Performance of a craving criterion in DSM alcohol use disorders

Abstract

Objective: Adding a craving criterion, presently in ICD-10 diagnosis of alcohol dependence, has been under consideration as one possible improvement to the DSM-IV, and recently proposed for inclusion by the DSM Substance Related Disorders Work Group in the 5th revision (DSM-V) of diagnostic criteria for alcohol use disorders. To inform cross-cultural applicability of this modification, performance of a craving criterion is examined in Emergency Departments (EDs) in four countries manifesting distinctly different culturally-based drinking patterns (Mexico, Poland, Argentina, U.S.). Method: Exploratory Factor Analysis (EFA) and Item Response Theory (IRT) were used to examine psychometric properties and individual item characteristics of the 11 DSM-IV abuse and dependence criteria with and without craving for each country separately. Differential Item Functioning (DIF) analysis was performed to examine differences in the difficulty of endorsement (severity) and discrimination of craving across countries. Results: EFA found craving fit well within a one dimensional solution, and factor loadings were high across all countries. Results from IRT analyses indicated that both discrimination and difficulty estimates for the craving item were located in the middle of the corresponding discrimination and difficulty ranges for the other 11 items for each country, but did not substantially increase the efficiency (or information) of the overall diagnostic scheme. Across the four countries, no DIF was found for difficulty but significant DIF was found for discrimination (similar to other DSM-IV criteria). Conclusions: Findings suggest that while craving performed similarly across EDs in the four countries, it does not add much in identification of individuals with alcohol use disorders.