Evaluating the Computerized Assessment and Referral System (CARS) Screener: Sensitivity and Specificity as a Screening Tool for Mental Health Disorders among DUI Offenders

摘要

Background/Purpose: A growing body of evidence suggests that people who are arrested for driving under the influence (DUI) of alcohol are more likely to screen positive for psychiatric disorders than the general population. Additionally, psychiatric comorbidity has been shown to increase vulnerability to criminal re-offense. However, DUI programs face many barriers to screening for psychiatric disorders. This paper evaluates the sensitivity and specificity of a screening tool developed for these programs, the Computerized Assessment and Referral System (CARS) Screener. Methods: We used data from 381 DUI offenders in Massachusetts, as well as a secondary data source, the National Comorbidity Survey-Replication (NCS-R: N=9,282) to examine the accuracy of the CARS Screener when compared to full assessment. Results: Based on both sets of analyses, we found that the CARS Screener offers a sensitive and specific method to screen for many psychiatric disorders. Specifically, the CARS Screener has a high sensitivity and specificity for bipolar disorder, intermittent explosive disorder, depressive disorders, generalized anxiety disorder, alcohol and drug use disorders, gambling disorder, post-traumatic stress disorder, panic attacks, and social phobia. Conclusion: The CARS Screener appears to be an effective tool that will help DUI programs better understand and address the mental health issues facing their clients.