When etiological relationship is of interest, the incidence rate is a preferred measure. The aim of the present retrospective study was to estimate the incidence rate of self-reported hand eczema in a sample from the general population and to study the relation of this to age, sex, and atopy. A questionnaire was mailed to 3000 individuals aged 20-65 y, randomly selected from the population register of Göteborg, Sweden. This gave a response rate of 73.9%. Questions were asked about ever having had hand eczema, time of onset of the disease, history of childhood eczema, and history of asthma/hay fever. The crude incidence rate of self-reported hand eczema was 5.5 cases per 1000 person-years (females 7.1 and males 4.0). There was no difference, however, in incidence rate between women and men above 30 y of age. In a Poisson regression analysis, female sex, childhood eczema, and asthma/hay fever were all significantly associated with hand eczema, but only at ages below 30 y. A moderate influence of recall bias and a probable tendency to underreport imply that the incidence rates presented are to be considered as minimum rates.