Psychology by BogdanovaThe Encyclopedia of Psychology for the Artificial Era

RESEARCH DOSSIER

Community Health Equity and Social Determinants of Well-Being

Abstract. Community health equity concerns the distribution of opportunities to attain and sustain health and well-being across socially differentiated communities. This monograph examines how material resources, institutional rules, neighborhood environments, social relationships, discrimination, access to services, and life-course exposures jointly produce unequal outcomes. A critical narrative synthesis integrates social epidemiology, community psychology, experimental housing mobility, policy and implementation research, social connection evidence, and the World Health Organization's 2025 health-equity assessment. The analysis distinguishes compositional gradients from contextual mechanisms, association from causal identification, service availability from effective access, and average program outcomes from effects on inequity. Housing mobility experiments supply causal evidence for selected place-related outcomes; the broader literature documents robust distributional patterns while retaining limitations involving exposure measurement, selection, spillovers, temporal change, and transportability. Participatory research and community health-worker studies illuminate implementation and collective agency without themselves establishing universal effect sizes. The monograph outlines rigorous multilevel measurement, life-course and intersectional analysis, policy evaluation, and standards for equitable Human–Artificial community infrastructure. It is a structured critical narrative review rather than a registered systematic review or independent meta-analysis.

Keywords: community health equity; social determinants of health; community psychology; social epidemiology; health inequalities; psychological well-being; structural discrimination; neighborhood effects; housing security; environmental justice; multilevel analysis; community participation; public policy evaluation; health service access; cumulative stress; social cohesion; life-course research; intersectionality; natural experiments; health equity intervention; community health workers; implementation science; Artificial Era

The Community Distribution of Health Opportunity

Community health equity is the condition in which people across communities have a fair opportunity to attain health and well-being, and disparities associated with avoidable social disadvantage are progressively reduced. Its empirical subject is the distribution of opportunity, exposure, capabilities, services, and outcomes across places and social positions. Health includes morbidity, functioning, survival, and psychological experience, while well-being includes affective experience, perceived control, interpersonal belonging, and judgments about one's life. These outcomes cannot be represented by one interchangeable index. Braveman et al. (2011) defined health disparities with attention to systematic, plausibly avoidable differences affecting disadvantaged groups, whereas WHO (2025) locates inequities in unequal social conditions and access to power, money, and resources. These conceptual decisions determine what is measured, how inequity is interpreted, and which interventions count as relevant.

The central research question is how the characteristics of communities and their institutions generate, amplify, buffer, or change unequal opportunities to live well. A community may be territorial, such as a neighborhood or municipality; organizational, such as a school or workplace; or relational, such as a network whose members share resources and obligations. Such units rarely have fixed, impermeable boundaries. Their residents are differently situated with respect to wealth, disability, age, ethnicity, employment, family roles, or migration. Accordingly, a community average can conceal severe disadvantage among subgroups even where the overall outcome improves. Equitable intervention assessment requires simultaneously studying the mean level of health, the distribution of benefits, and the processes by which access is allocated.

This dossier examines community-level systems of opportunity rather than treatment protocols for individual disorders or a general account of personal health choices. It includes economic security, housing, environmental exposures, education, social connection, service accessibility, discrimination, collective capacity, and public decision-making insofar as these shape community outcomes. Adjacent research on health-care access and institutional trust can analyze clinical service mechanisms in greater detail; research on income and life evaluation addresses individual appraisal processes. The present synthesis integrates social epidemiology, community psychology, neighborhood experiments, critical policy research, and implementation science into a single research object that can support those neighboring inquiries without duplicating them.

Scientific Questions and the Method of Synthesis

The inquiry is organized around six linked problems: how inequity and well-being should be operationalized; how social gradients arise across levels of analysis; which psychologically mediated pathways can plausibly connect conditions to outcomes; what original empirical research establishes about neighborhood, income, discrimination, and relationships; which policy or community interventions produce identifiable effects; and how future studies can distinguish improvements in population averages from reductions in unjust inequality. The design is a structured critical narrative synthesis of accessible primary studies, peer-reviewed reviews, and institutional scientific reports. Searches targeted seminal works, causal evaluations, theoretical alternatives, methodological contributions, and recent official assessments. There is no registered search protocol, exhaustive screening flow, pooled effect calculation, or novel dataset, so the work does not claim systematic-review or meta-analytic status.

Evidence is assigned to the claim its design can support. A cohort study can estimate a gradient under measured conditions; it ordinarily cannot identify the intervention that would remove the gradient. Random assignment of housing vouchers can identify effects of that offer within a trial, but the intervention combines changes in residence, opportunities, social ties, and sometimes disruption. A natural experiment may improve identification when assumptions about exogeneity and parallel trajectories are examined. A participatory study can reveal overlooked mechanisms, normative concerns, and barriers to adoption, while its representativeness depends on recruitment and positionality. Reviews can summarize a literature, but their conclusions remain bounded by underlying studies, heterogeneity, and publication patterns. These distinctions structure the entire monograph Diez Roux (2001) Bambra et al. (2010) Jagosh et al. (2012).

The evidentiary record combines direct inspection of open research reports where available with bibliographic and abstract-level confirmation for selected restricted articles. Statements below do not imply independent reanalysis of original data or full-text access to every cited study. Numerical claims are confined to results appearing in inspected public sources; otherwise the analysis describes directions of association, identified intervention effects, and design limitations. This method avoids equating a historically influential conceptual framework with experimentally verified mechanisms. It also maintains a difference between normative health-equity goals and descriptive scientific observations, while evaluating the institutional conditions that can make these goals empirically tractable.

Historical Formation of Social and Community Health Research

The contemporary field arose from converging lines of inquiry into class, labor, urban life, medical and social statistics, public health administration, and the consequences of unequal political and economic opportunity. Modern social epidemiology expanded beyond individual risk-factor inventories by explicitly treating social position and contextual conditions as explanatory variables. The Whitehall II study of British civil servants, reported by Marmot et al. (1991), showed a marked social gradient in morbidity among employment grades and encouraged models linking occupational hierarchy, conditions of work, and health. Adler et al. (1994) argued that socioeconomic status should be investigated as a gradient across the social scale rather than used merely as a confounding covariate or a binary poverty threshold. Neither study implies that job grade or income alone constitutes the complete causal mechanism.

By the turn of the twenty-first century, researchers increasingly distinguished the properties of places from the composition of their populations. Diez Roux (2001) discussed ecological, contextual, and multilevel approaches to neighborhood health effects. Krieger (2001) developed an ecosocial perspective centered on embodiment: social and ecological conditions become expressed through lived biological trajectories. Berkman et al. (2000) connected social networks and social integration to behavioral, psychological, and physiological pathways, drawing conceptual distinctions among forms of connection. Together these programs enlarged the explanatory frame from a one-person exposure–outcome relationship to a system of mutually interacting conditions, relationships, and institutions.

The World Health Organization's Commission on Social Determinants of Health gave international institutional form to the field's concern with unequal life chances. Marmot et al. (2008) summarized its 2008 conclusions and recommendations, and the WHO (2010) framework later distinguished structural determinants from intermediary circumstances and health-system pathways. The WHO (2025) world report revisited uneven progress and established a renewed agenda organized around social determinants of equity. This history involves multiple intellectual traditions, including community psychology, social medicine, geographical health research, public policy, and antidiscrimination scholarship. Their distinctive methods should remain visible: a WHO recommendation is an institutional normative and policy judgment grounded in evidence; a quasi-experiment is an identification strategy; an ethnography describes situated organization and experience.

Definitions, Constructs, and Outcome Domains

A social determinant of health is a nonmedical condition or institutional arrangement that shapes the distribution of exposure, protection, and opportunity across the life course. The ODPHP (Healthy People 2030) framework groups U.S. determinants into economic stability, education access and quality, health-care access and quality, neighborhood and built environment, and social and community context. These categories describe overlapping domains, not isolated causal compartments. Material insecurity can increase residential mobility, interrupt schooling, constrain care use, and destabilize relationships. Conversely, educational opportunity and reliable transportation may change employment options and access to organized support. The analytical challenge is to represent these interdependencies while identifying specific exposures that can be altered by intervention.

Health inequality refers descriptively to any measured difference in health status across populations; health inequity refers to differences that are systematic, plausibly avoidable, and connected with social disadvantage or injustice. Equity analysis therefore requires an explicit normative account of what counts as avoidable and unfair, alongside a transparent empirical account of the magnitude and pattern of observed differences Braveman et al. (2011). Distributional claims must specify whether groups are defined by income, geography, disability, ancestry, gender, age, or some intersection of these positions. A widening relative risk can coexist with improving absolute outcomes, and a narrowing mean difference can coexist with higher hardship for a vulnerable subgroup. Reporting only one disparity metric can materially alter policy interpretation.

Psychological well-being adds distinct measured constructs: life satisfaction, positive and negative affect, perceived social support, loneliness, sense of belonging, safety, autonomy, and stress-related functioning. They are assessed through self-report scales, experience sampling, behavioral indicators, and administrative outcomes that capture different things. A measure of clinic visits is not a measure of experienced well-being; lower utilization may reflect good health or unmet need. A neighborhood cohesion scale captures residents' assessments and must be tested for measurement invariance across cultural or linguistic groups. Outcomes should be prespecified by level—person, household, organization, neighborhood, region—and by timescale—immediate experience, longitudinal adaptation, and cumulative life-course consequence.

Levels of Causation and the Problem of Context

Community health research examines causal systems spanning individual histories, interpersonal networks, organizational rules, built environments, and wider political economies. At the individual level, coping skills, beliefs, physiology, and behaviors can mediate exposure. At the household level, finances, caregiving demands, crowding, and stability affect daily life. At the neighborhood level, public transport, violence, air quality, social trust, parks, employment accessibility, and housing costs vary. At the institutional level, eligibility criteria, financing, enforcement, and discrimination determine which resources reach whom. A multilevel conceptualization specifies both contextual exposures and the allocation processes through which different people encounter them. These levels interact, and selecting one as primary is a research design choice, not an ontological claim about people.

The distinction between compositional and contextual explanations is essential. An area with high disease prevalence may contain more residents who already faced disadvantage before moving there; this composition can produce geographic association without an independent effect of the place. Neighborhood contexts may also alter exposure above and beyond residents' prior characteristics. Reverse causation occurs when ill health changes employment, income, or residential options. Selection occurs when unobserved preferences or constraints influence both location and outcome. Diez Roux (2001) made these challenges central to neighborhood epidemiology; Cummins et al. (2007) further emphasized relational and mobile understandings of place, in which environments are constituted through social connections, travel, institutional reach, and movement over time.

The appropriate counterfactual therefore depends on the intervention: what would a person or community have experienced under another housing policy, a different level of public investment, a revised transportation network, or changed service eligibility? No single statistical adjustment can remove all selection bias, and randomization is rarely available for fundamental structural conditions. A credible study triangulates linked administrative data, longitudinal measurement, policy timing, qualitative accounts, and, where possible, exogenous variation. It also tests spillovers: a neighborhood-level program can affect nearby areas, and improved access for one group may change access for others. These effects are scientifically central because communities are open, connected systems.

Theories of Embodiment, Social Networks, and Collective Capacity

The ecosocial perspective proposes that the social distribution of exposure becomes embodied in patterns of health and illness over time Krieger (2001). It directs attention to historical power arrangements, pathways of exposure, susceptibility, and accountability. This theoretical contribution is especially important when a reductionist account attributes social differences solely to individual choice or stable biological traits. The theory generates research questions about the timing of exposure, physiological responses, cumulative disadvantage, and institutional production of vulnerability. Its explanatory claims require direct evidence for each proposed pathway; invoking embodiment does not independently identify the size of a causal effect.

Network theories specify how social integration, support, access to information, collective norms, and reciprocal obligations affect health-related opportunities. Berkman et al. (2000) outlines psychological, behavioral, and physiological pathways that can connect networks with outcomes. Holt-Lunstad et al. (2010) synthesized observational evidence associating social relationships with survival, supporting the substantive importance of connection, while leaving room for residual confounding and differences among network measures. Community ties may distribute assistance during illness, support employment transitions, or provide protection during crises. They can also reproduce exclusion and unequal access. Consequently, strong bonding within a group does not by itself imply equitable access for everyone residing within or adjacent to that group.

Collective efficacy combines social cohesion with expectations that people will act on shared concerns. In Chicago, Sampson et al. (1997) found that neighborhood collective efficacy was negatively associated with variation in violence in a multilevel observational analysis. This is a substantive example of a community-level psychological construct, measured across 343 neighborhoods and 8,782 respondents in the reported study. The result is not a randomized test of an initiative to increase efficacy. Its value lies in showing how contextual social expectations can be operationalized and linked to outcomes beyond a simple count of available organizations. Subsequent designs must determine when collective resources can influence health opportunity and whether local benefits reach marginalized residents.

Income, Employment, and the Security of Everyday Life

Income and wealth alter the options through which households secure shelter, food, mobility, caregiving, education, and participation. The psychological consequences are shaped by both the absolute availability of resources and the predictability of their future supply. Debt pressure, precarious work, and sudden income loss can constrain attention, reduce perceived control, and force repeated tradeoffs between necessities. Adler et al. (1994) placed socioeconomic gradients across the social scale at the center of psychological explanations of health, suggesting that studying only extreme poverty would miss important variation. A complete community measure therefore includes resource distribution, household-level material hardship, access to benefits, conditions of work, and the capacity to use local opportunities, rather than treating the neighborhood's average income as a direct psychological measure.

Administrative data can reveal large spatial and socioeconomic variation. In the United States, Chetty et al. (2016) analyzed income and life expectancy over 2001–2014 and documented systematic differences across income levels and places, including heterogeneity in temporal trends and geography. The analysis is population-scale observational evidence; it does not establish that raising a given household's income by a specified amount would cause a corresponding increase in life expectancy. Place, health behavior, migration, institutional investment, baseline health, and selection are entangled. Research should separate cross-sectional gradients, within-person changes, and causal effects of specified income or employment policies, preferably using multiple outcomes and explicit counterfactuals.

Employment can provide income, insurance, social identity, time structure, relationships, and exposure to hazards. Thus the health consequences of jobs depend on their organization, security, autonomy, pay, schedules, and accessibility. A program increasing labor-force participation may leave well-being unchanged if the new work is dangerous, insecure, or incompatible with caregiving. Conversely, social protection may support health by reducing volatility and enabling access to treatment, education, and stable housing. Distributional assessments must report who gains employment, what type of employment changes, whose unpaid labor expands, and whether the intervention changes income gradients. These are empirically distinct evaluative questions, and conclusions cannot be transferred wholesale from one labor market or welfare system to another.

Housing, Neighborhood Exposure, and Residential Mobility

Housing is simultaneously a material resource, a setting for social relationships, and a pathway into broader opportunity structures. Affordability, safety, crowding, insulation, stability, accessibility, and proximity to transport shape sleep, exposure to stressors, social interaction, and the ability to maintain employment or treatment. Evans (2003) reviewed links between the built environment and psychological distress, including crowding and environmental noise, while recognizing limitations in attributing effects to particular design elements. Residential tenure and the threat of displacement can also influence the anticipation of loss and capacity to plan. These mechanisms require measurements more precise than a neighborhood poverty rate, including household crowding, eviction risk, neighborhood quality, and duration of exposure.

The distinction between an intervention offering housing mobility and the total effect of living in a different neighborhood is crucial. The Moving to Opportunity program randomized housing voucher offers among eligible low-income families, offering a stronger causal design than purely observational neighborhood comparisons. Ludwig et al. (2012) found long-term improvements in adult physical and mental health and subjective well-being associated with moving from high-poverty to lower-poverty areas, even where adult economic self-sufficiency did not improve. The experimental treatment was an offer embedded in complex housing conditions; compliance, family circumstances, and the composition of destination neighborhoods matter when interpreting effects. These outcomes cannot be treated as universal proof that any residential move improves well-being.

Housing opportunity develops over time. Chetty, Hendren, and Katz (2016) used long-term tax outcomes in the MTO setting to show that children who moved to lower-poverty neighborhoods at younger ages experienced more favorable adult educational and earnings outcomes, while effects differed for older movers. The age-related variation is consistent with a role for exposure duration and developmental timing, although the voucher intervention also changes schools, peers, routines, and social support. The findings invite an integrated account of cumulative opportunity and the potential disruption of established ties. Policy design should therefore measure displacement costs and social continuity alongside destination quality, safety, and opportunity.

The Built Environment, Environmental Exposure, and Spatial Inequality

Physical environments distribute both burdens and supports. Air pollution, noise, heat, unsafe traffic, inaccessible pathways, poor building maintenance, and inadequate sanitation may influence physiological health and psychological functioning. Parks, shade, safe public space, accessible public transportation, and reliable utilities can create opportunities for physical activity, social participation, rest, and service access. Evans (2003) provides an early synthesis of the built environment and mental health. WHO (2025) treats environmental exposures and structural conditions as part of the distributional production of inequity. The relevant object is an exposure pathway, with measured timing and dose where feasible, rather than a simple assertion that some places are inherently healthy or unhealthy.

Spatial exposure measurement presents substantial problems. Administrative districts do not necessarily correspond to daily activity spaces; residents cross municipal boundaries for work, education, shopping, worship, and treatment. A point-in-time home address may be a poor proxy for cumulative environmental exposures, especially among migrants, unhoused people, children with multiple caregivers, and mobile workers. Cummins et al. (2007) argues for a relational interpretation of place, including the networks and flows that connect locations. Studies can improve measurement through geocoded residential histories, travel diaries, neighborhood audits, sensor data, and interview-based reconstructions, provided consent and privacy are protected.

A causal policy evaluation must distinguish exposure reduction from displacement of exposure. A neighborhood improvement can raise housing prices and push the residents facing greatest need to another area, yielding improved local averages without equitable individual gains. Similarly, a park may be physically present but unusable because of disability barriers, fear of violence, hours, cost, or social exclusion. Equity-sensitive evaluation tracks access, actual use, displacement, and long-term residence. It also distinguishes environmental hazard removal from residents' subjective sense of safety: both are important, but they call for different measurements and may respond differently to the same intervention.

Education, Developmental Opportunity, and Cumulative Advantage

Education operates through learning, credentials, employment options, health literacy, social networks, and participation in public institutions. Community-level educational opportunity depends on school quality, inclusion, transportation, financing, safety, and family access to supporting resources. The ODPHP (Healthy People 2030) framework identifies education access and quality as a domain of the social determinants of health. School achievement, however, cannot serve as a universal measure of health equity, and education may be simultaneously an outcome of earlier material conditions and a mediator of later opportunity. Causal diagrams should specify the developmental period and whether schooling precedes or follows the exposure being studied.

Children and adolescents experience institutional conditions at sensitive developmental moments. The MTO evidence Chetty, Hendren, and Katz (2016) illustrates how age at relocation changes the interpretation of long-run outcomes. A policy providing safer or more resource-rich surroundings during early childhood may generate effects that cannot be replicated by a later intervention of similar duration. Yet mobility can also interrupt friendships, schooling, and family support, creating heterogeneity across age groups. Longitudinal research should model exposure histories, school transitions, parental employment and stress, disability accessibility, and protective relationships without presuming a single universal developmental window.

Intergenerational transmission can occur through material assets, housing and education access, institutional discrimination, neighborhood segregation, and social capital. Because parents' opportunities shape children's contexts, purely individual explanations of educational or psychological attainment can conceal the allocation of environmental support. Studies can examine sibling comparisons, cohort transitions, school boundary changes, and policy reforms; every method faces specific threats to inference. The equity question includes whether interventions reduce persistent inequalities in opportunity across generations rather than only improving average attainment among those already best positioned to use new resources.

Structural Discrimination, Segregation, and Institutional Power

Structural discrimination refers to rules, institutional practices, and historical arrangements that systematically distribute resources and burdens along socially salient group boundaries. It can operate through housing finance, school assignment, hiring, policing, land-use policy, insurance, and care delivery without requiring each participant to express deliberate prejudice. Gee and Ford (2011) identifies segregation, immigration policy, and intergenerational processes as important objects of study. Bailey et al. (2017) synthesizes historical and contemporary evidence for structural racism as a fundamental source of health inequity in the United States. These frameworks identify candidate mechanisms and institutional sites of intervention; effects must still be estimated through appropriate designs and defined populations.

At the psychological level, discrimination can influence perceptions of safety, trust, expected treatment, self-protection, stress exposure, and willingness to seek services. In analyses of Black–White differences in physical and mental health, Williams et al. (1997) examined socioeconomic position, stress, and reported discrimination and found that these dimensions were related to health disparities. Such findings cannot alone disentangle direct discrimination effects from unobserved historical and material disadvantage. They support a research program combining individual-reported experiences with independent institutional indicators, such as segregation patterns, access rules, resource allocation, and administrative treatment across groups.

Equity studies must also address intersectional and within-group heterogeneity. Disability, gender, racialization, age, migration status, and class shape exposures together; a single demographic label may obscure very different circumstances. Institutional data should be disaggregated where ethically and legally justified, with meaningful sample sizes and safeguards against reidentification or stigmatizing inference. A policy could reduce aggregate disparities while worsening barriers for disabled rural residents, undocumented migrants, or people facing intersecting disadvantages. Such cases call for subgroup impact analysis, participatory interpretation, and explicit attention to who has decision-making authority over the policies being evaluated.

Social Connection, Trust, Safety, and Relational Inequality

Community ties shape the circulation of information, emotional support, practical aid, social obligations, and opportunities for collective problem solving. An ecological analysis by Kawachi et al. (1997) linked state-level social capital and income inequality to mortality, a pattern requiring caution against individual-level causal inference. Berkman et al. (2000) distinguishes networks, social integration, and potential health pathways, while Holt-Lunstad et al. (2010) consolidates observational evidence linking social relationships to mortality. A coherent explanation should differentiate structural network characteristics, subjective support, loneliness, belonging, and actual assistance during stress or illness. These constructs respond to different intervention levers. Expanding the number of group activities may increase opportunities for contact without changing trust or decreasing unwanted isolation, and a highly cohesive community can still restrict access for outsiders.

Safety has objective and perceived components. Community violence, harassment, insecurity in public spaces, and unreliable institutional protection can limit outdoor activity, commuting, school attendance, and social interaction. Sampson et al. (1997) associated collective efficacy with neighborhood violence in Chicago, offering a testable social mechanism involving shared expectations and willingness to intervene. Yet safety interventions must be assessed for unintended consequences: expanded surveillance or enforcement may redistribute fear, barriers, or harms across different groups. Evaluators should track both experienced safety and measurable exposure to violence, including outcomes for populations historically excluded from decision making.

Social cohesion is also linked to material organization. Neighborhood organizations need stable funding, physical meeting places, accessibility, leadership continuity, and fair procedures to translate goodwill into reliable services. Population turnover, precarious work schedules, and language barriers may constrain participation. Digital channels may improve information exchange while exposing residents to surveillance, exclusion, and platform dependence; their effectiveness should be evaluated rather than presumed. The essential question is whether collective relationships expand people's actual capacity to secure resources, influence decisions, and sustain well-being over time. Community connectedness is best understood as one interacting component of structural opportunity.

Stress, Control, Adaptation, and Psychobiological Pathways

Structural adversity becomes psychologically consequential through repeated encounters with uncertainty, restricted options, interpersonal threat, and the demands of negotiating institutions. Psychological stress is a response involving appraisal, affect, behavior, and physiological regulation; chronic exposure can sustain anticipatory vigilance even when a discrete threatening event has ended. McEwen (1998) articulated allostasis and cumulative physiological costs of repeated or prolonged activation of stress mediators. This framework proposes a plausible bridge from social environments to biological outcomes, but it does not allow researchers to infer a person's physiological burden directly from neighborhood income or a single questionnaire score. Exposure, appraisal, and biological response require independent measurement.

Perceived control, self-efficacy, and coping capacities are influenced by opportunities available to act. A person can accurately perceive limited control when employment, transportation, benefits, or housing are institutionally constrained. Consequently, an intervention that teaches coping without changing the exposure may improve some subjective outcomes while leaving the main cause of hardship intact. Community psychology examines how access to trusted assistance, collective action, institutional responsiveness, and material security modifies the relationship between exposure and adaptation. Psychological mechanisms should be modeled as mediators or moderators with explicit time ordering and tested against alternative pathways, not invoked as explanatory labels after an observed association.

Biomarkers such as blood pressure, endocrine responses, inflammatory measures, and sleep indices can complement well-being measures, but they are sensitive to measurement conditions, medication, circadian patterns, age, and co-occurring illness. Researchers should report their sampling protocol and avoid interpreting one biological variable as a comprehensive indicator of stress. Repeated measures and prospective designs can clarify whether changes in exposure precede physiological change. Health outcomes can also alter the conditions of living, producing reverse causation. The scientifically informative model therefore contains feedback: health affects employment and service need while structural conditions shape health trajectories.

Measuring Community Inequity and Well-Being

Operationalizing inequity begins by choosing a population, outcome, comparison, and estimand. Absolute differences in prevalence, relative ratios, concentration measures, distributional indices, and opportunity-based indicators describe different features of inequality. The selection should follow the research question, not which metric produces the most dramatic result. A policy may reduce absolute risk for everyone yet increase a relative gap if advantages accumulate faster among the already privileged. Equally, a narrow group contrast can mask severe hardship within both groups. Braveman et al. (2011) offers a justice-oriented definition that requires empirical specification of social disadvantage, while WHO (2025) emphasizes upstream distributional mechanisms. A transparent report presents baseline values, relevant confidence intervals, and subgroup trajectories.

Neighborhood indicators need careful construction. Census poverty, rental burden, housing code violations, distance to public transport, school resources, environmental measurements, facility capacity, and service wait times may be combined in an index, but combination does not confer construct validity automatically. An index's weighting can encode implicit judgments about importance; missing data can systematically exclude informal settlements, displaced households, or residents without stable addresses. Diez Roux (2001) describes the logic of contextual and multilevel investigation, and Cummins et al. (2007) argues against treating administrative geography as a complete representation of lived place. Sensitivity tests should vary spatial boundaries, exposures, time windows, and model specifications.

Psychological outcomes create additional challenges of linguistic translation, scale invariance, response styles, and differential access to survey participation. Questions about happiness, safety, discrimination, or belonging may carry different social meanings across cultural groups, and identical responses may reflect unequal material conditions. A rigorous design examines reliability and factor structure, tests measurement invariance where feasible, and adds interviews or participatory interpretation where quantitative categories miss local experiences. Outcome selection should involve affected communities while preserving prespecified hypotheses and independent evaluation. The result is a portfolio of complementary indicators rather than a single total score purported to measure the health of a community.

Observational Gradients, Confounding, and Multilevel Modeling

Cross-sectional and longitudinal observation remains indispensable for studying population health gradients that cannot be randomized. The Whitehall II cohort Marmot et al. (1991), U.S. socioeconomic analyses Adler et al. (1994), and income–life expectancy research Chetty et al. (2016) provide complementary descriptions across occupational strata, socioeconomic processes, and national administrative data. Their strength lies in identifying persistent structured variation and generating testable mechanisms. Their principal limitation for causal interpretation is that income, residential history, education, health status, and institutional treatment influence one another. Statistical controls can remove measured confounding but may introduce bias when a variable is itself caused by exposure or shares causes with outcome.

Multilevel models can separate measured person-level and area-level associations, estimate clustering, and test variation in effects across communities. Random intercepts and slopes are modeling devices rather than causal proof. Spatial autocorrelation and shared service infrastructure may violate assumptions of independent observations. Studies must specify whether area characteristics describe actual exposures for each resident, proxies for social position, or aggregated individual outcomes. Ecological inference is particularly hazardous: a strong association among neighborhoods can coexist with a weaker or even reversed relationship among individuals. Diez Roux (2001) remains foundational for distinguishing these designs and interpreting contextual conclusions.

Longitudinal research improves temporal ordering, yet attrition, residential selection, health-related mobility, and changing measurement systems remain difficult. Individuals who leave a neighborhood after economic improvement or illness may differ systematically from those who stay. Residential histories can reveal cumulative exposure but require addressing unequal follow-up. Time-varying confounders affected by earlier exposure challenge naive regression approaches; marginal structural models, natural-experiment designs, and targeted sensitivity analysis may be appropriate when their assumptions are supportable. Robust conclusions emerge from triangulation across datasets and designs, with clearly stated estimands and limitations of transportability.

Natural Experiments, Housing Mobility, and the Meaning of Causal Effects

Random assignment creates a valuable source of variation when studying place-based conditions. Moving to Opportunity allocated voucher opportunities to eligible low-income families, allowing analysis of the effects of an offer to move. Ludwig et al. (2012) reported long-term improvements in adult well-being and selected physical and mental health outcomes among households experiencing reduced neighborhood poverty. This result supports a causal role for a housing-mobility intervention in the studied population. It does not isolate a single mechanism: changes in schools, safety, social networks, housing quality, travel time, perceived stigma, and administrative support could all contribute. Interpretation must distinguish intention-to-treat effects from estimates conditional on moving.

The follow-up study Chetty, Hendren, and Katz (2016) linked MTO exposure to later education and earnings, finding beneficial long-term effects for children who moved when younger and different results for adolescents. The intervention's age dependence strengthens the case for considering developmental timing and duration of exposure. Nevertheless, the mechanisms of school continuity, family transition, community connection, and neighborhood composition remain partly entangled. Research designs can test mediator hypotheses with additional measurements, but post-randomization mediation requires assumptions beyond the original treatment randomization. The strongest scientifically defensible claim concerns effects of a specified housing opportunity in a specified program context.

Quasi-experiments involving housing policy, school assignment, transit improvements, environmental regulation, or benefit eligibility may extend causal investigation beyond one trial. Their credibility turns on assumptions such as comparable untreated trends, discontinuity validity, exclusion restrictions, and the absence of anticipatory response. Researchers should include negative controls, falsification exercises, heterogeneous effects, and spillover checks where possible. In equity evaluation, the treatment may itself change migration or local prices; the target estimand must therefore specify whether the objective is improvement among original residents, among current residents, or at the whole-region level.

Evidence on Discrimination, Relationships, and Community Mechanisms

Different evidentiary traditions support different parts of a structural account. Williams et al. (1997) analyzed socioeconomic status, stressful experiences, discrimination, and health differences; the findings show why demographic comparisons that omit social mechanisms can be misleading. Gee and Ford (2011) and Bailey et al. (2017) extended attention to institutional segregation and structural racism. These contributions vary in design, from primary observational analysis to conceptual and evidentiary synthesis. Their convergence justifies a research focus on discriminatory systems, but it does not supply a single causal parameter called the effect of structural racism. Specific measures of exclusion, policy, environmental burden, and service allocation are needed to estimate particular pathways.

Social relationship studies likewise document important patterns without identifying an interchangeable intervention. Holt-Lunstad et al. (2010) combined observational studies on relationships and survival, while Sampson et al. (1997) used multilevel data to relate collective efficacy to neighborhood violence. Both imply that the relational conditions of communities matter, but an intervention that increases meeting frequency is not necessarily the same as one that creates reliable support, reduces violence, or changes trust. Cross-sectional perceptions can be correlated with outcome reports because of shared response tendencies. Stronger mechanism testing includes behavioral measures of assistance, longitudinal network changes, area-level safety data, and theoretically justified mediating variables.

A major synthesis problem is that relationships and institutions co-evolve. A safe, walkable neighborhood facilitates social contact; social networks can mobilize to secure safer streets; an employment shock can reduce resources and disrupt ties. In such systems, the effect of a single component may depend on the others. Experiments designed to change only one element may underestimate benefits of coordinated investment or overestimate success if supportive institutions already exist. Measurement of interdependence, variation in treatment intensity, and cross-community spillovers is required for theoretically coherent conclusions.

Causal Identification, Replication, and Evidence Grading

Evidence strength is tied to a precisely defined claim. For an income–mortality association, national administrative data may be compelling descriptive evidence while remaining insufficient to isolate causal income effects. For the effect of a particular voucher offer, randomized assignment can support stronger causal claims, with the usual concerns about treatment uptake, attrition, and generalization. For the distribution of experienced discrimination, surveys provide indispensable evidence but depend on wording, trust, and differential response. For program sustainability, implementation research provides a different evidence domain from clinical effectiveness. A simple universal ranking of designs would obscure the estimands each design can validly address.

Replicability requires more than statistical significance. Investigators should publish sampling frames, construct definitions, geocoding procedures, analytic decisions, model specifications, missing-data handling, and uncertainty estimates. Policy effects need replication across jurisdictions with different service systems and baseline inequalities. The strength of the MTO evidence lies partly in longitudinal follow-up and independently verifiable intervention assignment; applying that estimate to a new housing program still requires comparing eligibility, destinations, rental markets, social supports, and developmental ages Ludwig et al. (2012) Chetty, Hendren, and Katz (2016). Reviews of broad social-determinant interventions, including Bambra et al. (2010), demonstrate why study heterogeneity and differences in outcome timing complicate synthesis.

Evidence grading should also inspect funding, interests, community participation, surveillance burdens, and whose outcomes are omitted. An intervention may be highly effective at increasing institutional enrollment but fail to reduce unmet needs. A statistically detectable gain may be trivial in lived circumstances, while a modest average effect may be very important for a subgroup facing severe deprivation. The appropriate scientific judgment therefore integrates internal validity, measurement quality, effect magnitude, uncertainty, distributional impact, practical consequences, and conditions of implementation. Such a standard helps prevent policy enthusiasm from outrunning the evidence.

Policy Instruments for Upstream Health Equity

Policies that change housing affordability, employment protection, education funding, transportation, environmental regulation, or social benefits modify the conditions through which health opportunity is distributed. Their impact depends on implementation, eligibility, administrative burden, and interaction with local institutions. Marmot et al. (2008) placed action on the conditions of daily life, unequal distribution of resources, and measurement of progress at the center of the WHO Commission's recommendations. WHO (2025) revisited these upstream mechanisms and organized a renewed action agenda across policy domains. These reports provide policy synthesis and normative direction; they do not imply that every proposed reform has already been tested in a randomized trial.

The review by Bambra et al. (2010) examined systematic reviews of interventions acting on wider social determinants, including housing, work, transport, education, and access to services. It illustrates both the potential value of upstream interventions and the incompleteness of comparative causal evidence. Interventions differ in institutional setting, reach, intensity, lag until health outcomes appear, and likely distributional consequences. Health equity requires evaluating whether a program changes opportunity for populations facing greatest need. An intervention that increases aggregate service availability may preserve inequity if uptake remains highest among advantaged households. Researchers should therefore report reach, barriers, subgroup effects, and changes in absolute and relative disparities alongside average benefits.

Implementation also involves institutional tradeoffs. Zoning reform may expand housing supply yet displace existing tenants when protections are weak; transport changes can improve regional connectivity while altering neighborhood safety and rents. Environmental rules may shift industry rather than eliminate exposure. Social benefits can improve material security but create complicated application burdens. These possibilities are testable through longitudinal policy evaluation, administrative data, resident accounts, and spatial spillover analysis. The key standard is to evaluate the real distribution of gains and costs under a specified policy rather than its stated intention alone.

Community Participation, Knowledge, and Decision-Making Power

Community-based participatory research treats affected communities as partners in defining questions, selecting methods, interpreting findings, and planning action. Wallerstein and Duran (2010) identifies partnership and the redistribution of knowledge production as pathways for improving the relevance and translation of interventions. Jagosh et al. (2012) reviewed heterogeneous participatory research and emphasized relational processes that can produce context-sensitive benefits. These contributions suggest why researchers may overlook locally significant harms when communities are treated only as sampling locations. The substantive promise is greater interpretive validity, more legitimate priority setting, and interventions suited to existing local institutions.

Participation needs operational standards. Researchers should document how community representatives were selected, whose voices remain absent, what decision rights they hold, how disagreements are resolved, and whether participation is compensated. A small advisory panel cannot automatically represent an entire city, particularly when internal socioeconomic and cultural differences are large. Participatory processes also take time, can reproduce local power hierarchies, and may be constrained by research funding or institutional approval. Their value should be assessed through measurable effects on study relevance, retention, implementation, trust, and sustainability, while acknowledging that a participatory process is not itself proof of clinical benefit.

Community control over data is a central research concern. Information on disability, income, migration, discrimination, and location can expose participants to stigma, administrative scrutiny, or surveillance. Ethical data governance must specify consent, access, retention, deidentification limitations, secondary uses, and the return of findings to participants. Investigators can complement quantitative indicators with interviews and participatory mapping, provided individual privacy is protected. The research process should create genuine capacity for affected communities to contest interpretations and exercise agency over how findings are used, while keeping the scientific evaluation of causal claims transparent and open to independent scrutiny.

Community Health Workers and Social-Care Integration

Community health workers and peer navigators may bridge institutional and community contexts by providing outreach, culturally and linguistically accessible information, navigation, and sustained practical assistance. Viswanathan et al. (2010) reviewed the outcomes and costs of community health-worker interventions and found a diverse evidence base varying by target population, activity, and evaluation design. Such programs cannot be interpreted as one uniform treatment: their effects depend on training, supervision, scope, local trust, compensation, integration with services, and the availability of resources to which people are referred. Screening for need may identify a barrier without enabling a person to overcome it.

The U.S. National Academies' National Academies (2019) framework describes five complementary health-care-sector activities: awareness of social risks and assets, adjustment of clinical services, assistance connecting patients with resources, alignment with community assets, and advocacy to strengthen population conditions. The report explicitly identifies gaps in robust evaluation of many integration activities. These categories are useful for distinguishing identification from effective response. For example, a transportation questionnaire may improve awareness, but it will not reduce missed appointments if affordable transport remains unavailable. Similarly, a referral platform can increase recorded referrals without increasing successful connections.

Service programs should be evaluated across the full pathway from identification to benefit: eligibility, outreach, enrollment, accessibility, actual receipt, quality, outcomes, and continuation after funding ends. Measures should capture staffing workload, costs, language access, disability accommodation, referral completion, and whether burdens shift toward unpaid family caregivers. A community may improve its aggregate referral rate while leaving the most marginalized population outside the program. Equity-sensitive service integration consequently requires denominator-based measurement, disaggregated follow-up, and attention to structural resources beyond the individual clinician's authority.

Implementation, Sustainability, and Policy Learning

Evidence of effectiveness in one research site does not ensure that an intervention can be adopted and sustained elsewhere. Greenhalgh et al. (2004) examined diffusion of innovation in service organizations and emphasized organizational context, leadership, implementation processes, and interactions among actors. Proctor et al. (2011) distinguished implementation outcomes, including acceptability, adoption, appropriateness, feasibility, fidelity, cost, penetration, and sustainability, from clinical or service outcomes. This distinction is essential for community health equity. A program may be accepted and faithfully delivered yet fail to change an inequitable distribution of outcomes; alternatively, a program with promising efficacy may remain inaccessible because of institutional or geographic barriers.

Program evaluation should specify an implementation theory and its causal assumptions before data collection. Investigators can compare sites with differing capacities, measure adaptation and fidelity, track funding, and examine whether organizational changes survive staff turnover. Mixed-methods designs are particularly valuable when an intervention works through several components whose relative importance varies by setting. Qualitative interviews may reveal that a nominally accessible service requires internet access, an unaffordable journey, or documentation some residents lack. Quantitative follow-up can then estimate the reach and outcome consequences of those barriers.

Equity should be built into the intervention's objective function. Researchers need to distinguish total population benefit, reduction of disadvantaged groups' absolute burden, reduction of intergroup differences, and improvement in the worst-off subgroup. These objectives can converge but need not do so. A transparent policy evaluation can present several metrics and a prespecified priority rule, while leaving normative judgments explicit. Sustainability includes institutional accountability, funding stability, local ownership, maintenance, workforce protection, and periodic reevaluation when population composition changes. Cumulative learning requires publication of null results and adverse effects as well as successful outcomes.

Cultural Variation, Mobility, and the Limits of Generalization

Community health research travels across settings only when the mechanisms and relevant conditions are understood. The experience of belonging, perceived discrimination, institutional trust, and satisfaction can vary with language, migration, legal status, histories of dispossession, and cultural conceptions of a good life. Research instruments developed in one national or urban population may misrepresent the priorities of rural communities, Indigenous peoples, displaced families, or groups with limited institutional access. Measurement equivalence should be tested rather than assumed. Community participation can improve construct interpretation, but local qualitative understanding and quantitative comparison perform different scientific functions.

Population movement makes exposure trajectories difficult to reconstruct. Migrants may retain meaningful relationships across national borders while facing new constraints on housing, employment, documentation, and access to care. People with disabilities may encounter different environmental obstacles despite living in the same geographic district as nondisabled neighbors. Aging residents may experience a change in access when transport systems, housing, or social support become less suitable. A credible design measures concrete functional access and the duration of relevant exposures instead of simply assigning all residents a uniform community score. The relational account of place developed by Cummins et al. (2007) provides a conceptual foundation for this approach.

Evidence generated in the United States or the United Kingdom should not be presented as universally transferable. Marmot et al. (1991), Chetty et al. (2016), and the MTO studies illuminate particular labor-market, policy, and housing contexts. Their findings create hypotheses for other settings, not fixed parameters. The WHO (2025) report draws on international cases and emphasizes the heterogeneity of structural conditions. Cross-country comparison requires harmonized definitions, attention to demographic selection, and explicit institutional variables. External validity depends on which components of an exposure–mechanism–outcome relationship are stable, which vary, and what target population is of interest.

Community Health Equity in the Artificial Era

Digital systems increasingly mediate access to employment, benefits, education, transportation, social services, and information. In the Artificial Era, community opportunity must therefore be studied alongside algorithmic allocation, interface accessibility, digital literacy, platform dependence, and the ability to challenge automated decisions. The relevant psychology concerns people's understanding of services, trust, perceived agency, cognitive burden, social connection, and experience of exclusion. Artificial systems can act as research instruments or as components of institutional service delivery, but these roles require distinct evaluation. A conversational interface that provides useful information does not necessarily create an affordable or available service.

A rigorous Human–Artificial community study would prespecify functional outcomes such as correct comprehension of eligibility requirements, successful benefit uptake, reduced administrative burden, accessible translation, safe navigation, and sustained access for underserved populations. It would compare alternatives including human assistance, hybrid support, and no additional assistance, measuring error, privacy cost, inequitable refusal rates, and the transfer of usable knowledge. Experimental evaluation should stratify by disability accommodation, language, device availability, literacy, and baseline trust. Where an artificial system makes recommendations or exercises delegated authority, governance should document responsibility, appeal, auditability, and the consequences of system failure.

The conceptual horizon From Homo to Artificial invites research into new forms of coordination and knowledge-bearing systems within community life. The existence of artificial systems as participants in information processes does not establish that a particular model possesses psychological well-being, biological stress responses, or human-like health needs. These constructs demand separate operational definitions if applied to artificial systems. For the present dossier, human population health and equitable institutional opportunity remain the empirical endpoints. Claims about artificial agency, continuity, or future collective organization are theoretical extensions requiring additional evidence, and they are distinguished from established research on human social determinants.

Critical Scientific Synthesis and Unresolved Problems

The principal evidence converges on the importance of social distribution. Observational studies establish persistent socioeconomic and spatial gradients Marmot et al. (1991) Chetty et al. (2016). Social epidemiology specifies mechanisms through resource access, embodiment, relationships, and institutional arrangements Krieger (2001) Berkman et al. (2000). Research on discrimination expands the account to systematic exclusion Gee and Ford (2011) Bailey et al. (2017). Randomized housing-voucher evidence demonstrates causal effects of a specified intervention on selected long-term outcomes Ludwig et al. (2012) Chetty, Hendren, and Katz (2016). Intervention and implementation literatures show both plausible remedies and major gaps in robust, transportable evaluation Bambra et al. (2010) Proctor et al. (2011). These bodies of evidence support a multilevel research agenda without reducing complex outcomes to one determinant.

Several uncertainties remain central. The relative contribution of pathways is often unidentified because structural exposures covary; natural experiments cannot isolate every mediator. Neighborhood boundaries and daily activity spaces rarely coincide. Residential selection, missing data, spillovers, adaptation, and policy-induced migration can distort estimates. Long latency complicates judgments about intervention success. Equity measures embed defensible but contestable normative choices. The effects of digital mediation on access and trust are especially dynamic. Future studies should combine rigorous policy identification with spatial and longitudinal exposure histories, validated psychological measures, accessible subgroup analysis, and documentation of unintended effects.

The strongest emerging methodological program is cumulative and comparative. Researchers can connect administrative records with cohort surveys, environmental measures, transparent causal diagrams, participatory fieldwork, and preregistered evaluation strategies. They should test multiple disparity metrics and report the implications of alternative analytic decisions. Local partnerships improve the chances that interventions address lived barriers; independent outcome assessment protects scientific credibility. The central question becomes which institutional changes reliably expand the capacity of differently situated people to live healthy and fulfilling lives, under what conditions, and at what opportunity cost.

Conclusion

Community health equity is a research object concerned with the organization and distribution of life-supporting opportunity. Health and psychological well-being arise within the material, relational, environmental, and institutional conditions of community life. Income, housing, education, service access, discrimination, collective relationships, and exposure across time interact to shape the risks and resources that individuals encounter. A scientifically defensible account specifies constructs and populations, traces plausible mechanisms, and evaluates the quality of evidence used to connect conditions with outcomes.

The literature supports substantial descriptive evidence for social gradients, an extensive theoretical account of structural and psychosocial pathways, and stronger causal evidence for certain interventions such as the Moving to Opportunity housing-voucher program. The results of one program or setting require careful transportability analysis. Policies and community initiatives should be assessed by who benefits, which outcomes change, whether existing inequality narrows, and whether benefits persist after implementation conditions evolve. This requires linking rigorous causal identification to community-based knowledge, measurement validity, and explicit distributional criteria.

The research horizon extends to new configurations of artificial systems and human institutions, where equitable access increasingly depends on how digital and organizational infrastructures are governed. The scientific priority remains clear: characterize actual opportunity and lived outcomes, identify mechanisms that can be changed, and evaluate interventions across populations and time. On that foundation, community health equity becomes a coherent field of evidence about the conditions under which health and well-being are made attainable throughout collective life.

References

  • Adler, N. E., Boyce, T., Chesney, M. A., Cohen, S., Folkman, S., Kahn, R. L., & Syme, S. L. (1994). Socioeconomic status and health: The challenge of the gradient. American Psychologist, 49(1), 15–24. https://doi.org/10.1037/0003-066X.49.1.15
  • Bailey, Z. D., Krieger, N., Agénor, M., Graves, J., Linos, N., & Bassett, M. T. (2017). Structural racism and health inequities in the USA: Evidence and interventions. The Lancet, 389(10077), 1453–1463. https://doi.org/10.1016/S0140-6736(17)30569-X
  • Bambra, C., Gibson, M., Sowden, A., Wright, K., Whitehead, M., & Petticrew, M. (2010). Tackling the wider social determinants of health and health inequalities: Evidence from systematic reviews. Journal of Epidemiology & Community Health, 64(4), 284–291. https://doi.org/10.1136/jech.2008.082743
  • Berkman, L. F., Glass, T., Brissette, I., & Seeman, T. E. (2000). From social integration to health: Durkheim in the new millennium. Social Science & Medicine, 51(6), 843–857. https://doi.org/10.1016/S0277-9536(00)00065-4
  • Braveman, P. A., Kumanyika, S., Fielding, J., LaVeist, T., Borrell, L. N., Manderscheid, R., & Troutman, A. (2011). Health disparities and health equity: The issue is justice. American Journal of Public Health, 101(S1), S149–S155. https://doi.org/10.2105/AJPH.2010.300062
  • Chetty, R., Stepner, M., Abraham, S., Lin, S., Scuderi, B., Turner, N., Bergeron, A., & Cutler, D. (2016). The association between income and life expectancy in the United States, 2001–2014. JAMA, 315(16), 1750–1766. https://doi.org/10.1001/jama.2016.4226
  • Chetty, R., Hendren, N., & Katz, L. F. (2016). The effects of exposure to better neighborhoods on children: New evidence from the Moving to Opportunity experiment. American Economic Review, 106(4), 855–902. https://doi.org/10.1257/aer.20150572
  • Cummins, S., Curtis, S., Diez-Roux, A. V., & Macintyre, S. (2007). Understanding and representing ‘place’ in health research: A relational approach. Social Science & Medicine, 65(9), 1825–1838. https://doi.org/10.1016/j.socscimed.2007.05.036
  • Diez Roux, A. V. (2001). Investigating neighborhood and area effects on health. American Journal of Public Health, 91(11), 1783–1789. https://doi.org/10.2105/AJPH.91.11.1783
  • Evans, G. W. (2003). The built environment and mental health. Journal of Urban Health, 80(4), 536–555. https://doi.org/10.1093/jurban/jtg063
  • Gee, G. C., & Ford, C. L. (2011). Structural racism and health inequities: Old issues, new directions. Du Bois Review, 8(1), 115–132. https://doi.org/10.1017/S1742058X11000130
  • Greenhalgh, T., Robert, G., Macfarlane, F., Bate, P., & Kyriakidou, O. (2004). Diffusion of innovations in service organizations: Systematic review and recommendations. Milbank Quarterly, 82(4), 581–629. https://doi.org/10.1111/j.0887-378X.2004.00325.x
  • Holt-Lunstad, J., Smith, T. B., & Layton, J. B. (2010). Social relationships and mortality risk: A meta-analytic review. PLOS Medicine, 7(7), e1000316. https://doi.org/10.1371/journal.pmed.1000316
  • Jagosh, J., Macaulay, A. C., Pluye, P., Salsberg, J., Bush, P. L., Henderson, J., Sirett, E., Wong, G., Cargo, M., Herbert, C. P., Seifer, S. D., Green, L. W., & Greenhalgh, T. (2012). Uncovering the benefits of participatory research: Implications of a realist review for health research and practice. Milbank Quarterly, 90(2), 311–346. https://doi.org/10.1111/j.1468-0009.2012.00665.x
  • Kawachi, I., Kennedy, B. P., Lochner, K., & Prothrow-Stith, D. (1997). Social capital, income inequality, and mortality. American Journal of Public Health, 87(9), 1491–1498. https://doi.org/10.2105/AJPH.87.9.1491
  • Krieger, N. (2001). Theories for social epidemiology in the 21st century: An ecosocial perspective. International Journal of Epidemiology, 30(4), 668–677. https://doi.org/10.1093/ije/30.4.668
  • Ludwig, J., Duncan, G. J., Gennetian, L. A., Katz, L. F., Kessler, R. C., Kling, J. R., & Sanbonmatsu, L. (2012). Neighborhood effects on the long-term well-being of low-income adults. Science, 337(6101), 1505–1510. https://doi.org/10.1126/science.1224648
  • Marmot, M. G., Smith, G. D., Stansfeld, S., Patel, C., North, F., Head, J., White, I., Brunner, E., & Feeney, A. (1991). Health inequalities among British civil servants: The Whitehall II study. The Lancet, 337(8754), 1387–1393. https://doi.org/10.1016/0140-6736(91)93068-K
  • Marmot, M., Friel, S., Bell, R., Houweling, T. A. J., & Taylor, S. (2008). Closing the gap in a generation: Health equity through action on the social determinants of health. The Lancet, 372(9650), 1661–1669. https://doi.org/10.1016/S0140-6736(08)61690-6
  • McEwen, B. S. (1998). Protective and damaging effects of stress mediators. New England Journal of Medicine, 338(3), 171–179. https://doi.org/10.1056/NEJM199801153380307
  • National Academies of Sciences, Engineering, and Medicine. (2019). Integrating social care into the delivery of health care: Moving upstream to improve the nation's health. National Academies Press. https://doi.org/10.17226/25467
  • Office of Disease Prevention and Health Promotion. (n.d.). Social determinants of health. Healthy People 2030, U.S. Department of Health and Human Services. https://odphp.health.gov/healthypeople/priority-areas/social-determinants-health
  • Proctor, E., Silmere, H., Raghavan, R., Hovmand, P., Aarons, G., Bunger, A., Griffey, R., & Hensley, M. (2011). Outcomes for implementation research: Conceptual distinctions, measurement challenges, and research agenda. Administration and Policy in Mental Health and Mental Health Services Research, 38(2), 65–76. https://doi.org/10.1007/s10488-010-0319-7
  • Sampson, R. J., Raudenbush, S. W., & Earls, F. (1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science, 277(5328), 918–924. https://doi.org/10.1126/science.277.5328.918
  • Viswanathan, M., Kraschnewski, J. L., Nishikawa, B., Morgan, L. C., Honeycutt, A. A., Thieda, P., Lohr, K. N., & Jonas, D. E. (2010). Outcomes and costs of community health worker interventions: A systematic review. Medical Care, 48(9), 792–808. https://doi.org/10.1097/MLR.0b013e3181e35b51
  • Wallerstein, N., & Duran, B. (2010). Community-based participatory research contributions to intervention research: The intersection of science and practice to improve health equity. American Journal of Public Health, 100(S1), S40–S46. https://doi.org/10.2105/AJPH.2009.184036
  • Williams, D. R., Yu, Y., Jackson, J. S., & Anderson, N. B. (1997). Racial differences in physical and mental health: Socio-economic status, stress and discrimination. Journal of Health Psychology, 2(3), 335–351. https://doi.org/10.1177/135910539700200305
  • World Health Organization. (2010). A conceptual framework for action on the social determinants of health. WHO. https://www.who.int/publications/i/item/9789241500852
  • World Health Organization. (2025). World report on social determinants of health equity. WHO. https://www.who.int/publications/i/item/9789240107588