Measurement of lifetime exposure to alcohol: data quality of a self-administered questionnaire and impact on risk assessment

Abstract

Measurement of alcohol consumption in epidemiologic research is usually limited to “current” drinking status, pertaining to consumption during a short reference period prior to baseline measurement. Use of such an indicator in the assessment of health risks can lead to biased results. It is therefore important (a) to measure exposure to alcohol over a longer period of time and (b) to evaluate possible effects different measures have on risk assessment. In the paper results are presented of a comparison of a standard lifetime drinking history interview and a self-administered mailed questionnaire, which can be used in large-scale prospective studies. In the second part, the data on lifetime consumption were used to simulate the effect that use of a lifetime consumption indicator instead of a current consumption indicator would have on an assumed J-shape risk curve. Our results show that dose-response relationships will be affected when the correlation between current and lifetime consumption drops below .50.