Validity of two questions on alcohol use in a health survey questionnaire.

The aim of this study was to investigate whether consumers of high and low levels of alcohol could be identified by two questions about alcohol use in a postal questionnaire survey. A sample of 2,300 persons aged 18 64 years from Stockholm county were sent a masked postal questionnaire comprising 30 questions about their health and functioning. Two questions concerned their alcohol consumption. One year later the subjects underwent a psychiatric health examination, which included an assessment of their alcohol use. The two questions about alcohol consumption identified high alcohol consumers with a relatively high sensitivity and specificity, of 64% and 87%, respectively, and thus are useful for identifying high alcohol consumers in health surveys using questionnaires.