MEASUREMENT ERROR IN ALCOHOL-CONSUMPTION - THE SWISS HEALTH SURVEY

It is planned to include in the Swiss Health Survey (SHS) the question s on the consumption of alcohol that have been part of the Trend Surve ys of the Swiss Institute for the Prevention of Alcohol and Drug Probl ems (SIPA). Since both surveys use different questioning methods (SHS: telephone interview/self-administered questionnaire; SIPA Trend Surve ys: personal interviews) a comparative study was carried out to test p ossible effects of the methods used on responses about alcohol consump tion. A split-sample design was used, whereby 1097 persons were asked about their alcohol intake by personal interview, and 1154 persons com pleted a self-administered questionnaire with the same questions. Cons iderable differences were found: on an average, consumption reported b y personal interview was 38.5% higher among males (including abstainer s: 35.9%) and 18.0% higher among females (including abstainers: 17.4%) than that reported by self-administered questionnaire. Consumption fi gures as assessed by personal interview correspond better to the per c apita consumption figures as estimated by sales. Preliminary conclusio ns concerning the design of future surveys of alcohol consumption in S witzerland are presented.