The social determinants of mental health and disorder: evidence, prevention and recommendations

The paper argues that social determinants are the most modifiable targets to prevent onset of mental health problems and improve population mental health. Structural conditions such as income, employment, education, housing, food security, discrimination, childhood adversity, social support, neighborhood context, and access to care shape mental health and are unequally distributed due to policy and power structures. The authors highlight growing service demand, treatment gaps, and rising youth mental health problems, alongside persistent inequities exacerbated by shocks (e.g., pandemics, economic crises, conflict, climate change). The research question centers on how social determinants causally influence mental health and how preventive strategies can intervene to reduce inequalities. The paper’s purpose is to synthesize evidence on determinants and prevention, present a population-level prevention framework, and recommend actions grounded in social justice, given that many disorders begin by adolescence and disproportionately affect marginalized groups.

The review synthesizes evidence on individual/familial and contextual social determinants. Socioeconomic disadvantage shows strong gradients across disorders in HICs and LMICs; early-life exposure elevates risk 2–3 fold for child mental health, with effects tied to education, food insecurity, and chronic financial stress. Prenatal/perinatal adversities (maternal stress, malnutrition, obstetric complications) predict later neurocognitive and mental health problems; famine-linked prenatal malnutrition is associated with psychosis; evidence for vitamin D as causal is not supported by recent causal analyses. Childhood adversity (abuse, neglect, household dysfunction, bullying) is common (roughly 40% with at least one ACE) and linked to broad psychopathology, self-harm, psychosis, and suicide; estimated population-attributable fractions around 28–30% for children and adults suggest large preventable burdens. Migration is associated with elevated psychosis (in migrants and descendants) and high PTSD in refugees; mixed patterns for other disorders reflect selection and substantial post-migratory adversities (discrimination, isolation, deprivation). Ethnoracial minoritized groups experience higher distress and psychosis, more negative care pathways, and disparities linked to interpersonal and structural racism; discrimination prospectively predicts common mental disorders and psychosis risk. LGBTQ+ people face minority stress from stigma, discrimination, and hostile policies, elevating depression, anxiety, and suicidality; support buffers risk, and intersectional disadvantages worsen outcomes. Sex-based inequalities include higher depression/anxiety in women and higher externalizing, SUD, and suicide in men; gendered social risks (e.g., intimate partner violence) and social roles contribute. Loneliness (subjective) and social isolation (objective) are longitudinally and bidirectionally associated with depression and anxiety and linked to suicide risk; mechanisms and trials remain underdeveloped. Contextual determinants include neighborhood deprivation/urbanicity, with consistent associations for non-affective psychoses; debates on social causation vs selection persist, but some longitudinal studies controlling genetic liability still find area effects. Income inequality correlates with worse population mental health at national and subnational levels. Social capital (especially cognitive dimensions) generally associates with better mental health, but effects can be subgroup-specific and nonlinear; ethnic density often shows protective effects for psychosis and suicide (particularly for Black and Latino groups) though very high segregation can be harmful. Physical environments (housing quality/stability, built environment, air/noise pollution) are unequally distributed and linked to mental health; longitudinal evidence supports benefits of housing regeneration and harms of residential mobility in childhood; particulate air pollution (PM2.5) associates with depression/anxiety and short-term PM10 with suicide; green/blue space evidence is mixed; climate change likely exacerbates social adversities affecting mental health though direct causal evidence is limited.

This is a narrative, evidence-informed review. The authors focus on pervasive social determinants across the life course and those common across major mental disorders, prioritizing high-quality evidence (systematic reviews, longitudinal studies, RCTs and quasi-experiments) where available, and drawing primarily from Global North literature while acknowledging gaps and context-specific determinants in LMICs. They introduce and apply a prevention framework (universal, selective, indicated; with secondary and tertiary components) to organize intervention evidence. The paper explicitly notes limitations of observational studies (selection, confounding), the need for stronger causal inference, and the overrepresentation of HIC evidence.

- Social determinants are unequally distributed and causally implicated in mental health disparities across the life course. Early-life exposures have large impacts: children in socioeconomic disadvantage are 2–3 times more likely to have mental health problems; childhood adversity accounts for an estimated 28.2% of child/adolescent and 29.8% of adult psychiatric disorders (population-attributable proportions). - Migration and minoritization: Migrants and descendants have elevated psychosis risk; refugees/asylum seekers show high PTSD prevalence. Ethnoracial minoritized groups experience higher psychosis incidence and adverse care pathways, with discrimination and structural racism as key drivers. - Loneliness and social isolation show longitudinal, bidirectional links with depression/anxiety and suicide outcomes; eliminating loneliness could potentially prevent an estimated 11–18% of depression cases (causality assumed). - Neighborhood disadvantage/inequality: Robust associations between urbanicity/deprivation and psychosis persist even after some genetic controls; income inequality associates with worse mental health across settings. Social capital generally relates to better mental health; ethnic density can be protective for psychosis and suicide, with nonlinearities at high segregation. - Physical environment: Housing regeneration improves depression/anxiety; childhood residential mobility predicts later emotional/behavioral problems and psychosis. Air pollution (PM2.5) is associated with depression/anxiety; short-term PM10 with suicide; emerging links for other pollutants and psychosis. - Prevention framework and interventions: • Parenting programs (e.g., Triple P, Incredible Years) reduce disruptive behavior and internalizing symptoms; one review estimated number needed to prevent one adolescent anxiety case as ~10; benefits extend to parents; evidence of cost savings. • School-based programs: CBT-based universal/selective programs yield small but durable reductions in depression/anxiety; one review estimated up to 50% prevention of internalizing disorders at 6–9 months. Good Behaviour Game reduces conduct problems and showed lower suicidal thoughts/attempts at age 21/22 after first-grade implementation. Suicide prevention programs (e.g., SEYLE) halved suicidal ideation and attempts at 12 months. • Economic interventions: Increases in income improve mental health; guaranteed income in Manitoba reduced hospitalizations (notably mental health). Cash transfers in LMICs and HIC low-income contexts improved youth depression and child behavior; effects strongest when lifting families out of poverty; conditionality may harm some subgroups (e.g., adolescent girls). • Early-life home visiting (nurse-led) for high-risk families improved child behavior, reduced adolescent substance use and crime, and was cost-effective with long-term dividends. • Neighborhood interventions: Moving To Opportunity reduced distress/anxiety in some subgroups but had mixed adolescent mental health outcomes, illustrating potential unintended harms. Greening vacant land reduced depressive symptoms and improved self-worth. • Selective interventions for refugees: Psychosocial and community-based (bridging/linking social capital) interventions reduce PTSD and internalizing symptoms, though long-term sustainment varies and evidence quality is variable. • Secondary/tertiary strategies: Social prescribing shows promise but evidence quality is low; Individual Placement and Support consistently improves competitive employment in severe mental illness; family interventions reduce relapse in psychosis and benefit youth depression/suicidality; trauma-focused CBT and EMDR show moderate effects for PTSD, including in displaced populations, especially with cultural adaptations. - Seven social-justice–framed recommendations: center social justice; invest in multi-domain pay-off interventions; target critical life-course windows; prioritize poverty alleviation; strengthen causal inference; establish inclusive longitudinal monitoring; ensure parity of investment across primary, secondary, tertiary prevention.

Findings support embedding social determinants within the biopsychosocial model and prioritizing primary prevention alongside treatment. By focusing on modifiable social risks—especially in prenatal, early childhood, and adolescence—prevention can interrupt intergenerational cycles of disadvantage and reduce inequities. Universal strategies shift population risk distributions (per Rose’s prevention paradox), while selective and indicated strategies can address concentrated risks among marginalized groups. Effective interventions (parenting, school-based CBT and suicide prevention, cash transfers, home visiting) demonstrate that acting on social determinants yields measurable mental health improvements and broader social benefits. However, causal inference remains challenging; many data come from HICs; and complex, place-based interventions can produce heterogeneous or unintended effects, underscoring the need for rigorous, context-sensitive evaluation and equitable access. Aligning prevention with social justice aims can maximize impact by addressing structural drivers (poverty, discrimination, segregation) and ensuring inclusive monitoring and parity of investment across all prevention levels.

The paper consolidates evidence that social determinants generate and sustain mental health inequalities and that multiple primary prevention strategies—often beginning early in life—can reduce population burden. It sets out a preventive framework and seven actionable, social-justice–oriented recommendations: center social justice; invest in multi-benefit interventions; target critical life-course windows to break intergenerational transmission; prioritize poverty alleviation; strengthen causal inference; implement inclusive longitudinal monitoring; and ensure parity between primary, secondary, and tertiary prevention. Future research should expand high-quality evidence in LMICs, employ modern causal methods, conduct large-scale trials of complex social interventions, and routinely measure mental health outcomes across sectors to inform scalable, equitable prevention.

- Evidence base is disproportionately from high-income countries; context-specific determinants and intervention effects in LMICs are under-studied. - Many associations derive from observational studies with potential confounding, selection, and genetic liability; stronger causal inference and triangulation are needed. - Heterogeneity of measures and methodologies limits synthesis; few large RCTs of complex social/structural interventions exist. - Some school-based and neighborhood interventions show small effects, limited durability, subgroup-specific benefits, or potential unintended harms; rigorous long-term evaluations and economic analyses are scarce. - Monitoring systems often rely on help-seeking samples and may underrepresent disadvantaged/minoritized groups, impeding equitable prevention planning. - Affiliations/data for some author institutions and certain intervention components were not fully detailed in the excerpted text.