Measurement of lifetime alcohol intake: utility of a self-administered questionnaire

Prior epidemiologic research revealing cardioprotective effects of alcohol intake has systematically neglected lifetime exposure to alcohol, which may cause serious bias in conclusions regarding drinking and heart disease risk. Departing from use of an earlier interview schedule, the authors of the present 1996-2001 cohort study developed a self-administered Lifetime Drinking History questionnaire (LDH-q). A total of 16,211 Dutch men and women older than age 45 years participated by completing the baseline questionnaire. A random sample of 3,255 men and women was used to determine the reliability and validity of the LDH-q. Test-retest reliability was assessed by means of the intraclass correlation coefficient and kappa scores. Correlations between lifetime and current intake scores were used to assess discriminant and convergent validity. Both reliability and validity appeared to be reasonably high compared with results obtained by using interview formats to measure lifetime alcohol intake. Reliability of the LDH-q was higher for men than for women, probably because of the more frequent and regular drinking of men. Indices of validity were similar for men (0.75) and women (0.70). Results show that the LDH-q can be a useful instrument in large-scale cohort studies.