WITHDRAWN: Comparing Behavioral Health Outcomes and Treatment Utilization of Justice-Involved and Non-Justice-Involved Native Adults

Background and Purpose: Native populations are overrepresented in the criminal justice system with an incarceration rate more than double that of white Americans and in some states justice involvement is 7 times that of whites. While this is the case, there remains a scarcity of data and research on Native Americans involvement with the justice system, and what their lives look like prior to, during and after justice-involvement. The current study fills the gap by exploring rates of substance use and mental health diagnoses and treatment utilization among justice-involved and non-justice involved Native samples. Further, we examined correlates associated with past-year service utilization.

Methods:Data from the 2015-2019 National Survey on Drug Use and Health (NSDUH; N=214,505 adults aged 18+) was used. NSDUH is an annual, nationally representative survey administered to U.S. non-institutionalized civilians that collects data on substance use, service utilization, and other behavioral health-related constructs. For this study, the subsample examined 3,068 Native adults, defined as American Indian or Alaskan Native in the NSDUH dataset. Past year criminal justice involvement is a recoded variable that indicates if a respondent reported either being on probation, parole, or being arrested in the last 12-months.Substance use disorder is operationalized as meeting DSM-IV criteria for any substance abuse or substance dependence. Any past-year mental illness was a recoded variable of reported mental illness diagnosis in the last year. The healthcare attendance variables are dichotomized variables capturing any visit (i.e., to the emergency room) in the past year. Mental health treatment refers to the receipt of past year mental health treatment (i.e., inpatient or outpatient). The substance use treatment construct refers to receiving any form of substance use treatment. T-tests, chi-square tests of independence, and multivariable logistic regression models were run to test for differences on service utilization, mental health status, and substance misuse between native non-justice involved and justice-involved samples.

Findings: Among Native adults in the sample, 446 reported justice-involvement in the past year. Justice-involved Native adults were significantly more likely to use the emergency department (48.54 vs. 37.42; p<.01), utilize mental health treatment (23.92 vs. 12.98; p<.001), and access substance use treatment (22.44 vs. 1.91; p<.001) in the past year. Additionally, justice-involved Native adults reported significantly higher rates of mental illness (27.91 vs. 19.60; p<.05) and past-year substance use disorder (42.08 vs. 9.21; p<.001). In regression models, justice-involvement was significantly associated with past year substance use treatment (AOR=7.48, 95% CI=3.71-15.03) and past-year mental health treatment (AOR=2.91, 95% CI=1.80-4.69).

Conclusion and Implications: The findings from this work highlight the differences among substance use, mental illness, and treatment utilization between justice-involved and non-justice-involved Native individuals. Findings suggest that justice-involved Native individuals fair worse in all areas and argue for the consideration of interventions that incorporate both culturally-sensitive and justice responsive models to improve the behavioral health outcomes of Native populations. Lastly, while justice-involved Native populations utilize treatment services at higher levels, there is reason to believe that there continues to be a mismatch of need and utilization and further exploration is warranted.