A Systematic Analysis of Global Development Trends, Challenges, and Priorities of SDM Research in Mental Health

People with mental illness often experience difficulties maintaining social relationships and increased loneliness. Shared decision making (SDM) in mental health has demonstrated benefits such as increased knowledge of options and conditions, greater involvement in decisions, and higher satisfaction with clinical interactions. Despite these benefits, SDM adoption and implementation in mental health lags behind physical health due to stigma, concerns about patient decision-making capacity, and clinician fears of liability. A key, under-discussed barrier is the absence of a coherent SDM model developed specifically for mental health with input from people with mental illness, their families, and clinicians. Definitions, measures, and core components of effective SDM interventions remain inconsistent. This umbrella review aims to clarify what constitutes an effective SDM approach in mental health and to identify core targets and elements required for successful utilization and implementation of SDM.

Across included reviews, SDM models cited were predominantly adapted from physical health (e.g., Charles et al.’s medical decision-making model, the Integrative Model of SDM, the Three Talk Model, and the SDM 3-Circle Model). Reviews commonly listed components such as providing decision support, presenting options, discussing pros/cons, eliciting patient values/preferences, and making or deferring a decision. However, components essential in mental health clinical interactions—such as building trust, establishing a therapeutic alliance, supporting personal recovery, activation, and empowerment—were not adequately addressed by these models. Measurement tools frequently used (e.g., Autonomy Preference Index, Observing Patient Involvement in Decision Making, Decision Self-Efficacy, Decisional Conflict) were not developed or validated specifically for people with mental illness, and there was no consensus on primary outcomes for SDM in mental health. The literature has focused narrowly on patient–clinician dyads and psychopharmacological decisions, with limited attention to broader decision domains or involvement of family/carers and peer support workers.

Design: Umbrella review (systematic overview of systematic reviews and meta-analyses), reported per PRISMA 2020, with a priori registration (PROSPERO: CRD42020190700). Working definition of SDM: a health communication approach focusing on patient–clinician–family/carer interactions around treatment decisions to improve clinical and functional outcomes through personalized care. Search strategy: Comprehensive searches in nine databases (CINAHL, PubMed, Scopus, MEDLINE, EMBASE, Cochrane Library, Web of Science, Scopus, PsycInfo) using terms including shared decision making, decision support system, decision aid, informed choice/decision. Inclusion criteria: Systematic and scoping reviews (2010–2021) of SDM interventions in adult mental health across any setting, delivered by various professionals, targeting patients, clinicians, and/or caregivers. Reviews could include quantitative, qualitative, or mixed-methods primary studies. Outcomes were categorized as affective-cognitive, behavioral, and health. Screening/selection: Initial title/abstract screening by one reviewer with 10% double-screened; full-text screening by two reviewers with consensus resolution and senior arbiter as needed. Quality appraisal: AMSTAR 2 for methodological quality of reviews; GRADE ratings extracted and adjusted as needed. Overlap across reviews assessed via the Corrected Covered Area (CCA) index. Due to heterogeneity, results were synthesized narratively rather than meta-analyzed. Data extraction/synthesis: Two reviewers independently extracted review characteristics, PICO, settings, number/design of primary studies, SDM taxonomy, quality assessments, and findings. Barriers/facilitators and limitations were collated; potential publication biases considered; recommendations compiled.

- Search results and inclusions: 7,383 records identified; 25 full-text reviews assessed; 10 systematic reviews met inclusion criteria. The 10 reviews contained 168 studies; 105 non-overlapping unique articles were identified, of which 100 were primary studies (5 primary studies had overlapping publications). - Participants and settings: Across 10 reviews, 47,846 participants (47,071 patients; 676 clinicians; 5 family carers; 90 other carers; 4 other). Studies spanned 16 high-income countries. First study in 1999; most recent in 2020. Median total sample size per study was 167.5 (range 10–13,734). - Overlap: CCA indicated 4% overlap across reviews; 20/105 studies appeared in at least two reviews; overlap between individual reviews ranged 0%–60%. - Methodological quality (AMSTAR 2): Overall poor quality—critically low: 60%; low: 30%; high: 10%. Only 30% provided a priori protocols; 80% conducted comprehensive searches; 30% listed excluded studies; 30% performed meta-analyses. - SDM conceptualization and components: All reviews referenced broader SDM literature; most listed at least half of 14 standard components. Most frequent was “provide decision support” (10/10 reviews); least frequent was “check/clarify understanding/summarize” (1/10). Mental-health-specific elements (trust, therapeutic alliance, empowerment, recovery support) were not addressed by cited models. - Measures and outcomes: Wide range of tools and outcomes used (e.g., satisfaction, knowledge, adherence, symptoms, QOL, readmissions, recovery), but no instrument developed/validated specifically for people with mental illness; no consensus on SDM outcomes prioritized in mental health. - Interventions and diagnoses in primary studies: Interventions included decision support tools (DSTs) only, multicomponent interventions with/without DSTs, and shared care planning/preferences elicitation. Settings: mental health hospitals (N=15), general hospital (N=1), community mental health (N=56), community health (N=28). Diagnoses: schizophrenia-spectrum (N=31), mixed SMI cohorts (N=29), depressive disorders (N=27); rare: eating disorders (N=1), borderline personality disorder (N=1). SDM components frequently implemented: enhance participation/activate and empower (N=37), discuss patient values/preferences (N=32), establish therapeutic alliance (N=11), define goals/actions (N=11), make the decision (N=9). - Effectiveness signals: Studies reported significant improvements across SDM-related outcomes (e.g., involvement, knowledge), health behaviors (e.g., adherence), satisfaction with care, symptoms, QOL, recovery, trust, and social functioning. Interventions using decision aids often improved information exchange; several non-DA interventions improved SDM-related, behavioral, and health outcomes. - Global trends: 43% of primary studies were published 2015–2020, indicating growth. Research predominantly focused on dyadic patient–clinician psychopharmacological decisions in community settings, with limited inclusion of family/carers or peer support workers and limited attention to broader decision domains. - Equity and culture: All studies were in high-income countries; interventions seldom addressed socio-cultural contexts, stigma, or decision-making capacity; evidence base biased toward White and Western populations.

The review clarifies that SDM in mental health research has largely adapted models and measures from physical health, emphasizing decision aids and information exchange within dyadic medication-focused encounters. While such approaches can yield improvements in knowledge, involvement, adherence, and some clinical outcomes, they neglect critical mental-health-specific dimensions such as trust-building, therapeutic alliance, empowerment, recovery orientation, and fluctuating decision capacity. Findings support expanding SDM beyond psychopharmacology to include psychosocial and social prescribing decisions and moving from dyadic to polyadic models that involve family members, carers, and peer support workers. Person-driven, recovery-oriented outcomes (empowerment, self-efficacy, hope) should be prioritized alongside clinical metrics. Moreover, measurement tools need co-development and validation with people with mental illness to ensure relevance, including assessment of decision-making capacity and attention to stigma. Addressing cultural and socio-economic diversity gaps is essential to align SDM outcomes with what matters to patients and families, particularly marginalized groups.

This umbrella review provides a comprehensive synthesis of SDM interventions in mental health and proposes a shift toward a polyadic, person-centered framework that includes family members, peer support workers, and non-psychiatric professionals. It highlights the need to broaden decision domains beyond medication, prioritize recovery-oriented outcomes, and develop and validate SDM measures specifically with and for people with mental illness. The work sets a new direction for SDM research and implementation in mental health toward person-driven measurement approaches and participatory methods.

- Heterogeneity of measures, settings, and samples precluded meta-analysis; narrative synthesis was used, and judgments about effectiveness relied on statistical significance rather than effect sizes. - Potential language bias due to inclusion of English-language publications only. - Exclusion of studies involving participants under 18 years. - Variability and generally low methodological quality of included reviews (per AMSTAR 2) limit confidence in pooled conclusions. - All included studies were from high-income countries, limiting generalizability to low- and middle-income settings.