The longitudinal stability of the Addiction Severity Index.

The Addiction Severity Index (ASI) is a structured interview widely used by substance abuse clinicians and researchers for client screening, determining treatment needs, and assessing treatment outcomes. Previous researchers have evaluated inter-rater agreement, test-retest reliability, and concurrent validity. The present report describes the stability of ASI scores in longitudinal work. In the context of an ongoing treatment outcome evaluation study involving seven assessors, inter-rater agreement, inter-rater reliability, as well as intra- and inter-rater accuracy were assessed repeatedly during a 2-year period. The results show the scores derived from the ASI to be stable across assessors and over time. The relationship between stable scores and resources required for training are discussed.