|Kimberlee J. Trudeau, Joanna Burtner, Kruti Joshi, Albert Villapiano, Megan Jones, & Stephen F. Butler|
Presented at ARM 2016
Research Objective: In the fragmented US healthcare system, patients seeking substance abuse treatment may not receive a comprehensive mental or physical health evaluation. This study compared economic, medical, and legal outcomes of adults evaluated for substance use disorders by schizophrenia diagnosis (Dx) status.
Study Design: A U.S. cross-sectional analysis was conducted using substance abuse assessment data. Participants were classified into 3 cohorts: 1) Probable Dx defined as those reporting schizophrenia diagnosis or >2 of the following: thought disorder diagnosis, psychological hospitalization, and hallucination without illicit substance use; 2) Possible Dx defined as those reporting thought disorder diagnosis or hallucination without illicit substance use; or 3) No Dx defined as those not Probable or Possible Dx. Analysis of variance and Chi-square tests were used to examine differences in cohort means and proportions, respectively. Logistic regression estimated odds ratios (OR) with 95% Wald confidence intervals (CI).
Population Studied: Participants ages >18 who completed the Addiction Severity Index-Multi-media Version (ASI-MV), a multi-domain substance abuse assessment across 44 states between 2009 and 2015.
Principal Findings: Of the total sample (N=368,365), 59% were Caucasian, 65% were male, 38% were between ages 21 and 30, and 71% had a minimum of a high school education. The total sample was classified into 3 cohorts: Probable Dx (9%), Possible Dx (4%), and No Dx (87%). Economic: Relative to the No Dx cohort, the Probable Dx cohort had 3 times greater odds of receiving money for physical disability (OR = 3.39; 95% CI (3.30 – 3.49)) and 11 times greater odds of receiving money for a psychiatric disability (OR = 11.54; 95% CI (11.16 – 11.931)). Medical: In addition, the Probable Dx cohort was at least 2.5 times as likely as the No schizophrenia cohort to report a visit to ER in past 30 days (OR = 2.64; 95% CI (2.57 – 2.70)), and the Possible Dx cohort was at least 1.5 times as likely (OR = 1.57; 95% CI (1.51 – 1.63)). Legal: Those with Probable Dx were about 2.5 times as likely to have been arrested for violent crimes (OR = 2.59; 95% CI (2.51 – 2.67)) compared to those in the No Dx cohort.
Conclusions: Overall, severity of issues in each of these three life outcomes (economic, medical, legal) increases based on probability of schizophrenia diagnosis (i.e., No Dx to Possible Dx to Probable Dx). Characteristics and outcomes among the Possible and Probable Dx cohorts appeared similar; both demonstrating poorer economic, medical, and legal outcomes relative to the No Dx cohort. This suggests that access to mental health screening, particularly symptoms associated with schizophrenia, may reveal emerging conditions in those entering the healthcare system through substance abuse treatment settings.
Implications for Policy or Practice: This study’s insights reinforce the need for screening, evaluation, and care coordination for mental health concerns for those who present to substance abuse centers. Based on these findings, the policy recommendation is to triage individuals with Probable Dx promptly for mental health screening and further evaluation, and provide them with comprehensive support.