Incarceration rates differ between races/ethnicities and are highest among African Americans. Racial discrimination exasperates negative health effects, but racial identity may mitigate some its effect. Discrimination and identity may affect the likelihood of incarceration in opposing directions.

Objectives

Guided by Critical Race Theory, we sought to quantify racial/ethnic differences in discrimination and identity, and predict the number of days incarcerated by discrimination and identity scores.

Methods

We used the National Epidemiologic Survey on Alcohol and Related Conditions-III (2012-2013) to identify incarceration histories and create discrimination and identity scores (n=34,956). We used zero-inflated Poisson regression to predict number of days incarcerated across discrimination and identity scores. Weights and stratum were applied to account for the parent survey’s complex sampling design.

Results

Mean discrimination scores (0-6) were highest among African Americans (weighted mean = 1.83) and lowest among Whites (weighted mean = 0.64). Identity scores (0-8) were also highest among African Americans (weighted mean = 6.90) and lowest among Whites (weighted mean = 5.97). Our fitted models demonstrated a positive and strong relationship between discrimination scores and predicted days incarcerated, which varied between groups (Predicted days incarcerated at discrimination scores 0 and 6 – African American, 20-72 days; Hispanic, 11-46 days; Native American, 8-65 days; White, 9-83 days).

Conclusion

Discrimination score – but not identity score – maintained a strong association with predicted days incarcerated in our fully adjusted model. Interventions aimed at preventing incarceration should be tailored to the unique histories of those at the greatest risk for arrest and detainment.