Article Text
PDF
Abstract
Background Severe mental illness (SMI) incorporates schizophrenia, bipolar disorder, non-organic psychosis, personality disorder or any other severe and enduring mental health illness. Medication, particularly antipsychotics and mood stabilisers are the main treatment options. Medication optimisation is a hallmark of medication safety, characterised by the use of collaborative, person-centred approaches. There is very little published research describing medication optimisation with people living with SMI.
Objective Published literature and two stakeholder groups were employed to answer: What works for whom and in what circumstances to optimise medication use with people living with SMI in the community?
Methods A five-stage realist review was co-conducted with a lived experience group of individuals living with SMI and a practitioner group caring for individuals with SMI. An initial programme theory was developed. A formal literature search was conducted across eight bibliographic databases, and literature were screened for relevance to programme theory refinement. In total 60 papers contributed to the review. 42 papers were from the original database search with 18 papers identified from additional database searches and citation searches conducted based on stakeholder recommendations.
Results Our programme theory represents a continuum from a service user’s initial diagnosis of SMI to therapeutic alliance development with practitioners, followed by mutual exchange of information, shared decision-making and medication optimisation. Accompanying the programme theory are 11 context-mechanism-outcome configurations that propose evidence-informed contextual factors and mechanisms that either facilitate or impede medication optimisation. Two mid-range theories highlighted in this review are supported decision-making and trust formation.
Conclusions Supported decision-making and trust are foundational to overcoming stigma and establishing ‘safety’ and comfort between service users and practitioners. Avenues for future research include the influence of stigma and equity across cultural and ethnic groups with individuals with SMI; and use of trained supports, such as peer support workers.
PROSPERO registration number CRD42021280980.
- shared decision making
- compliance
- medication safety
- mental health
- patient-centred care
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
This is an open access article distributed in accordance with the Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits others to copy, redistribute, remix, transform and build upon this work for any purpose, provided the original work is properly cited, a link to the licence is given, and indication of whether changes were made. See: https://creativecommons.org/licenses/by/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Medication optimisation is challenging for both people living with severe mental illness and their prescribing clinicians; medication non-adherence is common.
WHAT THIS STUDY ADDS
Effective optimisation of medication requires a person-centred approach embedded throughout a service user’s journey from initial diagnosis to effective medication co-management with practitioners.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Research is needed in multiple aspects of medication optimisation, including transition from acute care to community, the role of trained peer support workers and practitioner awareness of unique needs for individuals from ethnic and cultural minority groups.
Introduction
Medication optimisation
Severe mental illness (SMI) is a significant global healthcare burden with rates increasing throughout the world.1 The term SMI includes diagnoses of schizophrenia, non-organic psychosis, bipolar disorder, personality disorder and any other severe and enduring mental illness.2 Medications are a key treatment for SMI, but medication side effects can contribute to chronic physical illness (eg, diabetes, cardiovascular disease), a diminished quality of life and a decreased lifespan.3 4 Complex medication regimens are often used to treat SMI; dosing can be a delicate balance between overprescribing and underprescribing, based on individual service user’s (SUs) unique needs.5
Given the complex nature of SMI medication management and the need to consider issues such as risk of relapse, serious side-effect profiles and potential drug-drug interactions, medication safety is of paramount importance to SUs and practitioners.6–8 Psychotropics have many safety challenges. Metabolic adverse events including weight gain, changes in cholesterol and triglyceride levels and increases in blood sugar are a particular concern.9 Many psychotropics have anticholinergic effects; these include constipation, falls and confusion.10 11 This confusion could potentially worsen the negative and cognitive symptoms of schizophrenia and increase the risk of medication errors.7
Since 2008, the global Institute for Healthcare Improvement (IHI) and its country affiliates, such as the UK’s Health Foundation, have advocated for inclusion of SUs and person-centred care approaches when identifying best practices, strategies pertaining to patient safety and quality of care delivery12 as well as promoting shared decision-making (SDM) in healthcare systems.13 The original Triple Aim IHI framework consisted of three pillars for advancing quality and safety: enhanced population health, positive SU experiences and cost-effectiveness.12 The original framework has expanded to Quadruple Aim, including staff experience.12 These IHI frameworks highlight how the SU voice is an integral component of all healthcare quality and safety initiatives. Medication optimisation, a hallmark of medication safety, is defined as ‘a person-centred approach to safe and effective medicines use, to ensure people obtain the best possible outcomes from their medicines’.14 Effective medication optimisation involves the perspectives of SUs with lived experience of taking medications.15 Multidisciplinary care delivery for SMI is more effective when SUs play a central role in medication decision-making.16
Failure to optimise medication is often attributed to SU non-adherence, practitioner underprescribing or overprescribing or overtreatment, including polypharmacy.2 16–18 Management of SMI is particularly challenging with reported non-adherence rates as high as 50%.19 20 Non-adherence21 and overprescribing occur more frequently in ethnic minority communities, as do physical illnesses, such as diabetes and cardiovascular disease.14 22 In general, the lowest possible medication dose is recommended to control SMI symptoms,23 however higher doses are often prescribed by practitioners concerned about relapse.24 Poorly managed SMI increases relapse rates, hospitalisation and is associated with unemployment, homelessness, disrupted education, substance misuse, physical health problems, self-harm and excess mortality.2 17 18 25 A systematic review and meta-analysis of studies from Asia, Europe and North America found that non-adherence within the SMI population is the strongest predictor of relapse.26
Shared decision-making
Person-centred approaches, such as SDM between key practitioner groups (eg, pharmacy, medicine, nursing), SUs with SMI and family carers, are associated with increased SU medication adherence and improved practitioner prescribing practices.7 17 19 20 There is, however, limited research on what needs to happen, how and when in the SU-practitioner relationship to promote person-centred SDM and ultimately, medication optimisation for SUs with SMI.4 17 27 Assumptions are often made about intervention effectiveness only from practitioner’s viewpoints.28 The implementation of SDM can be hindered by practitioners’ beliefs about SDM. A Netherlands-based study compared practitioner reports of SDM use with direct observations of their SU interactions.29 Practitioners reported using SDM as their usual decision-making style, but in observations, there was low engagement with SUs. The authors described practitioners as ‘unconsciously incompetent in SDM’. Therefore, developing knowledge on how to implement effective person-centred approaches that promote medication optimisation is needed.
Present research
We conducted a realist review on medication optimisation with community dwelling SUs living with SMI. We focused on community dwelling as most SMI SUs live in the community, where there are greater opportunities for them to exercise control over their medication regime (eg, by omitting doses or via non-adherence).
We synthesised data from academic literature and drew on perspectives of community-based stakeholders with lived experience of SMI, informal (family) carers and mental health practitioners caring for SUs with SMI. A realist review can uncover important contextual factors affecting outcomes.30 We constructed a programme theory comprising a series of testable hypotheses, known as context-mechanism-outcome configurations (CMOCs), to explain a potential SU-practitioner journey from initial diagnosis to trusting therapeutic alliance, SDM and medication optimisation.
Realist reviews have become increasingly popular within the quality and safety literature to explain how and why interventions work. Realist reviews have been used to investigate junior doctors’ antimicrobial prescribing31; safety-netting practices in primary care32 and medication management for community dwelling seniors on complex medication regimens.33 Realist reviews address research questions about what works, for whom, under what circumstances and how, and are a valuable methodological alternative or complement to other forms of evidence synthesis, such as systematic reviews.30 34
Research objectives
The overall objective for this realist review was to use published literature, alongside lived experience and practitioner stakeholder groups to understand: What works for whom and in what circumstances to optimise medication for community dwelling SUs with SMI.
Methods
We conducted a five-stage realist review. Our review protocol was published35 and registered a priori with PROSPERO (CRD42021280980). We used academic literature as well as feedback and advice from our stakeholder groups (lived experience group (LEG) and practitioner group (PG) to refine our programme theory, and create a series of testable CMOCs.36 The LEG comprised six lived experience stakeholders from Birmingham and Solihull Mental Health NHS Foundation Trust (BSMHFT) Lived Experience Advisory Research (LEAR) Group and two additional individuals with lived experience from outside LEAR (who were recruited to facilitate discussions). The PG comprised healthcare practitioners from the UK caring for SUs with SMI. The practitioners were recruited from personal networks and via social media advertisements.
Our CMOCs describe specific contexts associated with important outcomes related to medication optimisation, such as therapeutic alliance formation, and they articulate the mechanisms that trigger these outcomes. In this project, we have defined the following:
Setting: adults with SMI taking medication living in the community.
Intervention: any intervention to optimise medication usage for SUs living with SMI.
Context: situational factors that create conditions necessary to trigger underlying mechanisms.
Mechanisms: hidden, psychological processes that link specific contexts to intended outcomes.
Outcomes: quality of life, adherence, adverse events, disease symptoms, economic.
Stage 1: objectives, initial programme theory
Objectives
To conduct a realist review using published literature alongside lived experience and practitioner stakeholder input to understand: What works for whom and in what circumstances to optimise medication use for community dwelling SUs with SMI.
Development of initial programme theory
We developed an initial programme theory (IPT), a testable explanation of how and why medication optimisation is supposed to work for people living with SMI. Prior to the application for funding, IM initially consulted with stakeholders with lived experience from BSMHFT. These conversations helped shape the initial funding application and formed the tentative foundations for the IPT. Through retelling accounts of their own experiences of SDM within mental health services, stakeholders stressed the importance of the therapeutic relationship in generating positive outcomes. Once the project started, JH liaised with subject matter experts known to the project team to gain a wider understanding of SMI, medication utilisation and SU-practitioner SDM. Additionally, an informal literature search primarily using Google Scholar was conducted. This step was used to gain an understanding of the factors influencing SDM within SMI. Additional publications recommended by the project team were also considered. The IPT was developed from these sources.
Stage 2: literature search
A formal literature search was conducted in January 2022 by CD across eight bibliographic databases (MEDLINE, Embase, PsycINFO, CINAHL, Cochrane Library, Scopus, Web of Science Core Citation Indexes (Science Citation Index Expanded, Social Sciences Citation Index, Arts & Humanities Citation Index, Emerging Sources Citation Index, Conference Proceedings Citation Index, Book Citation Index) and Sociological Abstracts). Our searches combined free text and subject heading terms for SMI, with terms describing medication or medication optimisation, and a comprehensive list of terms reflecting our project focus on SDM and SU-practitioner relationships. The LEG and PG helped to identify key concepts used in our search strategy. Our original protocol indicated that we would run searches in Google Scholar, but this was deemed unnecessary following screening, in light of the volume of literature already retrieved. In response to PG and LEG feedback, additional targeted searches were conducted in June 2022 to identify material relating to internet use, peer support and tapering medication.
Full details of our search strategies are available in online supplemental file 1.
Supplemental material
Stage 3: screening and inclusion
Inclusion criteria focused on community dwelling adults (aged 18+ years) living with SMI and taking antipsychotic medication. Studies limited to inpatient settings or focused on diagnoses outside of SMI were excluded. Full details of the inclusion/exclusion criteria can be found in table 1.
Review: inclusion and exclusion criteria
Screening
The results of the main search were screened by title and abstract by JH using RAYYAN (a web-based tool designed to assist with screening of title and abstract). In line with realist methodology, titles and abstracts were screened to determine relevance to the developing programme theory. A random 10% sample was screened in duplicate by MMP. Uncertainties were resolved via discussions with JH and MMP. Full-text screening was initially completed in EndNote X9 by JH on documents published from 2014 onwards. The decision to focus on this timepoint was based on a significant increase in documents with SDM content during this time period. For full-text articles to be included at this stage, they had to contain data relevant to the developing programme theory with details on the SDM process in relation to prescribing, switching, tapering and taking antipsychotic medication. All full-text documents were assigned a star rating of one to five in EndNote by JH, based on a global judgement of each document’s likely relevance, richness and rigour to confirm or dispute the developing programme theory. One-star documents were deemed irrelevant and rejected. Two-star documents were deemed ‘unsure’, and these were subsequently discussed with MMP and reallocated. Three-star documents were deemed irrelevant for programme theory development but potentially useful for background material. Four-star documents were deemed relevant to CMOC development and programme theory refinement. Five-star documents were deemed the most conceptually rich and relevant to CMOC development and programme theory refinement.
Pre-2014 documents and documents obtained via citation searching and personal networks were purposively screened and analysed by CD, JH, MMP and HH but were not categorised with a star rating as they were chosen due to perceived high relevance, richness and rigour.
Stage 4: data extraction, analysis
Data extraction
Document characteristics were extracted to an Excel spreadsheet by HH (online supplemental file 2).
Supplemental material
Data analysis
A coding framework was iteratively and inductively developed and tested by MMP, JH and HH to organise relevant data. The coding framework was developed in conjunction with the LEG and PG. The framework was tested using a subset of papers from the review. MMP, HH and JH tested each iteration of the framework, and codes were added to ensure the framework encapsulated all relevant data. The framework comprised three conceptual buckets that reflected a more abstracted level of data, for example, decision-making, medication management interventions and therapeutic alliance. In total, there were nine revisions to the framework. The final framework in online supplemental file 3 demonstrates the different levels of abstraction.
Supplemental material
Coding of post-2014 five-star full-text documents was completed in NVivo by MMP and HH with a 10% check in duplicate by JH. Extracts of data were coded to nodes (termed parent nodes in NVivo) reflecting conceptual buckets. Extracts of data were coded against subnodes (called child nodes in NVivo) and multiple nodes if appropriate.
Once all five-star papers were coded and discussed with the PG and LEG, a pragmatic decision was made to focus the review on decision-making and therapeutic alliance. Coded data based on the coding framework were initially extracted and imported in Microsoft Word by MMP and HH. Tentative CMOCs were developed and refined through ongoing discussions with JH, MMP and HH. Approximately 30 hours of collaborative discussion took place over MS Teams.
Stage 5: data synthesis, CMOC development and programme theory refinement
Each tentative CMOC was individually reviewed, verified against the underpinning evidence and refined. Similar CMOCs were amalgamated where appropriate. These CMOCs were iteratively refined by checking remaining data from NVivo nodes and extracting relevant examples. Further refinement of the programme theory and CMOCs occurred using data from relevant four-star papers, pre-2014 papers, papers from additional searches and discussions with the project team, LEG and PG. The finalised set of CMOCs and a refined programme theory were discussed and validated with the PG, the LEG and the wider project team. The refined CMOCs, supporting evidence and document origin (eg, post-2014, via citation search or personal networks) can be found in online supplemental file 4.
Supplemental material
Results
Our main search identified 1118 unique results. After title and abstract screening, 144 documents published from 2014 onwards were screened in full text. Twenty-nine papers were assigned a five-star rating and coded in NVivo. Thirty-three papers were assigned a four-star rating. Following the decision to narrow the focus, of these 62 papers, 27 were rejected leaving a total of 35 four-star and five-star papers in the review. Eighteen papers were identified via additional searches, citation searches and personal contacts. Seven of the pre-2014 papers were deemed relevant taking the total number of papers contributing to the review to 60 (35+18+7). The inclusion criteria associated with SMI varied between the included papers. Twenty-eight papers included schizophrenia or schizoaffective disorder,37–64 3 papers included borderline personality disorder,38 58 59 9 papers included bipolar disorder,43 49 51 54 59–61 65 66 19 papers included psychosis/psychotic illness,38 40 43–45 51 59 61 67–77 20 papers had a broad definition, for example, SMI or people taking antipsychotic/psychotropic medication,37 38 66 75 78–93 3 papers included less severe forms of mental illness alongside SMI in their population samples,51 60 66 2 papers did not have a patient population and focused on consultant psychiatrists94 and secondary care mental health pharmacists.95 Several papers included more than one type of SMI. Our searching and screening processes are summarised in figure 1.
Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020 flow diagram for new systematic reviews which included searches of databases, registers and other sources. Adapted from Page et al.114
Table 2 includes our refined programme theory and 11 CMOCs underpinning the theory. Our refined programme theory describes a journey of medication optimisation for SUs with SMI that begins with initial diagnosis and culminates in a therapeutic alliance characterised by underlying trust, mutual information exchange and SDM. There are potential barriers and facilitators along the way, represented by positive and negative CMOCs. The journey includes practitioners, SUs with their family and social network and other information sources (eg, internet, peer support workers).
Refined programme theory and underpinning context-mechanism-outcome (CMOC) configurations
In table 2, CMOCs 1 (first contact) and 2 (relief) are associated with initial diagnosis. The literature highlights the importance of positive first encounters with healthcare services. Negative, coercive experiences can derail practitioner-SU trust formation, while positive experiences can decrease internalised stigma and reassure SUs that their condition is treatable.
CMOC 3 (dismissal) depicts how dismissal and devaluing of SUs by practitioners impedes the establishment of trust, which is a foundational component to therapeutic relationships.
CMOC 4 (being heard) illustrates how development of the therapeutic alliance is dependent on respectful, supportive practitioners willing to listen to and seriously consider SUs’ needs and concerns.
CMOCs 5 (practitioner information exchange) and 6 (seeking more information) represent SUs’ desire for credible, trustworthy information about diagnosis and medication that is personalised to them and the role of medication in treating their illness, including possible side effects. Regardless of information obtained from practitioners, SUs typically seek out additional information to obtain new knowledge, a greater sense of reassurance and more control over medication and life decisions.
As described in CMOC 7 (confiding and negotiating in a safe way), a hallmark of strong and effective therapeutic alliances is the ability of practitioners to support SUs, even if they disagree with their medication decisions. SUs feel safe in this type of alliance and are more apt to collaboratively plan their care with a trusted practitioner.
In contrast, CMOC 8 (perceived risks) illustrates why some practitioners might find it challenging to meet SUs’ wishes to taper their medication due to risk about potential adverse outcomes, such as relapse and its consequences.
CMOCs 9 (family and social supports), 10 (fear and guilt) and 11 (peer supports) are related to non-practitioner sources of support for individuals with SMI. The family can be a safety net and positive support for SUs (CMOC 9), or the family can be fearful of making medication changes (CMOC 10), resulting in negative SU emotions, such as fear and guilt. In these situations, SUs may feel pressured to conform to family wishes to avoid conflict and ensure that family support is not withdrawn. Peer support workers (CMOC 11) are a promising source of support to SUs, because they validate SUs’ feelings, given their lived experience with SMI. However, research on peer support worker roles in medication optimisation was lacking.
Discussion
The patient safety literature has demonstrated that higher levels of safety are achievable by ensuring the voice of SUs and other stakeholders are part of quality improvement efforts.96 Medication optimisation is an important component of patient safety, especially for SUs with SMI, where medication is a key treatment strategy.16 Our programme theory and CMOCs highlight how person-centred care approaches such as providing relevant, useful information and support through practitioners and others can lead to safe medication use (ie, medication optimisation). The CMOCs outlined above provide testable, causal explanations for outcomes, detailing by whom, when and how these happen.
Comparisons with existing literature and theory
A valuable aspect of the realist approach is the potential to use formal or substantive theories to further explain and buttress inferences about underlying mechanisms or drivers for individuals’ actions.30 Based on our reading of the included documents and recommendations from our stakeholder groups, we discuss below two theories of particular importance to the development of a therapeutic alliance, which is a necessary condition for effective optimisation of medication.
Supported decision-making theory
Supported decision-making theory emphasises practitioners’ roles in assisting SUs with decision-making based on their needs and preferences. Supported decision-making theory has legal and ethical associations with the 2006 United Nations Convention of Individuals’ Rights and Disabilities, which stipulated that no person should be appointed as a decision-maker for an individual who has the capacity to make their own decisions with appropriate supports.97 Supported decision-making encompasses person-centred planning, advocacy, communications, interpretive supports and representational supports (eg, peer support workers, family and social networks). The central question practitioners should ask themselves is: ‘What supports are needed to ensure this person can best exercise their rights?’97
Although research into supported decision-making and SMI is rare, qualitative researchers found that the timing and types of practitioner supports made a difference to individuals with SMI, particularly with respect to confidence and self-control.98 99 Researchers from one study created different thematic labels for SUs: SUs with capacity to make their own cogent decisions were ‘inward experts’, while SUs during periods of acute unwellness were ‘outward entrustors’, entrusting practitioners to guide their care management. SUs’ variable needs for supported decision-making required different practitioner roles, such as practitioners as facilitators (eg, sharing information openly and honestly) and as collaborators (eg, promoting SDM).98
In a recent systematic review of supported decision-making for SUs with SMI in clinical practice settings,100 a limited number of papers included in the review examined stakeholders’ perspectives of supported decision-making. Stakeholders, including SUs, family members and practitioners, all agreed to the importance of supported decision-making. Practitioner misconceptions about differences between SUs’ rights and preferences, however, were barriers to implementation success. If supported decision-making is a necessary condition for SU-practitioner SDM and medication optimisation, practitioners will need to understand SUs’ legal rights, and to engage in roles (eg, facilitator, collaborator) that promote SUs’ decision-making autonomy.98
Trust formation theory
Our findings make clear that, for SUs with SMI, ongoing alliance-building and confidence in their capacity to share in decisions and manage their lives depends on trust: trust in practitioners and trust/belief in themselves. Trust formation theory defines trust as ‘a psychological state comprising the intention to accept vulnerability based on positive expectations of the intentions or behaviour of another’.101 Our programme theory proposes that trust between SUs and practitioners evolves with the development of the therapeutic alliance. An exploratory study of the role of trust in medication management within mental health services102 supported our realist review findings that practitioners’ reluctance to share useful information in an open and honest way can create mistrust and worsen medication adherence. Ultimately, mistrust obstructs collaborative medication management.102 103
The wider literature provides some evidence of how mutual trust formation enables engagement, disclosure and collaboration in mental healthcare. Corroborating our findings, a qualitative study from the UK found that practitioners’ open communications and therapeutic listening promoted and sustained the development of mutual trust over time.104 More recent literature suggests that the development of a trusting therapeutic alliance is enhanced by practitioners’ awareness and respect for SU preferences, such as types of treatment options (eg, medications or psychotherapy), and influenced by the characteristics of practitioners they work with (eg, professional background, gender, age and ethnicity).105 Even when all preferences cannot be accommodated, eliciting, discussing and acknowledging SU preferences is associated with stronger alliances.105
The relationship between SDM, trust and information
CMOCs (4, 5, 7) associate SDM with active engagement of SUs and practitioners in open, transparent discussions and collaborative treatment planning based on mutual trust. CMOC 8, however, addresses the negative emotions and risks associated with the SU-practitioner relationship. The literature in our review focused predominantly on practitioner risks, such as concern for SU medication non-adherence and adverse SU outcomes. A recent review of qualitative studies106 discussed risk-taking from the perspectives of SUs and practitioners. With shared risk-taking, both parties jointly reflect and address inherent risks to any decision, particularly from a safety perspective. Some evidence suggests that mutually identifying and preventing or mitigating risks can actually strengthen the alliance and deepen trust.107 In the UK, the National Health Service recommends that risk assessment and management should be explained to SUs with SMI as soon as possible as part of SDM.108
CMOCs (6, 9, 11) pertain to non-practitioner sources of information and support, including the internet and social media, family and friends and peer supports with similar lived experience, although we identified very limited academic literature on how peer support workers can be used to optimise medication management.39 41 65 69 81 86 Questions exist with respect to peer support workers’ capacity to give accurate medication advice.109
Future research directions
Future research needs to address non-practitioner sources of information and supports, (eg, peer support workers, families and friends, internet and social media). Our LEG and PG stakeholders both endorsed the importance of peer support workers and their potential roles in medication optimisation.
An area of burgeoning interest is online decision support tools to improve information sharing and communication between SUs and practitioners. A recent systematic review found mixed evidence for the effectiveness of decision support tools with SUs with SMI.110 This review included tools to assist with prioritising treatment preferences, crisis planning and advanced directives. More conclusive research is needed to evaluate the efficacy of online support tools, especially how they help or hinder therapeutic alliance development, SDM and medication optimisation for SUs with SMI.
Our review has highlighted an important evidence gap relating to equity, diversity and inclusivity (EDI) for SUs with SMI from racial and ethnic minority groups, seniors and other vulnerable populations. These groups were rarely mentioned in our review’s included papers. When SUs with SMI are members of minority groups and/or vulnerable populations, stigma can be compounded.27 111 Ultimately, the success of SU-practitioner relationships depends on reducing mistrust among SUs who have been stigmatised by SMI and by race/ethnicity, while enhancing practitioners’ awareness and commitment to EDI. In England, the Race Equality Foundation is a national charity that tracks and reports racial inequality in public services (https://raceequalityfoundation.org.uk/). Researchers working with this charity identified persistent healthcare inequalities for English minority groups. ‘Traumatic, inappropriate and discriminatory experiences of services can have a detrimental impact on chances for recovery, particularly if the same risk factors of bereavement, family breakdown, incarceration, poverty and exposure to racism continue to be present’.27
Community-based care models may be a more cost-effective way of caring for complex and vulnerable patients. Only a limited number of new care models (eg, team-based care, integrated care) have been well-described or evaluated for SUs with SMI.112 In one mapping review of integrated physical-mental care models for SUs with SMI, a number of concerns were identified, including practitioners’ negative biases and stigma towards SMI.113 As our review illustrates, the initial context, particularly the presence of any negative biases towards SMI, can derail the development of a therapeutic alliance between SUs and practitioners.
Another important consideration for future research centres around the skill and knowledge of individual practitioners to be able to fully engage in SDM. Our practitioner stakeholders informed us that they are often not taught how to have difficult conversations with SUs, leaving them underconfident with some of the more challenging aspects associated with engaging in SDM with this population. However, none of the papers included within this review discussed these issues.
Strengths and limitations
We conducted our realist review within a 1-year timeline. To be as efficient as possible, we focused on the largest body of relevant literature published from 2014 onwards. We returned to pre-2014 literature after developing our CMOCs from the more recent literature to identify earlier data relating to the CMOCs.
Our research team were English speakers only, therefore we excluded any non-English documents from our searches. We adopted an inclusive approach to the population and used a broad definition of SMI. However, there may be differences in the approach to medication optimisation in the different diagnoses (eg, bipolar, psychosis, personality disorder). Future research should consider a more granular approach and compare and contrast across different types of SMI. We focused on SDM between SUs and practitioners. As such, documents pertaining solely to practitioner clinical decision-making were excluded from the review. We accept that we excluded papers pertaining to the dilemmas practitioners face when making unilateral medication decisions due to situations, such as SU acute unwellness.
A strength of this review was the engagement with PG and LEG stakeholders throughout the review process. Their engagement supported our interpretations of data, ensured our findings reflected their real-world experiences and highlighted gaps in the literature. Our review identified evidence gaps in relation to the relationship between race, ethnicity, vulnerable groups and medication optimisation in SMI and in the role of peer support workers. Realist reviews are an iterative process of developing theory and CMOCs, which may then be confirmed, refuted or refined by future research, including, for example, realist evaluation. This review’s programme theory and CMOCs produced testable hypotheses for future research with SUs with SMI and community-based practitioners who serve them. In the papers included in this review, there was a lack of information on the contextual factors and mechanisms influencing clinical decision-making in challenging circumstances, such as SU acute unwellness. This was also highlighted by Martínez-Hernáez et al88 who stated that are still differences between models of concordance, described as therapeutic alliance building, and real-life practice situations.
Conclusions
Medication optimisation is a person-centred approach that begins at time of initial diagnosis and ensures optimal information and supports are accessible to SUs, based on their needs and preferences. At the beginning of the journey and during periods of relapse, supported decision-making is necessary to overcome stigma and establish trust between SUs and practitioners. Practitioners engaging in SDM must ensure that the voice of the SU is present throughout the recovery journey in order to optimise medication.
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
Ethics statements
Patient consent for publication
Ethics approval
Not applicable.
Acknowledgments
We would like to thank our practitioner and lived experience stakeholder groups for their time and effort in helping us shape this research. Members of the lived experience group included but not limited to Max Carlish, Hameed Khan, Mustak Mirza, Barbara Norden, Janelle Peacock, Ian Tighe and Rosemary.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Twitter @DrJoHowe, @maidment_dr
Contributors JH and MMP contributed equally to this manuscript and are joint coauthors. Contributors IM and plus all co-apps conceived the research project and all the authors contributed to its design. GW and CD provided methodological support. CD devised and carried out the literature searches. JH, MMP and HH undertook data extraction, analysis, synthesis and development of context mechanism outcome configurations (CMOCs) and refinement of the programme theory. The other authors (IM, GW KA, MC, SJ, SO, RU and EP) contributed to interpretation CMOCs and programme theory. KA led the PPI Lived Experience Group involvement. JH, IM and HH led the practitioner group involvement. MMP and JH drafted the manuscript and the other authors (IM, GW, KA, MC, SJ, SO, RU and EP) provided feedback. All the authors read and approved the final manuscript. IM is the guarantor for the project.
Funding This study/project is funded by the NIHR Programme Development Award (203683).
Disclaimer The views expressed are those of the author(s) and not necessarily those of the NIHR or the Department of Health and Social Care.
Competing interests IM: for list of all grants see ‘Funding’ section of: https://orcid.org/0000-0003-4152-9704; GW: for list of all grants see ‘Funding’ section of: https://orcid.org/0000-0002-5384-4157; CD: for list of all grants see ‘Funding’ section of: https://orcid.org/0000-0002-7083-6589; SJ: (1) reviews the shared decision-making learning package on an annual basis for the centre for postgraduate education; (2) non-paid specialist adviser to Pfizer on the SHARE project. RU is (1) supported by the NIHR Oxford Health Biomedical Research Centre. The views expressed are those of the author and not necessarily those of the NIHR or the Department of Health and Social Care. (2) received the following: Paid speaker fee at non promotional educational event, Otsuka June 2022 Consultancy: Vitaris July 2022 Consultancy: Springer Healthcare December 2021. (3) Undertakes the following positions: Honarary General Secretary for BAP (unpaid), Deputy Editor for the British Journal of Psychiatry 2018-2023 (unpaid), Member of Scientific Advisory Group: MIND 2023- (unpaid).
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.