Issues in the definition and measurement of drinking outcomes in alcoholism treatment research.

This article reviews methodological and conceptual issues regarding the choice of drinking outcome measures in alcoholism treatment research. The following issues are discussed: Should drinking outcomes be conceptualized in terms of an underlying unitary disorder, or should provision be made for independent outcomes that cover a wide variety of dimensions? Which drinking outcomes are typically measured in treatment evaluation studies and how are they operationalized? What are the empirical associations among drinking outcome measures? If multiple outcomes are measured, which should be given primary importance? Over what period of time should treatment outcome be evaluated? What procedures can be used to detect, correct or prevent the response bias associated with verbal report methods? Because outcome measures need to fit the hypotheses and practical needs of a particular study, it is unlikely that complete standardization can be achieved across all studies. Nevertheless, given the importance of drinking outcomes and the need for economy, two primary dependent measures are recommended: (1) proportion of available drinking days abstinent; and (2) intensity of drinking, as defined by the total amount consumed (in ounces absolute alcohol) during the follow-up period divided by the number of actual drinking days. This article also proposes a strategy that may help to guide the selection of outcome measures in future research.