We have conducted a multicentre case-control study to assess the epidemiological importance of previously suggested risk factors for psoriasis, including family history of the disease, smoking and alcohol consumption. Newly diagnosed psoriatics, with a history of skin manifestations no longer than 2 years were eligible as cases; as controls we selected subjects with newly diagnosed dermatological conditions other than psoriasis. Interviews were performed by trained medical investigators using a structured questionnaire. Two-hundred and fifteen cases, aged 16-65 years (median age 38), and 267 controls, aged 15-65 years (median age 36), were interviewed and included in the analysis. Family history was a risk factor for psoriasis; the multiple logistic regression (MLR) adjusted-odds ratio was 18.8 (95% confidence interval 6.4-54.8) for a history in parents, and 3.2 (95% confidence interval 1.5-6.6) for a history in siblings. The risk of psoriasis was higher for current smokers than for those who had never smoked. The MLR adjusted odds ratio was 2.1 (95% confidence interval 1.1-4.0) for people smoking 15 cigarettes or more per day. The risk of psoriasis was higher for alcohol drinkers: compared with teetotallers the MLR adjusted-odds ratios were 1.3 (95% confidence interval 0.8-2.3) for subjects drinking one or two drinks/day and 1.6 (95% confidence interval 0.9 to 3.0) for those drinking three or more. However, the trend in risk was not statistically significant. Our study confirms the role of family history in psoriasis and provides some evidence of a dose-response relationship for an association between smoking habits and psoriasis.