Loza & Loza-Fanous (2000) found that the SAQ’s high-risk group always “failed” with more frequency and sooner than the medium and low-risk groups on all post-release criterion measures