Police Abuse and Mental Health in a Nationally Representative Sample of African-Americans

Background: Media reports over the past several years have captured the problem of police abuse coming to a head in the United States, drawing national and even global attention to this long-standing problem that disproportionally harms African-Americans. In response, the American Public Health Association released a statement declaring law enforcement violence to be a critical but direly under-examined public health issue. Limited research has shown that police abuse is associated with distress, depression, anxiety, and trauma symptoms in U.S. samples; however, no prior study has examined this topic using nationally representative survey data of African-American households –the demographic group purported to be at highest risk for this exposure. Thus, we analyzed the National Survey of American Life (NSAL) to explore the psychiatric significance of police abuse among African-American adults.

Methods:

Sample and procedures

Data were drawn from the NSAL, a national household probability sub-sample of African-American adults (N=3570), who were assessed using the World Health Organization Composite International Diagnostic Interview (WHO CIDI).

Measures

Police discrimination was measured using a dichotomous item: “Have you ever been unfairly stopped, searched, questioned, physically threatened or abused by the police?” Respondents were assessed for 12-month DSM-IV Axis I disorders using the WHO-CIDI. Diagnostic categories included mood disorders; anxiety disorders; and Post-traumatic stress disorder (PTSD). Lifetime suicidality was assessed in terms of ideation, plan, and attempt. Self-reported demographic covariates included age, sex, marital status, education, employment, and income. DSM-IV diagnoses of alcohol or substance use disorders were included as confounders, given that substance use frequently co-occurs with mental health issues and may draw police attention.

Analyses

Wald X2 tests were used to test for differences in prevalence of psychiatric outcomes among people with and without history of police abuse. Multivariable logistic regression was used to test for associations between police abuse and 12-month psychiatric disorders while adjusting for sociodemographic variables and 12-month alcohol/substance use. Multivariable logistic regression was used to test for associations between police abuse and lifetime suicidality after adjusting for sociodemographic variables and lifetime psychiatric disorders (mood, anxiety, PTSD, alcohol/substance).

Results: Police abuse was more prevalent among respondents with psychiatric disorders, and was associated with greater odds of endorsing 12-month psychiatric disorders after adjusting for sociodemographic covariates and alcohol/substance use disorders, compared with those who did not report any police abuse. Police abuse was also associated with greater odds of reporting lifetime suicidal ideation, plans, and attempts, after adjusting for sociodemographic covariates and lifetime psychiatric disorders.

Conclusion/Implications: This is the first study to our knowledge to show that police victimization is associated with major psychiatric disorders and suicidality in a nationally representative sample of African-American households. Police abuse is an alarmingly common occurrence among African-Americans, with measurable mental health correlates. Police abuse can be stressful and traumatizing, injuring the mind by way of the stress-response system. Future studies can use longitudinal designs to establish causal direction. Findings call for proper police officer training and accountability, community-based restorative justice approaches, as well as targeted mental health screenings for African-Americans.