Alcohol intake assessment: the sober facts.

Recent recommendations in regard to the level of alcohol intake have mainly been based on epidemiologic studies which relied on self-reported amounts of alcohol consumed. Therefore, it is important to be aware of the quality of self-reported measures of alcohol intake. Alcohol intake assessment methods were reviewed with respect to their capacity to rank individuals according to alcohol intake and their ability to explain the variation in the level of intake in population samples. In 33 methodological papers published after 1984, alcohol intake was assessed by five main methods: quantity frequency, extended quantity frequency, retrospective diary, prospective diary, and 24-hour recalls. The mean level of alcohol intake differed by 20% between these methods. It was also found that when researchers asked specifically about intake of beer, wine, and liquor, this resulted in 20% higher estimates of intake. These percentages were similar among populations with low and high mean alcohol consumption (4 vs. 10 drinks per week). It was found that ranking of individuals according to intake was satisfactory, with weighted correlation coefficients between methods ranging from 0.63 to 0.73. The authors conclude that, when there is sufficient evidence that alcohol intake is underestimated in a population, methods that enquire about both the frequency and amount consumed, for beer, wine, and liquor, separately, will yield the most realistic levels of intake.