Community Distress and Racial Inequities In Adult Protective Service Outcomes: Longitudinal Evidence from California Counties

Background/Purpose

While Child Protective Services research extensively examines systemic risk and racial disparity, a significant gap remains in understanding Adult Protective Services (APS). Drawing on ecological systems theory, this study investigates how county-level stressors, poverty, and geography are associated with the volume of completed APS responses. We specifically test whether the impact of community-level psychological distress and poverty on response closures differs between Black and White adults to determine if risk levels experienced by racial groups are tied to community factors related to well-being (i.e., the absence of community-level distress and economic deprivation).

Methods

Using multilevel mixed-effects negative binomial regression, we analyzed 3,958 county-month observations (October 2019–December 2023) nested within California’s 58 counties, with data obtained from the California Department of Social Services APS and County Services Block Grant Monthly Statistical Report. The negative binomial framework was selected to account for overdispersed count data. The dependent variable is the monthly count of closed investigations (i.e., the aggregate reporting variable provided by California data, which corresponds to the monthly volume of APS responses) adjusted for the eligible population. Predictors include county-level race rates, poverty, and psychological distress. We report results as Incidence Rate Ratios (IRR), which represent the relative change in the rate of closed investigations given a one-unit change in the predictor. To interpret interactions, we calculated predictive margins, which hold all covariates at their mean to estimate the predicted rate of closures across varying levels of community stress.

Results

Main effects models reveal significant baseline disparities: investigation closure rates for eligible White residents are significantly lower compared to the investigation closure rates for eligible Black residents (IRR = 0.61, p < .001). Community-level stress and urbanicity were associated with increased risk. While poverty serves as a universal risk, it did not moderate racial disparities. A significant interaction emerged between race and county-level stress (IRR = 1.11, p < .001). Predictive margins reveal two profiles: 1) Risk Stability: Black population rates of investigation remain consistently high regardless of community stress levels; and 2) Conditional Protection: White population rates experience lower investigation volume in high-stability (i.e., low-stress) environments, a protection that evaporates as community distress increases. Finally, rurality was associated with a 49% lower rate of investigation closures (IRR = 0.51, p < .05), suggesting that geographic barriers may limit the system’s ability to reach and close cases compared to urban centers.

Conclusions and Implications

These findings fill a gap in the APS research field and suggest that APS involvement for Black Californians is characterized by risk stability, remaining high even in low-distress environments. In contrast, White adults experience lower rates of response contingent on community well-being, which is a benefit that vanishes under high community stress. This indicates that community distress may trigger system involvement for populations typically shielded by social capital, while failing to alleviate the persistent high rates faced by Black communities. For policymakers, these results highlight the need for targeted resources in rural and high-stress counties to address systemic volatility and differential service provision.