Comparison of three alcohol consumption measures: a concurrent validity study.

Objective: The major objective of the current research was to compare three different alcohol consumption measures. Method: A 90-day timeline follow-back measure was compared with two averaging measures (a quantity-frequency scale and a grid measure) in a counterbalanced design with 42 male and 30 female subjects. Results: Reasonable agreement was found among the three measures; the grid approach yielded significantly higher estimates of drinking days and total consumption, but lower estimates of peak intoxication levels. Subjects reported significantly less confidence in the accuracy of their own reports on the timeline, relative to the averaging methods. Conclusions: The merits of the various measures and implications for their use in treatment outcome research are discussed.