Use of the ADAPTE method to develop a clinical guideline for the improvement of psychoses and schizophrenia care: Example of involvement and participation of patients and family caregivers

Abstract Introduction The aim of this study is to analyse different ways of participation during the development of a clinical guideline to improve the early detection of psychosis and to deploy a comprehensive treatment plan to improve prognosis and social integration. Materials and method The clinical guideline was developed using the ADAPTE method with the participation of 40 authors and 80 external reviewers. The process was divided into three major phases: set up, adaptation and finalization. During adaptation and completion, a total of 44 patients and 18 family caregivers were involved. Results and conclusions The different roles assumed by the patients and their family caregivers were described, depending on the panel in which they participated, with diverse grades of complexity: a user as author, integration of the results of qualitative research with the participation of local users and family caregivers, 13 users as individual external reviewers and the participation of users and caregiver organizations in the external review. In the guideline, contributions from patients during the qualitative research were included in an innovative way, placing them just behind the recommendations. On the other hand, the results of the family caregivers’ study were included in a specific area of uncertainty. Further, the expressed point of view was considered as the collective demands of users and family caregivers’ organizations in the cost‐benefit analysis made by the organizing committee. There were diverse ways to conduct direct patient participation during the guideline development, ensuring that their individual experiences contributed significantly to the final version.


| INTRODUC TI ON
Psychosis refers to a set of psychiatric disorders in which an individual's perception, thoughts, mood and behaviour are significantly altered. Schizophrenia is the most common form of psychotic disorder, and because of its complexity and potential severity, it should be a major priority for mental health services. 1 It is well established that delays in the detection of psychotic disorders usually worsen both the patient's recovery and prognosis. 2 In addition, management is suboptimal in some cases, with the use of polypharmacy frequently 3 and insufficient monitoring of the physical health of these people, resulting in excessive mortality with respect to the general population. 4 These problems may be due to the complexity of the situations associated with psychotic conditions, which can make it difficult for professionals and patients to make decisions. Moreover, suboptimal care for these disorders may also be associated with the lack of implementation of the best knowledge and the existence of unwarranted variations in the provision of effective services or interventions. [5][6][7][8] Clinical guidelines (CG) are instruments designed to facilitate decision making in complex clinical conditions, as can be considered the case of psychotic disorders.
Nevertheless, consistent results are not obtained in terms of the impact of guidelines implementation on reducing antipsychotic co-prescription in schizophrenia outpatients. 9 On the other hand, nurse-led interventions seem to improve the detection of physical comorbidities. 10 Eventually, more research is needed to achieve the clinical impact of guidelines on patient outcomes, and alternative ingredients for implementation strategies are necessary as an audit of clinical activities and feedback to doctors. 9 For example, despite the existence of multiple guidelines for medicine optimization for schizophrenia, the individual's experience of using antipsychotic medication and its implications on adherence and outcomes are scarcely explored, and limited evidence is available. 11 On the other hand, the participation of patients in the elaboration of CG is an internationally recommended standard by all organizations involved in CG development. Among the reasons that have led to this orientation, it is worth highlighting the provision of more patient-centred guidelines with recommendations that are more suited to patients' values and preferences and also facilitate the implementation process. 12,13 Nevertheless, patient-reported experience has been poorly developed in people with severe mental illness. 14 In Spain, in the past 10 years, no CG have focused on psychosis and schizophrenia management. Adaptation methods allow for the development of adapted CG using the best scientific evidence and at a much lower cost than it would have to devise a de novo guide.
Alternatively, the integration of patients and their family caregivers throughout the process could provide new approaches that contribute to the improvement of guideline adoption.
The aim of this study is to present the development and elaboration of the Clinical Guideline for the Treatment of Psychosis and Schizophrenia 15 using the ADAPTE method 16 and actively promote the participation of stakeholders, including patients and family members.

| Study design
A descriptive case study is presented to develop a CG for psychosis and schizophrenia using the ADAPTE method, with the involvement of patients and family caregivers in the process (see Table 1).
Multiple methods were used throughout the process for users' and family caregivers' participation, such as those intrinsic to the ADAPTE method, together with qualitative methods.
The ADAPTE method consist of 3 phases, organized through 24 steps to adapt guidelines to a local context considering needs, priorities, legislation, policies and resources. 17 These three phases are set-up, adaptation and finalization. In the set-up phase, an organizing committee (OC) and a multi-professional panel (MP) are constituted, and the main topic is decided. The second phase of the ADAPTE process includes the determination of specific health questions; the search for existing CG; the assessment of the quality, content, consistency and applicability of existing CG; decision making about adaptation; and the preparation of the first manuscript. Finally, during the third phase, the CG is subjected to an external review, and an update system is included.

| Participants and setting
Professionals from mental health clinical practice (psychiatrists, clinical psychologists, mental health nurses), experts on methods for guideline development and implementation, and patients and their families were involved throughout the process.
In the users' group, a purposive sampling technique with the following criteria was performed: a diagnosis of psychotic disorder made by a psychiatrist or a clinical psychologist according to DMS-V 18 criteria, use of public mental health services, and no active symptoms at the beginning of the study that could interfere with participation. Subjects were excluded if they were under the age of 18 or had mental retardation. Thus, the full list of users with a diagnosis of psychotic disorder was obtained from electronic medical records. Once subjects were selected, they were contacted by phone by their mental health nurses to explain the objectives of the study and request their participation. Patients who agreed to participate in the study were afforded sufficient time before the meeting was held to allow them time to read the participant information sheet and ask any questions they considered relevant before signing the informed consent form. The sample size was subject to the principle of saturation of information during data collection and analysis.
For the family caregivers' group, sampling was aimed towards relatives who have assumed the main caring role for people under mental health services. This selection was carried out by the nursing staff in charge of caring for patients and their families. They telephoned family caregivers following a process similar to the one described above for patients.
The guideline was developed in Málaga (Spain), involving the Regional University Hospital, the University of Málaga and the Andalusian School of Public Health. Health care is provided by the public health-care system which guarantees universal coverage and free access, including mental health services.

| Approaches to engagement
The qualitative phase aimed to foster the involvement of users and family caregivers. A content analysis approach was used, by carrying out two separate groups: users and their family caregivers. This approach was framed in the steps for continuous patient engagement in CG development proposed by Armstrong et al. 12 Since this framework is not specifically intended for CG adaptation, some steps were not used, such as nominating and prioritizing guideline topics, framing the questions or developing systematic reviews. On the other hand, they were engaged in the rest of the steps proposed by this framework: the creation of an analytical framework (helping refine or expand the scope of the topics or additional factors Both in the qualitative study carried out with users and in the one carried out with family caregivers, the interviews were semi-structured and supported by a guide (Annex I). Issues identified in the literature, together with expert consultations and findings from previous studies carried out in our service 19 were considered for the construction of the interview guide. Finally, the topics included in the guide were the impact of the disorder on the users' daily life and interpersonal relationships; assessment of their relationships with the professionals who provided care to them (both from primary care and mental health) and the process of care; types of interventions offered by the public health system (including both psychological and pharmacological options) and their perceived impact; and, finally, personal resources used to cope with the disorder. The questions were open-ended, and the interview was conducted in a flexible style, so that any topic that was not initially proposed was accepted and discussed. The interviews lasted between 90 and 120 minutes and were carried out by neutral interviewers who were experts in group and individual interview techniques and were not associated with the treatment team of the study participants. Likewise, all the interviews included an observer trained for this purpose who was also outside the treatment team. The observer took note of the situation of each participant and the non-verbal aspects that could help to understand the interactions between the participants. The interviews were conducted in a location that was also different from their usual treatment settings. The interviews were audio-taped and transcribed verbatim.
Users' and family caregivers' contributions were deductively codified according to the recommendations of each uncertainty area. The procedure was similar to that described in previous studies, 20 which began with a detailed reading of categories and TA B L E 1 Approaches to users' engagement in each ADAPTE phase and outcomes in the CG

Set-up phase
One user as author in the MP subcategories identified in the patients' reports and a subsequent process of recognizing commonalities with each recommendation.
This linking process was performed by members of the MP and the OC who carried out the qualitative analysis. The initial pairings were distributed to members of the OC to contrast and contextualize these assignments, resolving discrepancies by consensus. On the other hand, the contributions of family caregivers were included in an area of uncertainty specifically aimed at interventions that improve caregivers experience.
To ensure the credibility and validity of the results, the criteria of Four of these members had experience in developing guidelines using the ADAPTE method in mental health. [23][24][25] The OC appointed an MP for the development of the guide, which included a large number of professional health-care staff (16 psychologists, 7 of whom were clinical psychologists; 13 psychiatrists; 6 mental health nurses and 5 primary care physicians). This panel included a woman diagnosed with psychosis who had studied psychology (and thus had technical and scientific training) but had not practised such a profession and participated throughout the entire process. She used to lead peer-support groups for patients of mental health services for psychosis, and her main contributions focused on the contents in which she was an expert.

| Phase 2. Adaptation
Existing  Table 2). The selected CG were reviewed and evaluated by four independent evaluators using the AGREE II tool 35 and obtained the results detailed in Table 2.
Based on the results, the NICE guideline was finally selected as the main basis for adaptation according to the AGREE score and the spectrum of intervention areas covered by this guide. However, the SIGN guide was also used as a complement for some areas of uncertainty.
To formulate the initial CG draft, the content of the guide was divided into 16 areas of uncertainty and each area was assigned to a subgroup of participants in the MP, including the users involved  For example (see Table 3   Most recommendations from the external reviewers were related to the unavailability of some equipment or services in the Spanish context, such as crisis intervention teams or mental health patient advocates. In these cases, the OC considered that it would be preferable to keep these recommendations, because, although they could not be fulfilled at the publication date of the guideline, they were sensitive to many of the perceived demands of users and family caregivers, and could serve as sources of information for patients, caregivers, professionals and decision-makers on the models of health-care organizations that have been proven useful. Thus, the CG could guide institutional decision making regarding the future organization of mental health services.

| Phase 3. Finalization
The feature of mental health patient advocate services is quite unknown in our context and raised many concerns among reviewers, specially the non-professional ones. Nevertheless, given that it has been demanded by patient and family associations for years and will be implemented in the Public Health System of Andalusia in the coming months, the OC decided to keep it as a reference.
The rest of the modifications sought to make the recommendations more understandable or correct wording errors.
Additionally, users and caregiver organizations were also invited to conduct an external review of the CG. As a result, the CG was en- User testimonials: Most users report that the treatment was prescribed without asking for their preferences and with hardly any information. However, several of them acknowledged having agreed or negotiated the treatment with their therapist, until they found the one that 'best suited their situation'.
• They didn't ask me anything. They told me directly 'take this'. And they started with a small dose, and then they went up because it didn't work for me, until they got the dose that suits me. But they didn't tell me 'there is this alternative or the other'. He said to me 'take this' (female, 55 years old). • They said 'you are taking this treatment and it seems to be going well, but it could be even better. Recommendation: Treatment with antipsychotic medication should be considered an explicit individual therapeutic trial. Include the following: • Discuss and record the side-effects that the person is most willing to tolerate. • Record the indications and expected benefits and risks of oral antipsychotic medication, and the expected time for a change in symptoms and appearance of side-effects. • At the start of treatment give a dose at the lower end of the licenced range and slowly titrate upwards within the dose range given in the drug data sheet. • Justify and record reasons for dosages outside the range in the drug data sheet. • Record the rationale for continuing, changing or stopping medication, and the effects of such changes. • Carry out a trial of the medication at optimum dosage for 4-6 weeks.

TA B L E 3 (Continued)
Andalusian Health Service (http://www.junta deand alucia.es/servi cioan daluz desal ud/publi cacio nes/lista dodet alle.asp?idp=715). All participants reported that the experience was highly positive and they were actively involved throughout the full period. This was remarkable since some previous studies on patients' participation in mental health decision making have reported difficulties among these patients in being perceived as a 'competent and equal person'. 43 The supportive attitude among OC members facilitated participation, as was previously reported by similar processes on mental health guideline development. 44 A commonly reported barrier to users' involvement in health-care improvement processes is the difficulty in their implementation. 45 Similarly, in our study, challenges were found in the coordination of external reviewers because of the extensive number of components.

| D ISCUSS I ON
Likewise, during the qualitative phase with patients and their families, it was not easy to reconcile the schedules and availability of many participants, especially during the focus groups. Other user involvement barriers identified in previous studies included the lack of financial support for users and the considerable amount of time required to participate. 46,47 However, in our case, users with well-valuated user-clinician relationships did not show inconvenience in involvement, as reported before. 45 Nevertheless, this might pose a risk of self-selection bias. Some studies have reported how poverty and stigma emerge as barriers to users' involvement, 48 which cannot be ruled out in our study.
This study has several limitations. First, the adaptation process was carried out for a specific context (the Spanish health-care system), and some findings would not be generalizable to other contexts. Likewise, the CG was based on the NICE CG, which could be a limitation although it was selected via a formal review and quality assessment process. Some authors have reported an evolution in focus from identifying source guidelines for adaptation to identifying specific recommendations for adaptation. 41 Users did not participate in the selection of uncertainty as proposed by Armstrong et al 12 Additionally, we found that users with good user-clinician relationships participated more than other patients as reported in previous studies. 45 We did not perform a formal evaluation of satisfaction and the resources employed. Eventually, the final impact of patients' involvement on the improvement of services for people with psychosis and schizophrenia would require a long-term sustainability assessment.

| CON CLUS ION
The use of adaptation methods has allowed for the development of a CG for people with psychosis and schizophrenia adapted to the Spanish health context, which for more than a decade has not had any resource for decision making in this field of mental health clinical practice. The ADAPTE toolkit is an invaluable resource that guarantees the systematization of the entire process. On the other hand, the comprehensive participation of patients and their families has helped incorporate their values and preferences not only in the writing of the recommendations but also in their contextualization to the local setting and the contribution of new perspectives for decision making.

ACK N OWLED G EM ENTS
The authors are grateful to all the people who have contributed to the elaboration of the Clinical Practice Guide, either as authors or as reviewers, their time and dedication. We also thank the National Institute for Health and Care Excellence of the United Kingdom for granting the adaptation and implementation licence using its guide 20 as the main source.

CO N FLI C T O F I NTE R E S T
The authors have no conflicts of interest to declare.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data sharing is not applicable to this article as no new data were created or analysed in this study.