Agreement between two dietary methods in the measurement of alcohol consumption

This study compares estimates of alcohol consumption from two dietary methods: a self-reported diet diary and a quantitative food frequency questionnaire. In the 1984-85 University of Michigan Food Frequency Study, 228 black and white, male and female respondents, ages 23-51 years, kept diet diaries covering 16 days during a 1-year period. At the end of the year, the same respondents completed a dietary quantity-frequency questionnaire that included alcoholic beverages. The two methods showed excellent agreement in estimated group mean ethanol consumption and good agreement in the ranking of individual respondents. Respondents described alcohol consumption in abstract terms on the questionnaire in a way that was consistent with the consumption recorded in the diary. However, the two methods showed only poor to moderate agreement in classifying individuals as infrequent, moderate or heavier drinkers, suggesting that such classifications should be used cautiously. These results suggest that in general population studies a dietary quantity-frequency questionnaire and a diet diary covering several weeks can measure moderate levels of alcohol consumption similarly. The degree of consistency suggests that, for moderate levels of intake, measures of alcohol consumption derived from the two methods are equally valid.