Published online by Cambridge University Press: 02 January 2007
To evaluate the reliability of alcohol consumption measured by five different dietary methods in the Catalan Nutrition Survey (1992–1993).
The different questionnaires used were: (1) two 7-day food records, (2) five items addressing alcoholic beverages in a 76-item food-frequency questionnaire, (3) a 1-week recall of 13 alcoholic beverages, (4) a 13-item alcoholic beverages frequency questionnaire and (5) two 24-hour recalls.
Ninety-three of the 120 adults (aged 20 to 70 years) initially recruited completed the questionnaires and provided a peripheral blood sample.
Mean daily alcohol intakes as assessed by the questionnaires were very similar. Beer was the alcoholic beverage most frequently consumed, followed by wine and liquor. Eighty-two per cent of the study population consumed less than 20 g of alcohol per day. No biochemical parameters were significantly correlated with alcohol intake estimated from the questionnaires.
We found a satisfactory level of reproducibility and validity in the pattern of alcohol consumption across different levels and types of alcoholic beverage intake. We also found that the self-administered 13-item questionnaire and the 1-week recall were the best techniques to measure moderate or low alcohol consumption, suggesting that the two methods are the most suitable to assess overall alcohol intake in the general population.