Our knowledge of the association between alcohol intake and alcohol-related health outcomes depends, to a large extent, on the validity and reliability of self-reported alcohol intake. Weekly drinking measures are frequently used in epidemiological surveys, but it has been shown that respondents have problems in correctly reporting intake for a full week. The aim of this study is to investigate whether a beverage-specific question implies better recall and, thereby, eliminates or diminishes the previously reported association between the recall period and the self-reported weekly alcohol intake.
The data is derived from the Danish Health Interview Survey 2005, which is based on a region-stratified random sample of 21,832 Danish citizens aged =16 years (response rate: 67%). The data were collected via face-to-face interviews.
A beverage-specific question on alcohol intake on each day during the last week did not alter the strong association between the recall period and self-reported alcohol intake. However, the overall self-reported alcohol intake increased substantially when using the beverage-specific question instead of asking for the overall alcohol intake on each day. Moreover, the analyses indicated that interviews on Sundays should be avoided if the purpose is to assess alcohol intake for the previous day (Saturdays).
It seems problematic to recall alcohol intake even when the recall period is as short as 1 week. Weekly drinking measures should primarily be used when the main aim of the study is to assess the average volume of alcohol intake in a specific population.