Assessment of automated craving across substances and across cultures: stability-analysis of the Craving Automated Scale (CAS)

Abstract

Background The transition from hedonic to compulsive use in Substance Use Disorders (SUD) is a critical point in SUD progression and hence relevant for assessment and treatment. To measure the habitual patterns of substance consumption, the Craving Automated Scales (CAS) for alcohol (CAS-A), substances (CAS-S) and cigarette smoking (CAS-CS) were developed and introduced to different countries. In this study, we aimed to investigate the structural stability of CAS across substances and cultures. Methods This study analyzed the CAS-scores of a sample of 370 participants in Germany, China and the UK, including 262 opioid-users, 65 smokers and 43 alcohol-users. We performed stability analyses to check the stability (i. e. factorial invariance) of factor solutions. Based on confirmed stability of the general factor (gfactor) solution and the calculations rule obtained in the previous validation of CAS-alcohol (CAS-A), the factor structures of CAS-A, CAS-S and CAS-CS were compared. Results The gfactor solutions based on calculations rule shows good stability, with the mean stability coefficients of 0.990 and 0.977 for CAS-S and CAS-CS respectively. The gfactor patterns were similar for CAS-A, CAS-S and CAS-CS, as well as across samples (Germany, China and the UK), with most factor-loadings larger than 0.7. Based on these findings, CAS-S and CAS-CS were also associated with established clinical measures of SUD. Conclusions Our findings suggest the two-gfactor solution based on a proposed calculation rule has a high stability across substances and cultures. This could be in line with common neurobiological mechanisms underlying habitual substance use. Moreover, comparing CAS with established clinical tools suggests that CAS might assess the automated behavior in substance consumption in a more sophisticated way.