Volume of ethanol consumption: effects of different approaches to measurement.

OBJECTIVE: Eight different approaches to measuring alcohol consumption were compared in terms of average daily volume of ethanol intake, selected percentiles of the volume distribution, the proportion of drinkers exceeding a volume-based cutpoint for moderate drinking and the estimated association between volume of intake and alcohol use disorders.
METHOD: Data were drawn from the 1988 National Health Interview Survey and the 1992 National Longitudinal Alcohol Epidemiologic Survey. The eight approaches compared overall and beverage-specific questions, reference periods of varying lengths, and measures based solely on usual intake with those that incorporated different aspects of atypical heavy drinking.
RESULTS: Average daily ethanol intake ranged from 0.43 oz based on two questions on current usual frequency and quantity of drinking (assuming 0.5 oz of ethanol per drink) to 0.72 oz based on 21 questions that included usual and heaviest consumption of beer, wine and distilled spirits in the year preceding interview.
CONCLUSIONS: Estimated volume was highly sensitive to the number and types of questions upon which it was based, and changes in formulation that resulted in relatively small increases in mean volume often were associated with much larger increases in the proportion of drinkers exceeding some specified level of intake and in the estimated association between consumption and alcohol use disorders. These issues should be considered when deciding on the consumption items to be included in alcohol surveys.