Psychometric properties of CAST for early detection of problematic cannabis use in Spanish adolescents

Abstract

The high prevalence rates of cannabis use in adolescents and its early onset constitutes a major public health problem, raising the need for its early detection. The availability of validated tools to analyze early cannabis use is essential to detect problematic use at an early age. The Cannabis Abuse Screening Test (CAST) (Legleye et al., 2007) is widely applied in Europe; however, the CAST cut-off scores vary according to the setting, the screening objective, and the correction version (CAST-f or CAST-b), creating therefore confusion in its application. Moreover, the psychometric properties of the CAST as a tool for detecting problematic cannabis use are understudied. To fill this gap, such psychometric properties have been analyzed in a sample of Spanish adolescents while using different cut-off scores for CAST-f and CAST-b. Based on our findings, the optimal cut-off scores are 2 points for CAST-b and 4 points for CAST-f. The internal reliability of CAST-f (α = 0.83) and CAST-b (KR-20 = 0.80) are satisfactory. Factorial analysis suggested the assumption of a one-dimension model. The CAST seems to be a valid and reliable tool for early screening of problematic cannabis use in Spanish adolescents.