Users’ choice and change of allocated primary mental health professional in community-based mental health services: A scoping review

Background: The recovery model in mental health care emphasizes users’ right to be involved in key decisions of their care, including choice of one’s primary mental health professional (PMHP). Aims: The aim of this article was to provide a scoping review of the literature on the topic of users’ choice, request of change and preferences for the PMHP in community mental health services. Method: A search of the PubMed, Cochrane Library, Web of Science and PsycINFO for papers in English was performed. Additional relevant research articles were identified through the authors’ personal bibliography. Results: A total of 2,774 articles were screened and 38 papers were finally included. Four main aspects emerged: (1) the importance, for users, to be involved in the choice of their PMHP; (2) the importance, for users, of the continuity of care in the relationship with their PMHP; (3) factors of the user/PMHP dyad influencing users’ preferences; and (4) the effect of choice on the treatment outcomes. Conclusion: While it is generally agreed that it is important to consider users’ preferences in choosing or requesting to change their PMHP, little research on this topic is available. PMHPs’ and other stakeholders’ views should also be explored in order to discuss ethical and practical issues.


Service users' choice in light of the recovery paradigm
The recovery model in mental health, developed from the study of subjective experiences of the illness and healing process of persons with mental health problems, has gained wide recognition in mental health policies and practice (Anthony, 2007). This model focuses on the process of care, promoting service users' right to co-produce and choose with carers key decisions of their care (Slade et al., 2014). Increasing choice is expected to create better alignment between what service users want and what services subsequently provide (Aylott et al., 2019;Piat et al., 2019;Samele et al., 2007). Service users' right to choose or be involved in the choice of their primary mental health professional (PMHP) may be another relevant aspect.

The choice of their PMHP: ethical framework
The right for the users of community mental health services to choose their PMHP is in line with the principle of the respect of Autonomy, one of the four fundamental principles of biomedical ethics (Beuachamp & Childress, 1979. This ethical foundation of the freedom of choosing one's PMHP is also reflected in deontological professional codes. For example, article 27 of the Italian Code of Deontological Medical Ethics, entitled 'Free choice of the physician and of the place where to receive treatment' (Title III, Chapter 2, Art. 27), states, The free choice of the physician and of the place where to receive treatment is the basis of the physician-patient relationship. In the professional activity, both in public and in private settings, the free choice of the physician is a fundamental right of the citizen. Any agreement between physicians aimed at limiting citizens' right to free choice is forbidden. (Italian Order of physicians, surgeons and dentists, 2014) Users' choice of PMHP: national health policies While the right to choose one's primary care physician, subject to availability and to specific geographical boundaries, is nowadays a widespread and accepted practice in many Western countries and several studies have investigated its implications (Lagarde, Erens & Mays, 2015;Mays et al., 2014;Robertson et al., 2008;Tan et al., 2015), this does not appear to be a reality in mental health.
In the United Kingdom, the document 'Creating a Patient-Led National Health Service (NHS)' (Department of Health & NHS, 2005) stated that the strategic aim of the NHS is to promote patient-centred pathways to care and services not only in primary care, but also in mental health settings. Therefore, in 2012, the right for NHS patients to choose their mental health care provider for outpatient treatment was affirmed (Department of Health, 2012). However, several factors prevented the system from working as it was intended, such as lack of information and awareness about this right, of the principle of patient choice for outpatient treatment, misuse of care pathways, lack of direct access by many primary care physicians for out-ofarea referrals and delays in authorization for funding. So, parity of care between physical and mental health remains problematic and not working in practice (Veale, 2018).
The freedom of choice in health care has become an important topic also in Australia (Peterson et al., 2014), New Zealand, the United States and Canada, where the mental health choice agenda focused on promoting a wider and more informed choice (Warner et al., 2006). In the United States, the list of the 10 'Rules for quality mental health services in New York State' (Infusing Recovery-Based Principles Into Mental Health Services, 2004), commissioned by the New York State Office of Mental Health, mentions as the first rule that 'There must be no uninformed choice'. Despite this, true choice is limited by the range of available services and the complexity and lack of coordination between different agencies (statutory, voluntary and private; Samele et al., 2007).
In Scandinavia, and Sweden in particular, the 'New Public Managements' (NPM) programme promoted reforms which, starting from primary care, have encouraged the exercise of patient choice (Glenngård et al., 2005), in line with the concept of 'responsiveness' (Johansson & Eklund, 2003). The organization of public welfare services was transformed into quasi-markets, with patients no longer strictly referred to their own district's services as before. However, recent published data show that such implementations do not necessarily develop in the intended directions, and authors call for more global research on this widespread phenomenon (Fjellfeldt & Markström, 2019).
In Italy, Law 833 (1978) and its following implementations protect the right of choosing and changing one's primary care physician and paediatrician, according to the principle of interpersonal trust. Despite a very long tradition of community-based mental health services (Fioritti & Amaddeo, 2014), where the PMHP is still the key figure to coordinate the contribution to care by different professionals, service users cannot generally choose their PMHP (Barbato et al., 2014). Only anecdotal reports exist that some Italian mental health centres have locally implemented operative instructions guiding how to manage users' request for choice and/or change of their PMHP (e.g. Department of Bologna, Italy).

Users' choice of their PMHP: views of users and associations of users
Mental health service users have asserted their right to choose a provider that best suits their individual needs and preferences, and user organizations are vocal on these issues as part of the drive to achieve parity between mental health and physical health. According to the UK National OCD Charity website, for example, having a 'Right to Choose' the PMHP can be helpful for reasons such as that the user may wish to access treatment closer to work or another location or that the user may wish to access treatment at a neighbouring service provider that has a better track record of treating the specific disorders the patient suffers from or shorter waiting times (The National OCD Charity, n.d.). Moreover, several free online platforms are available to provide information to patients about their rights and on the pathways to choose their PMHP (NHS Choices, n.d.; mental health charities like, for example, Mind, n.d.; Rethink, n.d.).

Aim of the article
The aim of this article was to provide a scoping review of the scientific literature on the area of the choice and request of change of the allocated PMHP by users of community mental health services.

Method
A scoping review of the literature was undertaken according to the framework outlined by Arksey and O'Malley (2005), searching the question, 'What is known from the existing research about users' choice, request of change of, and preferences for the allocated PMHP (generally, a psychiatrist) in community mental health services?'. The review included the following key phases: (1) identifying the research question, (2) identifying relevant papers, (3) study selection, (4) charting the data and (5) summarizing and reporting the results (detailed review protocol available on request). As indicated in the website homepage (https://www.crd.york. ac.uk/prospero/#guidancenotes), PROSPERO does not currently accept registrations for scoping reviews, it was therefore unable to accept our application or provide a registration number.

Search strategy and data sources
An effective combination of search terms breaking down the review question into 'concepts' was constructed. For each of the elements used, possible alternative terms were considered. Since community mental health care includes various services with different names, our review adopted the search strategy described by Bonavigo and colleagues (Bonavigo et al., 2016) to assure that a range of service settings were represented.
PubMed, Cochrane Library, PsycINFO and Web of Science databases were searched on 28 December 2018 for papers published in English with the following keywords: (choice OR change OR refusal OR preference OR matching OR concordance) AND (PMHP OR therapist OR case manager OR psychiatrist OR mental health care provider) AND (mental health* OR mental ill* OR mental dis* OR psychiatr* dis* OR psychiatr* illness OR schiz* OR psycho* OR anx* OR depress* OR bipol*) AND (community mental health service OR community mental health centres OR community day service OR outpatients day service OR home help* OR early intervent* psycho* OR supported hous* OR supported home* OR supported accommodation* OR supported living OR sheltered hous* OR sheltered home* OR sheltered accommodation* OR sheltered living OR assisted hous* OR assisted home* OR assisted accommodation* OR assisted living OR visiting support OR visiting outreach OR outreach OR housing project OR community rehabilitation service) Terms were identified by searching titles, abstracts, keywords, medical subject headings and mapping terms to subject headings. Additional relevant research articles were identified through authors' personal bibliography. Reference lists from the relevant papers were also screened.

Inclusion criteria
Qualitative and quantitative empirical papers and opinion papers were included in the review. The search was restricted to articles published in English and referring to adults with mental disorders. Studies were only included if the setting was related to community mental health services. There were no restrictions on publication status or publication date. We also included studies focused on factors influencing patients' preferences on the investigated topics, since papers exclusively addressing patients' opinions and experience were few.

Exclusion criteria
We excluded papers if they did not clearly focus on our topic of interest in an explicit way (off-topic) and if they referred to minors. Papers on choosing and requesting to change PMHP which were not based in adult community mental health settings and did not mention users' preferences were excluded as well. Finally, reviews, book chapters and editorials, and papers focused on pharmacological treatments were also excluded.

Study selection
Papers were retrieved and included according to PRISMA statement recommendations (Moher et al., 2009). Duplicates were removed and titles were made available, abstracts were initially screened for inclusion by three authors (A.M., G.R., R.V.) independently and disagreements were resolved by consensus with a fourth reviewer (G.M.G.). In cases when a definite decision could not be made based on the title and/or abstract alone, the full paper was obtained for detailed assessment against the inclusion criteria. For each selected paper, three authors (A.M., G.R., R.V.) screened the full text, extracted and summarized data.

Charting the data
Data extraction was performed for the following study characteristics: year of publication, first author, journal, study design, sample size and population, findings, outcomes of interest about choice (how the concept of choice of the PMHP had been defined, understood or interpreted within different community-based settings). In a modified two-step narrative synthesis approach, we identified all instances where choice/change of mental health PMHP was used across the included studies and integrated them into a conceptual framework.

Collating, summarizing and reporting results
Extracted data from the reviewed studies were reported in tabular material, available as supplementary material. The software used for the data collection was Excel (Microsoft Corporation). The synthesis of the findings was performed according to a thematic analysis method, through the identification of important or recurrent themes. The findings were summarized under thematic headings in the 'Results' section. Mendeley bibliographic software (Mendeley Desktop, Version 1.19.3, © 2008-2018) was used to record and manage the references. A meta-analysis was not conducted due to the diversity of populations, study designs and measured outcomes.

Results
The initial bibliographic search yielded 2,774 records, which were reduced to titles and abstracts to be further screened. Of these, 2,683 were excluded because they were off-topic, 21 because of concerning minors, 13 were papers about drugs and pharmacology, 10 were reviews or comments or book chapters and 9 were duplicated records. Therefore, 35 full-text papers were eligible. A further nine off-topic studies were excluded. An additional 12 relevant research articles were included and identified through the authors' personal bibliography. Finally, 38 papers were included in the review. Figure 1 shows the flow of article selection.
What follows is a narrative summary of the findings derived from the included studies, reported under four main headings: (1) the importance for users to be involved in the choice of PMHP, (2) the importance for users of the continuity of care, (3) the factors of the users/PMHP dyad influencing users' preferences and (4) the effect of choice on the treatment outcomes.

The importance of the choice of the PMHP
Several studies have shown that mental health service users would like to have greater freedom of care choice. Hill and Laugharne (2006) found that service users rate as very relevant to be informed about their condition and treatments; 31% also stated that they would like to express their preference about their own PMHP in public services. A similar result was found among 368 subjects with chronic depression (Groenewoud et al., 2015).

The change of the PMHP: the importance of continuity of care
Service users generally seem to value the experience of continuity of care and stability in the relationship with their PMHP. A study of 323 patients just discharged from psychiatric hospital following compulsory treatment for severe psychosis showed that more frequent changes in the key mental health professional were associated with longer hospital stays (Puntis et al., 2016). Accordingly, a qualitative study of 10 service users of mental health centres showed that changes in the allocated PMHP were experienced as setbacks in treatment, giving rise to negative feelings (Biringer et al., 2017). Even in team-based mental health services, the continuity of the individual relationship with one case manager seemed to play an important role for users' comfort level; users often expressed their preference for working with a particular case manager over others (Stanhope & Matejkowski, 2010).

User preferences in the user-PMHP dyad
Ethnic and language concordance in the user/PMHP dyad. Self-reports and questionnaires by 371 users of an Italian Community Mental Health service showed that users with the least education and those who spoke only Italian had a strong preference for a therapist of their own nationality (Lorefice & Borus, 1984). In line with this result, Herman and colleagues, who surveyed 604 users of a low-income Hispanic population, found that the language and cultural awareness of mental health workers was one of the main factors influencing user choices (Herman et al., 2016).
Blank and colleagues examined ethnic matching between mental health professional and user for 677 seriously mentally ill African American and Caucasian patients of a rural community mental health centre in the United States. In general, same-ethnic group dyads tended to have greater service utilization. African American users that matched with a therapist of the same ethnicity were more likely to fail appointments; conversely, Caucasian consumers in dyads with a therapist of their own ethnicity were less likely to fail visits (Blank et al., 1994).
Moreover, many minority group users expressed preference for minority PMHPs. For example, Tien and colleagues investigated mental health users' preference among 15 male and 15 female Black users in a community mental health centre in Los Angeles: 60% preferred Black professionals; the major reasons for preferences were the perceived professional competence (98%) and attitudes (97%), not just the cultural, ethnic and linguistic compatibility (Tien & Johnson, 1985). The majority (79%) of a sample of 218 American Indians indicated they would prefer a Native provider (Aronson et al., 2017). A study among 26.943 people explored the effect of ethnic matching in minorities serving mental health centre programmes and showed that ethnic matching seems to be linked to fewer emergency service visits, especially in case of concurrent language matching (Snowden et al., 1995). Knipsheer and colleagues, interviewing 96 Surinamese outpatients in a Dutch community-based mental health service, found that ethnic matching was rated as relevant by users and was a strong predictor of satisfaction with the service (Knipsheer & Kleber, 2004a). Conversely, a study by the same authors on Mediterranean migrants (82 Turkish and 58 Moroccan outpatients) in Dutch mental health services showed that most users did not value ethnic matching as important, and that clinical competence and compassion were considered to be more relevant (Knipsheer & Kleber, 2004b).
Finally, Ziguras and colleagues found that ethnic/linguistic matching between user and PMHP was one of the main factors that positively influenced medication compliance (Ziguras et al., 2001). It may be that clients are more willing to accept advice from case managers who they feel have a better understanding of their cultural values and beliefs. This interpretation is consistent with the findings of many previous studies (Flaskerud, 1986;Flaskerud & Liu, 1991;Sue et al., 1991).
Gender concordance between user and PMHP. Manthey and collaborators (Manthey et al., 1982) found that neither therapists' gender nor gender matching with user was significantly related to the duration of counselling. Several pioneering studies on the topic of patients' preferences for the gender of the PMHP were conducted among male war veterans. A qualitative study among veteran men who had experienced military sexual trauma (MST) found that veterans had mixed provider's gender preferences, with 50% preferring a female provider, 25% a male provider and 25% reporting no gender preference (Turchik et al., 2013).
The personal style and attire of the PMHP. The PMHP's style and attire seem to influence users' preference in different ways. First of all, it seems to affect the patient preference. According to Priebe et al. (2017), long-term users preferred cautious treatment presentations, while recent users with little experience of mental health services preferred an optimistic style in the presentation of available treatments. Second, mental health service users expressed their preference for a supportive, flexible, respectful and professional relationship. Boundaries were identified as helpful for creating safety and respecting personal privacy, contributing to create an experience of connection within the helping relationship (Grant & Mandell, 2016). In longtime consumer-provider relationships, trust and partnership were considered as important in the context of shared decision-making in public mental health services (Woltman & Whitley, 2010). Third, patients seem to have a preference regarding the attire of their PMHP. In a study investigating 163 war veteran users in military services on their preferences on how the mental health professional should be dressed, it was established that while only a small portion of participants preferred a physician in uniform (5%), the majority (65%) had no preference (Gould, 2011). According to other studies among users affected by chronic mental illness, users' opinion was that the professional should dress in a comfortable manner (Nihalani et al., 2006). Conversely, another study concluded that, when the images of the professional were evaluated in terms of referral for treatment, trust in treatment and willingness to share their confidential matters, users preferred the 'traditional' white coat (Atasoy et al., 2015), as it is in other medical branches (Cha et al., 2004;Maruani et al., 2013;Najafi et al., 2012;Neinstein et al., 1985).

Effect of choice on treatment outcomes
The opportunity for users to choose the PMHP on the basis of information about her or his therapy style showed that a significantly higher proportion of users who could choose their professional kept their scheduled appointments in comparison with users who could not make a choice (Ersner-Hershfield et al., 1979;Larsen et al., 1983). Conversely, Manthey and colleagues, in a study among 14 users of a community mental health centre, did not find a statistically significant influence of the choice of the PMHP on therapy outcomes. However, users perceived positively being able to choose their therapist: they reported feeling respected, responsible for and in control of themselves, and more willing and hopeful about participating in therapy (Manthey et al., 1982).
Several studies explored the effects of ethnic matching on the treatment outcomes. A study of 2,935 Australian users showed that ethnic matching was associated with a lower level of contact with emergency crisis assessment and treatment team services for clients with a non-English-speaking background. The effect of ethnic matching was more pronounced for more recently settled groups or those with poorer English-language skills (Ziguras et al., 2003). Chao and colleagues explored the effects of Working Alliance (WA) and client-clinician ethnic match on recovery status among 67 patients. Clients in the ethnically matched group reported significantly higher WA compared to the nonmatched group, suggesting that, in a multicultural community, ethnic matching may help augment WA and address potential barriers to treatment engagement (Chao et al., 2012). Other studies, though, did not support this hypothesis (Chinman et al., 2000;Ortega & Rosenheck, 2002).
Finally, a study investigating 224 women who participated in a clinical trial of group treatment for post-traumatic stress disorder (PTSD) and substance use disorders reported that racial/ethnic match did not confer additional benefits for Black clients in terms of PTSD outcomes; on the contrary, White clients, with severe PTSD symptoms at baseline, matched with their therapist, had greater reductions in PTSD symptoms at follow-up than their counterparts who were racial/ethnically mismatched. For substance use outcomes, both Black and White patients who were light substance users at baseline benefitted from the individual racial/ethnic match with their therapist, which resulted in lower odds of heavy substance use posttreatment (Ruglass et al., 2016).

Discussion
Our goal was to review studies investigating users' opinion and preferences about the topic of the choice and the change of the PMHP in community-based mental health services.
While it is generally agreed that it is important to take into consideration users' preferences in choosing or in the request of changing one's PMHP, our review shows that available studies are few, small in size and generally old.
Critics of the choice of care in mental health services express concerns about the practical implementation and the potentially negative consequences to the patient (Samele et al., 2007): they argue that creating the type of infrastructure required to support patient choice could be highly complex (Goodwin, 2006) and that too much choice can be debilitating and may increase the risk of mistakes in decision-making or have negative psychological consequences to the patient (Bate & Robert, 2005;Valsraj & Gardner, 2007). In spite of this, our review shows that there are no studies that assess whether guaranteeing the choice of the PMHP in the real world is really that difficult. Addressing the perceived constraints may result in more choice options to reach therapeutic goals in a collaborative framework with patients (Galeazzi et al., 2007).
Existing research has mostly explored the factors related to the user -PMHP matching. Characteristics of the dyad which seem to influence users' preferences are matching (or differing) in age, ethnicity, language and gender. This trend in international research may represent a positive development, highlighting the increasing interest in a collaborative model of care in line with the recovery model (Anthony, 2007).

Limitations of the study
A limitation of our review is that we have been restrictive with respect to the setting, including only studies conducted in community mental health services. Moreover, we have focused on public outpatient settings, excluding researches conducted in private practice, where several studies investigating patients' preferences and key aspects of patients/therapist dyads were conducted (Alegría et al., 2013). Finally, the methodological differences in outcomes and data collection, and the heterogeneity of the mental health community-based services' organizations, create methodological difficulties that made the comparisons between the included studies not always feasible.

Conclusion
Concerns about practical and organizational aspects (Samele et al., 2007) and prejudices towards people with mental health problems about their capacity to choose their pathways of care (Bate & Robert, 2005;Valsraj & Gardner, 2007) could eventually play a role in the neglect of the topic of choice of one's PMHP we have found in this review, undermining what could be a significant opportunity for service users, carers and professionals.
Four main aspects concerning the choice and request of changing PMHP emerged: (1) service users seem to appreciate the option of choosing their PMHP; (2) users stressed the importance of the continuity of care in the relationship with the allocated PMHP; (3) some inconclusive research is available on the factors of the users/PMHP dyad influencing users' preferences, such as matching (or differing) in education, age, gender, ethnicity, nationality, language; and (4) research focusing on the effects of the option of choosing and changing one's PMHP on treatment outcomes is scarce.
PMHPs and other stakeholders' views on this topic should be further explored also by means of intervention studies comparing different systems for letting service users choose and change PMHP, in order to inform policies regarding choice and to appropriately manage users' requests.