Influence of the recall period on self-reported alcohol intake

Abstract

Objective: To estimate the accuracy of recall on self-reported alcohol intake. Design: Population-based random sample. Setting: The Danish Health and Morbidity Survey 1994. Subjects: The study is based on 6,354 persons chosen at random among the adult Danish citizens. The response rate was 77%. Main outcome measure: Number of alcoholic drinks consumed on each day during the last week. Results: For self-reported alcohol intake on the last Saturday, respondents with a recall period of one day have the shortest recall period (reference group) and respondents with a recall period of seven days the longest. Respondents with a recall period of seven days reported significant lower alcohol intake (Odds Ratio: 0.56). The reported alcohol intake declines when the recall period increases. The decline in recall of alcohol intake is very clear already after 2–3 days. Conclusion: The systematic decrease in reported alcohol use with increased recall period indicates problems in correctly reporting alcohol intake for a full week. Many surveys use recall periods that are longer than a week and, therefore, underreported alcohol intake is expected to be even higher.