Convergent Validity of the ASAM Criteria in Co-Occurring Disorders

Abstract

Objective: The complexity of substance use and psychiatric disorders demands thorough assessment of patients for integrated services. We tested the convergent validity of the software version of the American Society of Addiction Medicine (ASAM) second edition-revised criteria for patient placement by examining the software's ability to discriminate based on a variety of demographic and clinical factors. Methods: This prospective naturalistic multi-site study examined the software's assignment of patients to three types of treatment (addiction only services, dual diagnosis capable, and dual diagnosis enhanced) and whether these assignments indicated an ability to discriminate between patients with and without dual diagnosis based on clinical characteristics and severity. Ten addiction treatment clinics spanning three counties participated, and both patients and ASAM assessors were kept blind to the ASAM recommendation. Patients were assigned to their respective treatment options based on routine assessment by clinicians at intake, which they had in addition to the ASAM interview. Three months after treatment initiation a follow-up interview with ASAM was conducted. Results: There were 261 patients in the study, 96 (36.7%) were assigned to addiction only services, 42 (16.1%) to dual diagnosis capable, and 123 (47.1%) to dual diagnosis enhanced. Patients assigned to the two dual diagnosis groups were significantly more likely to be younger and have fewer years of work than other patients. There were significant differences in history of inpatient and outpatient psychiatric treatment across groups. For example, a larger percentage of those in addiction only services had never been in inpatient treatment, while more of those in the two dual diagnosis groups had three or more inpatient stays. Despite similar alcohol and drug severity scores, patients recommended by the software for dual diagnosis enhanced programs showed a gradient of significantly higher psychiatric (p <.001), legal (p <.04), and family (p <.001) Addiction Severity Index composite scores at baseline than patients in dual diagnosis capable and addiction only services. Conclusions: Results show a high prevalence of co-occurring program recommendations with statistically significant and clinically meaningful differences between patient groups. The convergent validity of the revised version of ASAM Criteria Software is supported by these results.