loose ecosystem of advisory committees, experts, lobbyists and interested groups that are variously used to provide governments with expert advice and input on matters of policy. The government creates some of these structures for this purpose; others are formed unilaterally but at ‘arm’s length’ from government. Some respond to requests for input from government, others volunteer it without being asked, hoping to convince government of the value and relevance of the knowledge offered. Some, as we shall see below, are more public in their work, others are more private. Common to all is that unlike the formal legislative and regulatory apparatuses of government common to civil society, this loose collection of experts and committees functions in the liminal spaces where regulation, guidelines and policies are developed, informed and debated. This helps explain why science advice to governments is more of an ‘art’ than a science.¹

This chapter focuses on how governments make use of expertise to inform health regulation, where the expertise comes from sources connected to, but somewhat on the periphery of, the formal processes of policy development through legislation or judicial review. Two examples are drawn upon from direct experience (and are therefore somewhat subjective): (1) the use of expert panels supported by scholarly academies that are organised to provide input to government and (2) advisory committees established by government with a focus on the former US National Bioethics Advisory Commission. Both types play particular roles in the ecosystem of a country’s policy advice regime but have different features. There is a both a rich scholarly literature² and a grey literature on other structures and examples.³ The main emphasis is that expert advice in its many iterations forms part of the regulatory apparatus that governments do make use of in developing regulation and other policy. However, these are often underappreciated and therefore difficult to assess with respect to impact.

22.2 The Role of Evidence and Evidence Gathering to Inform Policy

In his first inauguration on 20 January 2009, President Barack Obama announced, ‘We will restore science to its rightful place and wield technology’s wonders to raise healthcare’s quality and lower its costs’.⁴ It was a statement as much about his predecessor George W. Bush’s lack of support for science and evidence in decision-making, as it was about the future of the Republic. Obama’s announcement was, one might say, a liminal proposition: he was looking back on almost a decade’s worth of policy decisions, including restricting stem cell science, delaying the appointment of the FDA commissioner and the White science advisor, and outright ignoring science advice when making decisions about women’s health led some critics to refer to Bush as ‘science’s worst-ever enemy’.⁵ With that in the rear-view mirror, it was easy for Obama to look to a future of promise and hope. Indeed, on 21 June 2016, the White House released ‘100 examples of Obama’s Leadership in Science, Technology, and Innovation’, the first of which was that Obama ‘elevated the quality and rigor of the science, technology, and innovation advice in the White House’.⁶

A similar anti-science assertion was made about Canada’s prime minister, Stephen Harper, who was admonished for ‘muzzling scientists’ and reducing research funding during his decade in office.⁷ In a manner reminiscent of Obama, and shortly after he was elected in 2015, Canadian Prime Minister Justin Trudeau asserted, ‘We are a government that believes in science – and a government that believes that good scientific knowledge should inform decision-making’. Trudeau took a number of actions, including appointing a Minister of Science, prioritising the appointment of a Chief Science Advisor, and filled his first federal budget⁸ with sixteen references to ‘evidence’, ‘evidence-based decision-making’, and textual support such as: the government ‘understands the central role of science in a thriving, clean economy and in providing evidence for sound policy decisions’.⁹ Subsequent federal budgets have made similar references to the value of evidence to inform decisions.

The USA and Canada examples were hardly unique. Indeed, it became de rigour for governments to embrace the value of using evidence in policy development, often referring to ‘evidence-based’ or more accurately ‘evidence-informed’ policy as a goal. The UK,¹⁰ Australia¹¹ and New Zealand¹² are three of the most visible, and who have been emphasising the value of evidence to inform policy for decades, though the support waxes and wanes depending on who is in power. Moreover, the call to use evidence is hardly new, particularly in medicine and healthcare.¹³ The justification is that data provide an objective foundation on which to develop policy, avoiding perceptions of bias, ideology or subjectivity. This approach is satisfying at many levels, though there has always been, and more recently an apparent increase, in public scepticism about the role of experts and expertise.¹⁴

The assertion of the value of facts alone may also unwittingly camouflage two other values of equal import: first, the value of recognising the epistemic foundation for beliefs about facts. As my colleague Alessandro Blasimme and I argue elsewhere, science has become especially challenging for policy-making, precisely because liberal democracies lack a coherent way to accommodate pluralistic views about scientific innovation.¹⁵ This is consistent with the view that using evidence is very different in practice than in theory. As Ian Boyd, Chief Scientific Advisor, UK Department of Environment, Food and Rural Affairs once suggested:

People don’t even think about data in the same way: When I think of data I think of binary or hexadecimal numbers. This betrays something of my background, but it was a surprise to me when in Defra, the UK Department of State with responsibility for food and the environment, we started to talk about data and I found that other people saw data very differently. Everybody had different preconceptions about data. Some seemed to be very confused. It had become trendy to talk about data, but few people appeared to think about data.¹⁶

The second type of camouflage concerns the role that ethical values play, since it should be unremarkable to claim that the test of good public policy is the degree to which it is supported by good evidence, good ethics and good epistemology. Therefore, while asking for evidence may seem sensible and pragmatic, what may be delivered depends on many factors. For instance, governments might think they want evidence given its popularity in public discourse and as found in a typical hierarchy of evidence (systematic reviews, randomised clinical trials, etc.), when what they may need is something quite different. In some instances specific answers to specific questions,¹⁷ but equally, other types of assistance including problem framing, support for a position they are intending to adopt – or oppose – or being aware of best practices by other jurisdictions. They may also ask for advice because of the perception that seeking input from elsewhere shows a degree of transparency, or provides assurance to constituents that a fair process is being undertaken to consider relevant information before a law is passed or a regulation implemented (or rescinded). This accounts for the range of instruments that governments may use, reflecting an epistemic hierarchy ranging from anecdotes, cases and stories to more organised collections of data and information, to something approaching comprehensive knowledge. I turn now to two examples of the use of experts to advise government.

22.3 Learned and Academic Societies as Experts for Government

Collections of scholars and academic experts have a long and distinguished history. Among the oldest of these academic societies in Europe are the Compangie du Gai Sçavoir, founded in 1323 by seven wealthy patrons in Toulouse whose purpose was to promote the poetry of the Occitan language; the Academia Platonica – also known as the Neoplatonic Florentine Academy – in 1462–1522; and the Barber Surgeons of Edinburgh established in 1505. In their early years, these organisations functioned as private discussion groups for their own edification and enjoyment. Yet as my colleague Summer Johnson and I describe elsewhere,¹⁸ it was not until the seventeenth century when academies of science and medicine were sought by governments for assistance in establishing public policy. Three of the most prominent were Britain’s Royal Society, established by Royal Charter in 1662; Germany’s Leopoldina established in 1652; and France’s Royal Academy of Sciences, the latter beautifully depicted in Henri Testelin’s painting of Jean-Baptiste Colbert presenting the Royal Academy to King Louis XIV at Versailles in 1667 (see Figure 22.1).

Figure 22.1. Colbert presenting the Royal Academy to King Louis XIV at Versailles in 1667 – Henri Testelin.

Reprinted with permission from the Réunion des Musées Nationaux Grand Palais

In the intervening years, and particularly in the nineteenth and twentieth centuries, learned societies emerged across the disciplinary spectrum. The American Council of Learned Societies, founded in 1919, lists seventy-five national or international ‘member societies’ in the humanities and related social sciences.¹⁹ The UK learned societies Wikipedia page lists more than 230 organisations,²⁰ Canada’s Federation for the Humanities and Social Sciences lists more than 160 universities, colleges, and scholarly associations²¹ and France counts at least thirty-six separate organisations. As organisations that honour excellence by recognising their country’s distinguished intellectuals and practitioners – usually with the title ‘Fellow’ – academies constitute a significant brain trust for any government to draw on. Increasingly, they are called upon to contribute scholarship, testify before legislatures, and offer their expert input. It is becoming common in the health research environment to seek out this type of expertise.²²

In addition to the individual activities of academies in their respective countries, there are other arrangements of these groups.²³ The first are collections of academies that are regional or global in scope, including: the InterAcademy Partnership, the Network of African Science Academies, Association of Academies and Societies of Sciences in Asia, InterAmerican Network of Academies of Sciences, the European Federation of Academies of Sciences and Humanities, European Academies Science Advisory Council, Council of Academies of Applied Sciences, Technology and Engineering and the Federation of European Academies of Medicine.

The second is the unique subgrouping of national academies organised within a country to provide specific expert input to governments. Only seven countries have such organisations of organisations: Australia, Belgium, Canada, Finland, Germany, Switzerland and the USA. I offer some observations about the American and Canadian versions.

The US National Academies of Sciences, Engineering, and Medicine (NASEM) is perhaps the most well-known and longest serving, with the National Academy of Sciences established in 1863, the National Academy of Engineering established in 1964 and the National Academy of Medicine – formerly Institute of Medicine – established in 1970. NASEM receives approximately US $200M annually from US Government contracts and US $100M from private or non-federal contracts,²⁴ undertaking about 200 different assessments, reports and projects at any given time. Numerous NASEM reports have been used to support health research regulation including early studies on primate research²⁵ and human subjects research regulations.²⁶

Modelled after NASEM, the Government of Canada responded to a proposal developed by Canada’s three main academies – the Royal Society of Canada, the Canadian Academy of Engineering, and the Canadian Academy of Health Sciences – to fund the creation of the Council of Canadian Academies (CCA) in 2005 with a mission to undertake independent assessments of evidence to provide government decision-makers, researchers and stakeholders with high-quality information required to develop informed and innovative public policy (https://cca-reports.ca/). Each of the three Canadian academies had well deserved reputations as a result their distinguished fellowship and missions. By early 2020, the CCA had completed more than fifty assessments on diverse topics in health, environment, science and technology, energy and public safety ranging. Like the US National Academies, the CCA’s assessments fill an evidence gap to support policy decision-making, including topics where regulation and legislation is ripe for review or update. The CCA’s 2019 reports on Medical Assistance in Dying and When Antibiotics Fail provide concrete examples of the gap-filling expertise governments welcome in health policy. Its 2020 release of Somatic Gene and Engineered Cell Therapies report will inform several regulatory needs: health research, innovation and disruptive technology. Importantly, the CCA expert panels take no normative positions on the subjects they assess. Rather, the CCA undertakes assessments that answer different descriptive questions, including: What is the state of knowledge of …? What is the socio-economic impact of …? What are Canada’s strengths in …? What are the best practices that exist for …?

The work is less about practical advice-giving to government, and more about evaluating available evidence. It is challenging to assess the impact of this type of work, especially on specific issues arising in health research regulation. Rarely does an evidence-focused document lead directly to a new regulation or to revision or reform of an existing one, yet there is impact. Canada’s Ministry of Innovation, Science and Economic Development undertook a comprehensive evaluation and audit of the CCA in 2018, using the NASEM, the UK Royal Society, the Australian Academy of Learned Societies and Germany’s Leopoldina as international comparators, and found that the CCA ‘addresses a need for independent, objective, and transparent scientific knowledge to support evidence-based decision-making’ and that ‘the demand for CCA assessments will continue to grow given the federal government’s priority for credible scientific knowledge to support evidence-based decision-making’.²⁷ The evaluation lists several assessments by name that have supported federal and provincial government, industry and stakeholders. Positive though these might be, such metrics can be misleading since assessment work – and evidence generally – cannot always be tracked directly to a policy outcome, which the CCA evaluators noted: ‘that there is a challenge in measuring this type of impact given that the CCA does not formulate recommendations or policy advice that could be tracked and attributed directly to its assessments’.²⁸ This same claim can be applied to another use of experts to inform government policy development: government-based advisory committees.

22.4 The Role of Bioethics Advisory Committees

As early as the eighteenth century, specialised committees were established to report on particular topics for governments. One such panel chaired by Benjamin Franklin was convened to investigate claims made by Anton Mesmer about the healing power of animal magnetism.²⁹ (The claims were rejected.)³⁰ Today, thousands of committees, working groups, royal commissions and advisory structures have been established by governments, non-governmental organisations, philanthropic bodies and industry. Until recently, in the USA alone, there were more than 1000 federal advisory committees authorised through the Federal Advisory Committee Act – until President Trump signed an Executive Order, on 14 June 2019, requiring at least one-third of them be terminated.

Unlike the CCA, these advisory bodies are intended to advise, that is, to make recommendations. One sub-category of these groups is the bioethics advisory bodies that have become a regular contributor to domestic and international debate about bioethics issues, and health research in particular. The WHO maintains a database of these groups, which currently number more than 110 around the world. Among the more influential are the standing committees such as the Nuffield Council on Bioethics and France’s National Consultative Committee on Ethics, while others are ad hoc groups such as WHO’s Expert Advisory Committee on Developing Global Standards for Governance and Oversight of Human Genome Editing. I was fortunate to have a front row seat working for President Bill Clinton’s National Bioethics Advisory Commission (NBAC) in the USA, an ad hoc advisory committee functioning between 1996 and 2001 with a focus on health research and genomics. NBAC was one in a series of such USA commissions, each of which played a key role in informing health research regulation.³¹ As luck, and the advances of science, would have it, NBAC found itself occupying some of the most intriguing liminal spaces in health science regulation in a generation. Two are highlighted here.

Following the announcement of the birth of the cloned sheep Dolly, NBAC was asked by President Clinton on 24 February 1997 to ‘undertake a thorough review of the legal and ethical issues associated with the use of this technology, and report back to be me within 90 days with recommendations on possible federal actions to prevent its abuse’.³² One hundred and three days later, on 7 June 1998, NBAC delivered its report concluding, ‘at this time it is morally unacceptable for anyone in the public or private sector, whether in research or clinical setting, to attempt to create a child using somatic cell nuclear transfer cloning’.³³ NBAC made six recommendations for public action, many of which Clinton accepted, including maintaining the moratorium on the use of federal funds for attempts to create a child. Perhaps the most significant impact of this work was the international conversation that began in earnest in other countries, especially the UK, Canada and Australia, and in international organisations including UNESCO, CIOMS, the World Medical Assembly and the International Convention on Harmonisation. Dolly’s birth moved the debate about reproductive cloning from one about a possible future technology to one in which it was now plausible to conceive of the possibility that a variety of public and professional actors would seek ways to make use of this technology.

Less than a year later, NBAC would take up a second controversial topic: embryonic stem cell research, following from the joint scientific announcements in November 1998 that human embryonic stem cells and germ cells had been cultured and derived for the first time. As with Dolly, Clinton came to NBAC to request advice, calling on the commission to ‘consider the implications of such research at your meeting next week, and to report back to me as soon as possible’.³⁴ Yet unlike the cloning report, the Clinton White House had a different reaction to NBAC’s work on stem cell science – rejecting the commission’s recommendations before they were formally submitted, the context for which I have described elsewhere.³⁵

NBAC’s experience is not unique in the world. Bioethics-by-commission is an area of scholarly study joining the emerging literature on the use of expert commissions to advise government. Some of this literature reminds us of the risks of relying on experts only, without appealing to the public.³⁶ Two examples from health research are illustrative: the influential role of patient advocates in the early debates around HIV prevention and treatment trials³⁷ and a similar story in breast cancer research.³⁸ In both cases, research regulations were amended to account for patient perspectives, and helped launch a patient engagement movement in health research that thrives today.

22.5 The Future of Advising on Health Research Regulation in a Liminal World

It is simplistic to conceive of regulations – or policy generally – as linear: that their trajectory is arrow-straight beginning with evidence and concluding with a shiny new regulation. Policy issues emerge for governments in often-unpredictable ways, requiring different types of responses. One reason is that science and society move in unpredictable ways, often responding to a recent study or an emergent problem.³⁹ For example, gene therapy research was proceeding slowly but cautiously in the 1990s until the death of Jesse Gelsinger set back research for decades.⁴⁰ But another reason, alluded to above, relates the democratisation of science and its place in the public sphere with seemingly opposing consequences. Public confidence and trust in science seem to be on the rise,⁴¹ and yet public and social media are filled with anti-science, unfounded hyped-filled assertions about medicine and health.⁴²

In the midst of these developments, a new expert is also emerging: the federal science advisor. These experts are ‘of’ government, and therefore play a different role and are exposed to different challenges. Only five years have passed since Dr Anne Glover was removed as Chief Science Advisor to the European Union in October 2014.⁴³ Predictably – and reassuringly – the reaction from the scientific community opposing the decision was swift.⁴⁴

Appointing and dismissing chief scientists and science advisors is itself a political act for governments and one can read both too much and too little into these decisions. It took months for George W. Bush to appoint his science advisor, and even longer for Donald Trump to appoint his. On the other hand, Justin Trudeau made appointing his Chief Science Advisor a key commitment of his Minister of Science’s first mandate.⁴⁵ New administrations have the right to appoint or dismiss any un-elected position. While the worst thing one can say about appointing a science advisor, or advisory commission, is that these are optically useful moves but unlikely to improve the quality of (health) regulations, the more ominous spin about decisions to remove, not appoint – or worse – to staff them with anti-science personalities, is that they cast an odious shadow on all policy advice that emerges from their office. Examples seem to abound in the USA in the Trump era, including unqualified ‘industry-captured’ scientists nominated to the Environmental Protection Agency’s Science Advisory Board,⁴⁶ or the muzzling of a government scientist’s report on climate change by the White House.⁴⁷ A more relevant health research example has been the ping-pong policy on foetal tissue and embryo research in the USA, which has been vacillating between permissive and restrictive depending on the political philosophy of the White House and the majority party in the US Congress.⁴⁸

Health research regulation is fraught with ethical, social and cultural challenges, particularly where the object of regulation involves fundamental matters of human health, against a backdrop of medical experimentation. Not surprisingly, legislation often takes time to craft wisely, and the regulations that follow may take even longer. A proposed revision to the main health research regulations in the USA, called the Common Rule, was first developed in 1981, revised in 1991, again in 2017, but not fully implemented until 2019 – despite substantive input from advisory commissions and expert panels, professional societies and the general public.

As the examples suggest, health research regulation is not undertaken by a single policy instrument, a single mechanism of reform or informed by single discipline or set of inputs. Health research regulation by its nature involves evidence, but also values, technical expertise, and stakeholder contributions. These are the liminal spaces in which developing, crafting and implementing these regulations exist. Fifteen years ago, President Clinton’s science advisor Neal Lane recognised what was needed:

The successful application of new knowledge and breakthrough technologies … will require an entirely new interdisciplinary approach to policy-making that operates in an agile problem-solving environment works effectively at the interface of science, technology business and policy, is rooted in improved understanding of people, organizations, cultures, and nations, engages the nation’s top social scientists, including policy experts, to work in collaboration with scientists and engineers from many fields.⁴⁹

Lane’s foresight was prescient. Health research is brimming with inter- and multidisciplinary approaches, which has led to commensurate commitment to interdisciplinary governance emphasising scientific integrity.⁵⁰ It is also evident in the encouraging commitment of young people to the future of the planet, reflected in their active engagement in climate issues,⁵¹ and efforts in citizen science,⁵² and science diplomacy.⁵³ The future of health research regulation will be in good hands if society is open to advice from the expertise of experts and non-experts alike.