Developing the EPA guidance of pharmacological treatment of schizophrenia – results of a Delphi process

Abstract

Background

The development of guidelines is time-consuming and cost-intensive. The heterogeneity of clinical practice, evidence, and patients’ needs is an issue across Europe. An European core guidance for a specific psychiatric disorder may help to overcome this issue. Here, we present a progress report on the European Psychiatric Association (EPA) proof-of-concept approach to develop a European consensus guidance on the pharmacological treatment of schizophrenia.

Methods

All national psychiatric associations in Europe were contacted to provide their schizophrenia guidelines. Six guidelines were rated by three experts, experienced in the development of national and international guidelines, from three different countries (Italy, Hungary, and Germany), and the German schizophrenia guideline published in 2019 was found to have the highest quality. For this proof-of-concept approach, 45 recommendations on the pharmacological treatment of schizophrenia from the German guideline were evaluated in a two-step Delphi process to determine their acceptability throughout the European continent.

Results

44 experts participated in the first round and 40 experts in the second round of the Delphi process. Agreement among the involved experts was reached for 75% of the presented recommendations from the German schizophrenia guidelines. 11 out of 45 recommendations (24.4%) did not reach this level of agreement.

Conclusions

This progress report highlights the possibility of developing a pan-European core guidance on the pharmacological treatment of schizophrenia by adapting national guidelines and reconciling their recommendations. However, several barriers in this adaptation process, such as non-agreement in recommendations with strong scientific evidence in the reconciling process, were identified and must be considered when developing the final guidance.

Introduction

Medical guidelines are systematically developed tools to assist physicians, psychologists, and other health-care professionals as well as patients and relatives in the decision-making process of a given treatment. Thus, guidelines promote the transparency of medical decisions. In that regard, guidelines evaluate and summarize the scientific evidence, help to determine the right and individual treatment for a given patient by weighting risk–benefit ratios, and are considered to improve the quality of medical treatments [1]. However, the development of guidelines is complex, cost-intensive, and needs substantial knowledge of the concept of evidence-based medicine [2, 3].

There is a substantial heterogeneity in clinical practice across European countries, which is mirrored in differences in treatment guidelines [4]. To harmonize guideline recommendations across Europe and to optimize the resources used by national approaches, the European Psychiatric Association (EPA) aims to develop a European core guidance on the pharmacological treatment of schizophrenia. If successful, this process should be extended to other treatments such as psychotherapy or psychosocial treatments and other disorders. The report is presently in a progressive state, currently based on the German evidence- and consensus-based schizophrenia guidelines. The aim of this report is to eventually create an overall European guidance for schizophrenia. This European guidance shall be adapted to European country-specific requirements and conditions by considering each county’s guideline competences intimately involving National Psychiatric Associations (NPAs).

Currently, we have 19 national treatment guidelines on schizophrenia available from 44 NPAs of the EPA. Worldwide, there are many more published with differing quality and scope (see a brief overview elsewhere: [5–7]). Every guideline has its own emphasis, target group, evidence-evaluation strategy, and presentation, but most guidelines overlap in a significant number of recommendations. This applies in particular to aspects of antipsychotic treatment. Thus, this overlap may lay the foundation for a European core guideline for an evidence-based, standardized, ethical, and cost-effective treatment of schizophrenia throughout Europe, targeting patients’ benefits. In that regard, EPA decided as a very first step to create a “Guidance paper on the pharmacological treatment of Schizophrenia” to build a consensus on how to best treat this disorder pharmacologically within their member associations. If successful, this concept could be the basis for future development of EPA core guidance publications for major mental disorders allowing an up-to-date knowledge transfer from published science into routine clinical care. This harmonized process can then be followed by further development of these core guidance documents to European or national living guidelines. Living guidelines allow for a fast update of recommendations as soon as new and relevant research becomes available [8] reducing the gap between publications and recommendations. As detailed below, we were able to identify the German evidence-and consensus-based guideline [5] as the guideline with the highest scientific quality within EPA. This guideline was used as a starting point for the development, coordination, and discussion of the planned core guideline. In this process, the NPAs of the EPA, the Global Alliance of Mental Illness Advocacy Network (GAMIAN) Europe, and the European Federation of Associations of Families of People with Mental Illness (EUFAMI) have been involved. Here, we report on the progress of this development.

Methodology

All 44 NPAs of the EPA were invited to make their respective national schizophrenia guidelines available, mounting up to 19, which were collected via email by the EPA head office. Three reminders were sent out. Reasons for the gap between 19 guidelines and 44 NPAs were, for example,the lack of availability of clearly described national guidelines or non-responses of the respective NPA. Out of those 19 guidelines, eight guidelines would have been potentially eligible as they were published no more than 5 years ago (one further could not be translated during the project period), and included pharmacological and non-pharmacological treatments of schizophrenia. The EPA president (PF) selected three schizophrenia experts (SG, IB, AH) based on their experience in developing guidelines from three different countries (Italy, Hungary, and Germany). They independently rated the methodological quality out of six of these national schizophrenia guidelines stemming from Germany, Ukraine, Finland, the UK, Slovakia, and Switzerland using the AGREE-II tool [9]. The guidelines from Norway and Croatia arrived too late to be involved in this process. Thus, only six guidelines were evaluated by the experts. Based on the AGREE-II tool, the minimum value was 1 (strongly disagree) and the maximum value was 7 (strongly agree). The schizophrenia guideline of the German Association for Psychiatry, Psychotherapy and Psychosomatics (DGPPN) [5] received the highest mean final evaluation score of 6.00 ± 1.00 points and was therefore selected to be the basis for the subsequent Delphi process. The guidelines of Ukraine (2.67 ± 0.58), Finland (3.33 ± 1.53), UK (5.00 ± 1.00), Slovakia (3.67 ± 0.58), and Switzerland (4.67 ± 1.56) reached lower rankings. For the Delphi process, a consensus group was developed consisting of schizophrenia experts of which 44 were selected from 26 NPA presidents (no more than 2 from one country) and five were nominated both from EUFAMI and GAMIAN-Europe. In the first and second round of the online Delphi process, which took place between January and April 2023, the 45 recommendations (including two statements) on pharmacotherapy or biological treatment (except catatonia and comorbidities such a sleep-disturbances or agitation) from the schizophrenia guideline of the German Association DGPPN were rated (agree vs. not agree with the recommendation). The threshold criterion for a consensus recommendation was ≥75% of agreement in the second round, which matches recommendations of the literature ranging between 70 and 80% [10]. Ethical approval for this project was obtained prior to study start from the Medical Faculty, LMU University Hospital, Munich, Germany (reg. nr. 22-0887 KB).

Results

In total, 68 experts were named by the respective NPAs out of 32 countries plus respectively two from GAMIAN and EUFAMI. In the end, 44 experts (45.5% female) participated in the first round of the Delphi survey, with a mean age of 53.16 ± 8.77 years and a mean professional experience with people with schizophrenia of 25.64 ± 10.03 years. 40 participated in the second round of the Delphi survey. Please see Table 1 for more demographic information of the sample. 34 out of 45 recommendations (75.6%) reached a level of agreement above 75% showing a good consensus across Europe on how to offer evidence-based pharmacological treatments to people with schizophrenia. This was based on scientific evidence and a rating scale between “agree,” “disagree” or “agree with changes.” 11 out of 45 recommendations (24.4%) did not reach this level of agreement. Table 2 highlights the detailed results of the final Delphi process. Though not reaching the 75% level of agreement, most of those 11 recommendations had still a substantially higher frequency of agreements compared to non-agreement. Remarkably, seven recommendations (64%) with no agreement were based on meta-analyses or randomized-controlled trials, meaning that no consensus was reached despite a high-level scientific evidence, as they did not seem to meet the clinical experience in the given country. Moreover, two of these recommendations (18%) had the highest strength of recommendation (A) in the source guideline [3, 9]. Please see Table 2 for a comprehensive description of all recommendations and the voting results of the second Delphi round.

Table 1. Participating NPAs and other associations and their representatives. N = sample size; SD = standard deviation

Table 2. Recommendation survey results (Recommendations that have not reached the 75% agreement are highlighted in bold)

Discussion

Here, we present a progress report on developing an EPA core guidance for the treatment of schizophrenia based on national guidelines. This first step should lay the foundation for further guidance publications and help to currently develop state-of-the-art tools to guide clinicians, patients, and other stakeholders in times of scarce time and financial resources. Our proof-of-concept approach focused on the pharmacological treatment of schizophrenia but will be extended to psychotherapeutic and psychosocial treatments. We were able to show the feasibility of this approach and the agreement on 75% of all recommendations on the pharmacological treatment from the German schizophrenia guideline [5, 11] showing that it is possible to scale a national guideline to other countries. However, prior to the final adoption of European core guidance, a discussion panel in addition to the Delphi processes used here is needed. This can be explained by the fact that our experts did not agree on several evidence-based recommendations that have been rooted in strong scientific evidence. This must be especially questioned for recommendations with an A-level recommendation, such as using metformin to prevent weight gain and not-to-use mood stabilizers to augment antipsychotic treatment. One should be aware that for metformin not only meta-analyses highlight possible advantages of this approach [12, 13], but that also one guideline based on the GRADE approach supports this strategy [14]. At this stage, we may speculate whether the uncertainty of evidence or uncertainties [12] in the application has resulted in the here reported discrepancies. Neurostimulation using electroconvulsive therapy (ECT) or repetitive transcranial magnetic stimulation (rTMS) did also result in non-agreement. One could speculate whether the inconclusive data regarding rTMS, the non-availability in some countries, or the general scepticism regarding ECT (e.g. due to lack of information) may explain these findings. A relevant limitation of the Delphi process stems from the fact that some recommendations on pharmacological treatment from the German schizophrenia guideline combine multiple statements. Thus, an expert might agree with one but disagree with another statement and this information is not adequately captured by the rating process. This aspect must be taken into consideration during the development of the EPA core guidance for the treatment of schizophrenia. Interestingly, 18% (2/11) of the recommendations with less than 75% agreement pertain to the treatment of negative symptoms, perhaps reflecting the currently limited options of available pharmacological treatments for this domain of schizophrenia psychopathology [15]. In the used German guidelines, especially CBT and training of social skills received high recommendation levels [5, 11]. It is important to note that during the country-specific approval process of the German guideline, all recommendations received >75% agreement. Importantly, to develop a European core guidance, we must ensure that during the nominal group process no personal opinions, conflicts of interest or special interests influence the voting results.

However, we were able to show the feasibility of such an approach. This progress report will guide the next steps including developing a full set of EPA recommendations for the treatment of schizophrenia and upon finalization and acceptance by the NPAs other guidelines may be further developed in a related manner. Thus, we plan to implement an up-to-date guidance paper in terms of overall European guidance for the treatment of schizophrenia. We plan to adapt the guidance paper to European country-specific requirements and conditions, considering each country’s guideline competences in terms of feasibility and applicability. To reach this goal, each recommendation could be reviewed by two authors who can make recommendations for updates of the text and of the supporting references as well as of the strength of evidence with a good approval process prior to submission. Moreover, the NPA boards should also have the opportunity to review and approve the planned guidance. Changes will then be discussed, revised, and approved by all authors, and presented during an online meeting of the authors. This new guidance paper will be developed in such a way that it can be transferred to a living guideline. Living guidelines have experienced an upswing during the COVID pandemic. They are an optimization of the established guideline development process by adding the option that individual recommendations can be updated as soon as relevant new evidence is available [8]. Concepts of how to develop living guidelines on a national level are available (e.g. [16]) and to take the next steps on a European level, such manuals describing the process of developing a living guideline must be adapted as well. In general, we are aware that guideline and guidance implementation remains in many cases insufficient [17–20]. Several barriers including personal factors (e.g. lack of motivation, lack of awareness, lack of knowledge), guideline-related factors (e.g. guidelines are outdated), external factors (e.g. difficulties in accessing guideline), or lack of resources (e.g. no possibility to implement a treatment due to the financial situation in the given healthcare area) have been identified in implementing guidelines [19]. This must be kept in mind when developing a pan-European EPA core guidance – especially differences between countries in the national healthcare sectors, financial opportunities, regional features, and legal basis must be acknowledged. Thus, core guidance can only be an advice with a broad consensus on the main aspects of treatment, but not a complete guideline trying to address all aspects of treatment in each healthcare setting. In summary, this progress report shows the results of a two-step Delphi process regarding the voting of predefined recommendations across Europe. This progress report lays the foundation for a pan-European core and living guidance for the management of schizophrenia, but also points out that for such a process in future a further development of the rules and regulations of how to develop such a guidance is necessary.