The AC-OK Cooccurring Screen: Reliability, Convergent Validity, Sensitivity, and Specificity

Abstract

The principal barriers to universal screening for the cooccurring disorders of mental illness and substance abuse are training, time, cost, and a reliable and valid screen. Although many of the barriers to universal screening still remain intact, the lack of a cooccurring screen that is effective and can be administered in a cost efficient way is no longer an obstacle. This study examined the reliability, factor structure, and convergent validity of the 15-item AC-OK Cooccurring Screen. A total of 2,968 AC-OK Cooccurring Screens administrated to individuals who called or went to one of the nine participating mental health and substance abuse treatment facilities were administrated and analyzed. Principal axis factor (PAF) analysis was used in the confirmatory factor analysis to identify the common variance among the items in the scales while excluding unique variance. Cronbach's Alpha was used to establish internal consistency (reliability) of each subscale. Finally, the findings from the AC-OK Cooccurring Screen were compared to individual scores on two standardized reference measures, the addiction severity index and the Client assessment record (a measure of mental health status) to determine sensitivity and specificity. This analysis of the AC-OK Cooccurring Screen found the subscales to have excellent reliability, very good convergent validity, excellent sensitivity, and sufficient specificity to be highly useful in screening for cooccurring disorders in behavioral health settings. In this study, the AC-OK Cooccurring Screen had a Cronbach's Alpha of .92 on the substance abuse subscale and a Cronbach's Alpha of .80 on the mental health subscale.