AIMS: (1) To estimate the impact of drinking patterns on negative social (behaviour) consequences. (2) To test for the additional impact of overall volume of alcohol consumed on these social consequences. (3) To explore whether the impact on social consequences of drinking patterns is comparable for measures that do and do not explicitly mention alcohol consumption. DESIGN AND SETTING: An 8-year follow-up to a 1987 study of the Swiss general population carried out through face-to-face interviews; the follow-up data presented in this article was collected in 1995 through a mailed questionnaire. PARTICIPANTS: Nine hundred and fifty-three respondents from the 1987 survey who also completed the mailed questionnaire in 1995. MEASUREMENTS: Variables used were as follows: volume of drinking, eight drinking patterns differentiated by volume and frequency, four social consequences without mention of alcohol and six with mention of alcohol. All multivariate analyses control for sex, age and linguistic region. FINDINGS: High volumes of drinking per occasion predicted negative social consequences independently of overall drinking volume. This finding was independent of explicit mention of alcohol in item formulation. For unemployment and accidents, if assessed independently of alcohol consumption, no significant relationship with either overall volume of drinking or drinking pattern was observed. CONCLUSIONS: Patterns of alcohol consumption are an important determinant of social problems. Future research with better design is necessary to establish their exact risk relations with different social consequences.