DSM-IV alcohol abuse: investigation in a sample of at-risk drinkers in the community.

OBJECTIVE: Although a category for alcohol abuse has been included in the major nomenclatures since DSM-III, many questions have been raised about the definitions and validity of this category.
METHOD: We investigated DSM-IV alcohol abuse in 628 at-risk drinkers residing in households who never met criteria for DSM-IV alcohol dependence at the time of a baseline interview. Cross-sectional and prospective longitudinal information were used.
RESULTS: Subjects most often met criteria for abuse with the single criterion, recurrent hazardous use (usually driving after drinking too much, on average 4.5 times per year). Younger and white subjects were more likely to receive the abuse diagnosis at baseline, as were drug users and those who drank five or more drinks per occasion frequently. Very few subjects received a diagnosis of DSM-IV alcohol dependence at follow-up, and those who did were equally likely to come from the abuse and non-abuse baseline groups. A current abuse diagnosis at baseline strongly and significantly predicted an abuse diagnosis at follow-up, although a past-only (remitted) abuse diagnosis at baseline was not clearly predictive of abuse at follow-up.
CONCLUSIONS: The DSM-IV alcohol abuse category has some predictive validity. However, given the manner in which respondents met criteria for the disorder, its merits as a diagnostic category remain in question. Researchers should be cautious about combining alcohol dependence and abuse into the same category.