ICD-10 includes a craving criterion for alcohol dependence while DSM-IV does not. Little is known about whether craving fits with or improves the DSM-IV criteria set for alcohol-use disorders.

Data were derived from current drinkers (n=18 352) in the 1991–1992 National Longitudinal Alcohol Epidemiologic Survey (NLAES), a nationally representative survey of US adults >17 years of age. The Alcohol Use Disorder and Associated Disabilities Interview Schedule was used to assess the eleven DSM-IV dependence and abuse criteria, and alcohol craving. Exploratory factor, item response theory, and regression analyses were used to evaluate the psychometric properties and concurrent validity of DSM-based alcohol disorder criteria with the addition of alcohol craving.

The past 12-month prevalence of craving was 1.3%. Craving formed part of a unidimensional latent variable that included existing DSM-IV criteria. Craving demonstrated high severity on the alcohol-use disorder continuum, resulting in an improved dimensional model with greater discriminatory ability compared with current DSM-IV criteria. Correlates of the diagnosis did not change with the addition of craving, and past 12-month craving was associated with prior alcohol dependence, depression, and earlier age of alcohol disorder onset among those with current DSM-IV alcohol dependence.

The addition of craving to the existing DSM-IV criteria yields a continuous measure that better differentiates individuals with and without alcohol problems along the alcohol-use disorder continuum. Few individuals are newly diagnosed with alcohol dependence given the addition of craving, indicating construct validity but redundancy with existing criteria.