Validity of the Addiction Severity Index in a substance abuse treatment sample with psychiatric comorbidity

Background: The Addiction Severity Index (ASI) has been shown to be a reliable measure of the severity of problems faced by substance users. The validity of Interviewer Severity Ratings (ISRs) has been questioned, particularly among interviewers with less training. There have also been mixed results for the validity of the Psychiatric Status scale among people with mental health problems. This study examines the relationship between ASI composite scores (CSs), ISRs and data from the Symptoms Checklist-90-R (SCL-90-R).

Methods: 142 participants completed the ASI and SCL-90-R at intake into a drug treatment program in Long Beach, California. This program provides residential and outpatient drug treatment and integrated mental health services. The majority of participants were male (81%) and Latino (51.4%) or Black (43.7%). The most commonly used drugs were cocaine and amphetamine.

Results: The CS and ISR for the Psychiatric Status scale were positively correlated (r=.80, p<.001). The SCL-90-R was correlated with both the Psychiatric Status CS (r=.60, p<.001) and ISR (r=.50, p<.001). The CS and ISR were most strongly correlated with the depression subscale of the SCL-90-R. The ASI Medical Status CS and ISR were correlated with the Somatization subscale of the SCL-90-R (.33>r<.37, p<.001). Participants who accessed integrated mental services had higher Psychiatric Status CS, ISR and scores on the SCL-90-R.

Conclusions: This study provides additional support for the validity of the CS Psychiatric Status scale. Evidence was also found for the validity of the ISR. However, the CS was more strongly associated with scores on the SCL-90-R.