Substance abuse treatment in correctional versus non-correctional settings: Analysis of racial/ethnic and gender differences.

BACKGROUND: Alcohol and drug abuse are widespread in the US. Substance abuse treatment services are effective, but utilization of services is low, particularly among African Americans, Hispanics, and women. Substance abuse is strongly associated with incarceration, and African Americans and Hispanics make up a disproportionate percentage of individuals with substance abuse problems involved in the criminal justice system. High treatment need, low treatment uptake, and the association between substance abuse and incarceration have led, in part, to correctional institutions filling the treatment gap by increasingly providing safety-net treatment services. We sought to better understand racial/ethnic and gender differences in determinants of treatment location (jail or prison versus non-correctional settings) among treatment-seeking adults.
METHODS: We used repeated cross-sectional data from the National Survey on Drug Use and Health (2002-2016) to identify White, African American, and Hispanic past-year substance abuse treatment participants (n=6,435). We tested the modifying roles of race/ethnicity and gender on the association between several exposure variables and treatment locus using multiple logistic regression.
RESULTS: Ten percent of treatment participants utilized treatment services in a jail or prison, which varied by race/ethnicity (9% of Whites, 15% of African Americans, 12% of Hispanics) and by gender (11% of men, 9% of women). In our fitted models, we found that educational attainment and past-year employment status varied in effect size between African Americans and Whites. The associations for both variables were strongest among African Americans (Any college vs. Less than high school - adjusted Odds Ratio [aOR] = 0.23, 95% Confidence Interval [95% CI] = 0.08, 0.70; Ever unemployed vs. Never unemployed in the past year - aOR = 5.32, 95% CI = 1.94, 14.60). Health insurance status was significantly associated with treatment in a jail or prison only among Whites (Private vs. No insurance - aOR = 0.37, 95% CI = 0.19, 0.69). Co-occurring mental health diagnosis was significant only among African Americans (Any mental health diagnosis vs. none - aOR = 3.91, 95% CI = 1.38, 11.09). Employment and health insurance status were significant only among men (aOR = 2.18, 95% CI = 1.26, 3.77; aOR = 0.39, 95% CI = 0.22, 0.70, respectively).
CONCLUSION: We identified modifying roles for race/ethnicity and gender in the relationship between several factors and treatment utilization in a jail or prison versus non-correctional treatment settings. More numerous factors and stronger effect sizes were identified among African Americans and men in particular. Health promotion interventions promoting the uptake of substance abuse treatment should tailor services to align with the needs of those with the highest risk for incarceration.