The concept of service model fidelity in Talking Therapies

Abstract

Abstract

The concept of Service Model Fidelity is considered as a parallel process to Treatment Fidelity in evidence-based psychological therapies. NHS Talking Therapies (formerly IAPT) aimed to increase access to an expanded, upskilled workforce on a national scale. This included systematic training, supervision and front-line service delivery, emphasising treatment fidelity to evidence-based interventions. A further feature of NHS Talking Therapies was modernising and restructuring of the health system that housed these trained practitioners. The term ‘service model fidelity’ (Cromarty, 2016) was coined to emphasise service modernisation aspects as a distinct entity. A definition of the latter is included. Examples of service model fidelity and of service model drift, are outlined to distinguish these from therapist drift. This links to past literature recommending changes in traditional mental health service design and emergent evidence from NHS Talking Therapies. The latter examines publicly available data identifying characteristics of service design, which appear to be predictors of enhanced clinical outcome. Challenges in modernising health systems are discussed and conclusions are made highlighting the crucial role of service model when delivering evidence-based therapies. Suggestions for further research into service configuration to improve experiences of service users are considered. This includes ongoing exploration of service design being more than a qualitative feature, and increasingly appearing as a key factor in enhanced clinical outcome.

Key learning aims

1. (1) To identify service model fidelity as separate entity to treatment fidelity.

2. (2) To provide a clear definition of service model fidelity.

3. (3) To delineate therapist drift from service drift.

4. (4) To further examine the role of service model in delivering evidence-based interventions.

Introduction

This paper suggests that the concept of ‘service fidelity’ is considered alongside the importance of ‘treatment fidelity’ in evidence-based psychological therapies. Improving Access to Psychological Therapies (IAPT) piloted in 2006, expanded nationally in 2009 to re-shape UK psychological services (Clark et al., 2009). IAPT was a response to the depression report (Layard, 2006) calculating 10,000 therapists needed training to provide National Institute for Health and Clinical Excellence (NICE) interventions. IAPT operated stepped-care mental health (Bower and Gilbody 2005) introducing Low Intensity in front of traditional CBT treatments, now termed High Intensity. Several years later, successful outcomes of IAPT continued to be reported (Clark, 2018; Saunders et al., 2020). IAPT was rebranded as NHS Talking Therapies (NHS England, 2023) with a regularly updated Manual (NHS England, 2024) outlining key features of the model, and notes clinical outcomes achieved nationally, broadly correspond with those of clinical trials. It states services operate under three principles: (1) evidence-based psychological therapies at the appropriate dose; (2) appropriately trained and supervised workforce; and (3) routine outcome monitoring on a session-by-session basis:

While increasing available training, supervision and upskilling a workforce in evidence-based interventions remains undisputed, it is argued this must occur in the context of developing a flexible and responsive health system to deliver them. This involves emphasising two broad, yet inter-related, themes operating in NHS Talking Therapies:

1. (1) Clinical improvement – increasing skills and access to evidence-based interventions;

2. (2) Service improvement – modernising health service structures to optimise the above.

Attention to service improvement has increased via emergent studies reporting service design as a factor in predictors of recovery (Gyani et al., 2013; Clark et al., 2018). The latter concluded that service configuration could be as important as providing effective treatments. The term ‘service model fidelity’ (Cromarty, 2016) was coined to emphasise changes to service delivery and structure. The present paper includes a full definition of the term, highlighting service modernisation in NHS Talking Therapies, as a distinct entity. This describes features of service design, for the benefit of practitioners and for those that may not have a clinical background. These concepts would benefit managers, clinical governance officers, planners and policy makers in the UK and other countries, in emphasising importance of a robust service model, containing effective psychological therapies, to deliver more challenging KPIs than traditional mental health systems.

A clinical improvement rationale for NHS Talking Therapies has a basis in the tension between outcome research in CBT and its traditional failure to translate into standard clinical services. Findings indicate that services with more variables that increase treatment fidelity to known interventions, will be linked to superior clinical outcome (Freeston, 2008; Gyani et al., 2013; Shafran et al., 2009). This was described as adhering to a ‘high dose narrow bandwidth’ model (Cromarty, 2016) where increasing dosage of the primary intervention allows practitioners in NHS services to closer match clinical outcomes achieved in research trials; and to combat therapeutic drift (Waller and Turner 2016). To optimise this, a service improvement rationale for the system that contains well-trained and supervised therapists and access to it, may be equally important. Aspects of this have been referenced in pre-IAPT literature (Bower and Gilbody 2005; Clark et al., 2009; Layard 2006; Lovell and Richards, 2000) and post-IAPT (Clark et al., 2018; Gyani et al., 2013). Fidelity to known evidence counters therapeutic drift among practitioners (Waller and Turner, 2016). If the service model is not managed with similar fidelity, ‘service model drift’ may impair overall effectiveness and impact on clinical recovery rates.

Service model fidelity: definition

The need of a service to be managed to the extent it can function optimally, to meet the aim of enacting NICE interventions and all associated KPIs, as intended. Only taking on new remits, disorders, treatments or structural changes, in response to NICE or other specific Talking Therapies Guidance; including any changes to training, supervisor requirements and service configuration.

The role of service model fidelity

Service model fidelity means ensuring that service-level changes made to benefit users, are in keeping with clinical evidence, not purely based on financial, workforce or management needs. Maintaining flow of stepped-care between Low and High Intensity is a key aspect of service model fidelity. The definition incorporates formal guidelines and proposed changes stemming from therapists’ preferences or service-user feedback. This must cross-reference with evidence and not be detrimental to clinical outcome or safety to ensure creativity and innovation remain within treatment and service fidelity parameters.

Examples of service model fidelity are managers ensuring new supervisors undergo supervisor training or services responding to recommendations following new evidence. A specific example would be services implementing guidance and organising additional training in adapting CBT for older people (Chellingsworth et al., 2016) before placing practitioners in aged care facilities. Such opportunities for training that ensure service fidelity may provide additional workforce benefits.

Ensuring service model fidelity amplifies strengths of inexorably linked variables, including combined effects of training, supervision and treatment fidelity. In terms of empirical evidence, these are ‘under-powered’ in isolation, but situated within a robust and flexible service model, they may generate a critical mass of outcomes they cannot achieve alone. In controlled research, treatment fidelity is an independent variable (Robb et al., 2011). When applying high dose narrow bandwidth principles, linked to training and clinical supervision, within a flexible and robust service model, treatment fidelity arguably becomes the dependent variable.

The rationale for service model fidelity

Instead of simply increasing CBT training places in the existing system, IAPT introduced structural changes for practitioners to operate within, contributing to widened access and improved outcomes. The evidence base for CBT was well-established prior to introduction of NICE Guidelines (circa 2000). This poses the question, where were these trained practitioners in the health system before NHS Talking Therapies? A survey by Townend et al. (2002) verifies widely established CBT training courses, trained practitioners, emergent supervisor training and therapist accreditation, years prior to IAPT and NICE guidelines.

Transparent clinical outcome data introduced by IAPT must be viewed in the context of traditional mental health services, where CBT practitioners were previously situated (Bickman, 1999; Lillienfield et al., 2013; Lovell and Richards 2000). Previously, CBT units were commissioned on waiting-times and throughput, not rigorous performance indicators of functional recovery, loss of diagnosis and return to employment (Cromarty, 2016). Those with trained CBT practitioners typically had lengthy waitlists, with inflexible structures and referral pathways (Lovell and Richards, 2000). In terms of challenges in capacity building, Cromarty et al. (2008) reported that scaffolding new clinical systems had to go beyond simply increasing the number of trained therapists. A core of experienced trainers and supervisors is required at the outset. Newly trained therapists needed to consolidate practice and receive ongoing clinical supervision. Over time, they would be trained as clinical supervisors and contribute to supervision and training of further cohorts, allowing service expansion.

The transformation agenda of NHS Talking Therapies achieved this expansion on a national scale. It formalised and streamlined pre-existing training, qualifications and practice, on top of significantly increasing the number of new therapists. What it introduced nationally that did not previously exist, was service modernisation. This involved reconfiguration of psychological services into a stepped-care system (Bower and Gilbody, 2005), improving entry points and emphasising clinical recovery rates over throughput. Lovell and Richards (2000) noted treatments for community-based interventions remained ‘rooted’ in traditional, secondary-care models. They proposed ‘multiple access points and levels of entry’ (MAPLE). This advocated service re-design, stepped care, Low Intensity interventions and captured features of service model fidelity years ahead of IAPT roll-out. MAPLE pointed out that psychological services, instead of being community facing, with features such as rapid and self-referral, were rigidly organised with a single point of entry. The author, for example, was highly trained in treatment of post-traumatic stress disorder, in a service with an 18-month waitlist. The waitlist ranged from mild to complex and chronic cases, all with the same point of access. NHS Talking Therapies improved this by introducing a Low Intensity step among other changes. Lovell and Richards (2000) raised the point that modernising and restructuring services could optimise effectiveness of the skilled workforce. Gyani et al. (2013) and Clark et al. (2018) provide initial evidence that it does, by reporting publicly available data from NHS Talking Therapies services.

Gyani et al. (2013) report reliable recovery rates in year 1 of IAPT, achieved by 19,395 service-users from 32 sites. This identifies predictors of recovery at three levels: user level, service level, and functional compliance with NICE guidance. It noticed the high number of completed datasets for a routine cohort study, with outcomes for over 91% of people treated. This immediately highlights the benefit of IAPT services recording session-by-session routine outcome measures. It allows data capture in services, previously associated with controlled research, which traditional mental health systems did not achieve. Overall, 63.7% of cases showed reliable improvement. Most service-users received NICE-recommended treatments; and when a non-NICE treatment was delivered, recovery rates were reduced. Predictors of higher reliable recovery were: high average number of sessions, higher flow of step-up rates from low to high within the service, larger services, and a higher proportion of experienced staff. It concludes that fidelity to the IAPT clinical model is associated with optimised rates of reliable recovery. Considerable variability between IAPT services is reported but comparisons with previous mental health systems cannot be made as these did not yield transparent clinical outcome rates.

Clark et al. (2018) identified predictors of variability in clinical performance based on five organisational features. These were: (1) percentage of cases with a problem descriptor; (2) average number of sessions; (3) waitlist time to commence treatment; (4) the percentage of appointments missed was associated with more negative clinical outcome; and (5) percentage of referrals treated was positively associated with outcome. The study concludes that improvements in service implementation may be as important as the effective treatments they contain. Both studies note that operating stepped-care principles (Bower and Gilbody, 2005) such as reduced wait lists and flow of step-ups are predictors of improved outcome. Low and High Intensity services need to adhere to this to avoid service model drift.

Service model drift

Practitioners not stepping-up unsuitable cases, skipping recording of session-by-session outcome measures or introducing a therapy that is not NICE recommended, would constitute examples of therapeutic drift. Service managers directing practitioners to enact the above, is not therapeutic drift. This would constitute ‘service model drift’ by not managing services as intended, which is to utilise skilled staff and optimise clinical outcomes, based on best available evidence. Service drift can occur at both Low and High Intensity steps. A clear example would be funding staff to do training in a therapy that was not NICE-recommended. Specific Low Intensity examples are management decisions that interrupt or circumvent the flow of clinical case management processes. Examples would be introducing blanket policies to step-up cases with high scores on measures or the slightest element of suicidal ideation, contrary to clinical guidance or supervisor decisions to manage this within the system on an individual basis.

Further examples of service drift could include any changes due to waitlist pressures or financial constraints, e.g. employing supervisors not trained in CBT to supervise practitioners that are; or reducing NICE-recommended session numbers to ease waitlist pressure. Counter-analogies to these examples ask - should people wield scalpels without training in them? Should cancer services ration chemotherapy to three sessions per-person to reduce waitlists? These provocative questions highlight the role of service model fidelity, as managing mental health services involves high ethical standards and obligations once a clear evidence-base exists.

Discussion and conclusions

Like clinical recovery rates, systemic changes require ongoing maintenance and monitoring as health systems are live and dynamic entities. Service model fidelity needs to be self-correcting, allowing frequent monitoring of key performance indicators and adapting to new evidence. This systematic approach to jointly maintain service fidelity and treatment fidelity is integral to enhancing quality of interventions, client’s safety, and experiences within the system. This monitoring is critically important with a continually expanding evidence-base, where approaches such as low intensity CBT may evolve over time. For example, new guidance on low intensity formulation (Cromarty and Gallagher, 2023) must ensure this does not constitute drift into high intensity and disrupt flow of step ups. In high intensity, pressure to treat cases outside of service remit could result from individual clinician or supervisor decisions or be made locally as a managment directive. Making a service model fidelity rationale explicit among managers, practitioners and service users could increase overall understanding, engagement and satisfaction. Emphasising the science and ethics behind it, in management guidance, clinical training, continued professional development and service-user information could enhance this; and better facilitate ongoing management.

To manage ongoing service model fidelity alongside treatment fidelity, communication and feedback mechanisms within services require regular collaboration between key stakeholders, e.g. commissioners, practitioners, service-users, managers, and clinical leads. This task should be informed by the definition of service model fidelity and any new findings it incorporates. When evidence emerges (Clark et al., 2018; Gyani et al., 2013) on service design enhancing clinical outcomes, these findings must be adpoted. This would include stakeholders monitoring the following: predictors of recovery by auditing cases for a problem descriptor and compliance with NICE-recommended treatments; ensuring session-by-session measures are recorded including follow-up appointments; monitoring flow of step-ups in the system; monitoring waitlist times and missed appointments; management of supervisor and accreditation standards; and ensuring any structural changes are implemented based on closer adherence to existing or new guidance. The collaborative panel would reject changes that fall outside of this guidance.

Services that provide evidence-based interventions with flexible delivery methods benefit from being commissioned on clinical outcome; those with little or no evidence of outcome do not. The continued success of NHS Talking Therapies may be partly attributed to national service re-design, proposed by Lovell and Richards (2000) and enacted following specific proposals and piloting (Clark et al., 2009; Layard, 2006). Wholescale implementation and changes in commissioning were managed through a governmental organisation, the UK Department of Health, largely by-passing professional bodies. Accreditation was managed by a multi-professional therapy organisation, the British Association of Behavioural and Cognitive Psychotherapies (BABCP), not a single health professional body. When this has been successfully replicated outside of the UK (Cromarty et al., 2016; Knapstad et al., 2018) it has been in the form of pilot studies without equivalent training and accreditation standards. This also occurred in more traditional health systems, containing more private and insurance-based healthcare, which have led to recommendations but not nation-wide service transformation and recommissioning.

Cromarty (2016) outlined challenges in changing health systems, as proposed by previous innovations (Bower and Gilbody 2005; Clark et al., 2009; Layard, 2006; Lovell and Richards, 2000). This noted that the evidence base for behavioural and cognitive therapies has been in existence for decades, therefore, it cannot be disputed that proven interventions already exist. If these fail to translate into front-line services, the system and implementation plan is more likely responsible. It cautions that, ‘Advances requiring wholescale change, can pose an initial threat to traditional services and professions. Innovations can ultimately fail to have an impact in wider health settings if they are not resourced, implemented and managed successfully’. Advances requiring systemic change, instead of being welcomed, may initially encounter resistance from existing institutions and professions. Resistance to change may still occur even when changes work, are evidence-based, economically viable and benefit service users. It concludes by stating, ‘If the system fails, don’t blame the evidence base!’.

In summary, high dose narrow bandwidth interventions must be supported by service model fidelity, as the delivery system in which they operate is equally important. Treatment fidelity and clinical improvement can be optimised when integrated with an effective service-re-design model. Combining clinical improvement and service model fidelity combats therapeutic drift and ‘service model drift’.

The established evidence base, thorough training and clinical supervision that existed pre-IAPT were insufficient in optimising treatment fidelity in CBT services. To best enact the evidence base, attention must be paid to the system in which trained practitioners are housed. Practitioners rigorously applying a high dose narrow bandwidth model, would likely not achieve the same outcomes in a traditional system, due to its rigid design. The stepped-care structure of NHS Talking Therapies is a clear system-wide improvement. Service fidelity to this enhances user experience, with provision of evidence-based interventions, improved access, and responsiveness within the system. Clearer guidance, based on the service model fidelity definition, including feedback from practitioners and service-users, can ensure proposed changes remain within the model and exclude any that constitute service drift.

Further research into NHS Talking Therapies needs to build on findings from Gyani et al. (2013) and Clark et al. (2018) examining specific features of improved service design. This can explore increasing evidence that service model fidelity is a factor in enhanced clinical outcome of NHS Talking Therapies and not a purely qualitative feature.