Psychometric properties of the CAST and SDS scales in young adult cannabis users

Abstract

AimTo assess the validity of two cannabis use severity scales among young cannabis users and to evaluate their ability to detect Substance Use Disorders (SUD). Participants241 volunteers (18–25years), with a wide spectrum of cannabis use in the last 12months. MeasurementsThe Cannabis Abuse Screening Test (CAST) and Severity of Dependence Scale (SDS) were self-administered. The Psychiatric Interview for Substance and Mental Disorders (PRISM) was used as gold standard for cannabis use disorders according to DSM-IV. Reliability and validity were assessed for two different CAST coding algorithms (b—binary and f—full) and for the SDS. In addition, the cannabis use diagnostic criteria contained in the PRISM were grouped to approximate forthcoming proposed DSM-V criteria to further evaluate these scales. Findings26.6% (95% CI: 21.0–32.2) of the subjects met criteria for cannabis dependence, and 49.0% (95% CI: 42.7–55.3) for cannabis use disorders. For both scales internal consistency (Cronbach's alpha>0.71) and test–retest intraclass correlation coefficients (>0.80) were good. The score 12 in the CAST-full discriminated better than others between presence and absence of dependence (27.0%; 95% CI: 21.4–32.6) while the score for discrimination of SUD was 9 (51.5%; 95% CI: 45.1–57.8). For the SDS the values were 7 (22.0%; 95% CI: 16.8–27.2) and 3 (64.7%; 95% CI: 58.7–70.8), respectively. According to proposed DSM-V criteria, for moderate and severe addiction the values for the CAST-f were 7 (68.5%; 95% CI: 62.5–74.3) and 12 (27%; 95% CI: 21.3–32.6) and for the SDS, 3 (65.0%; 95% CI: 58.7–70.8) and 7 (22%; 95% CI: 17.0–34.3), respectively. ConclusionsThe CAST and SDS applied to young cannabis users are reliable and valid measures to detect cannabis use disorders when compared to both DSM-IV and proposed DSM-V criteria.