Objective: To evaluate habitual levels of physical activity in a nationally representative sample of adults in Ireland. Design: Cross-sectional survey using a self-administered questionnaire. Usual levels of work, recreational and household activities were evaluated in relation to anthropometric, demographic and socio-economic characteristics. The amount and intensity of all activities were quantified by assigning metabolic equivalents, (METS) to each-activity. Setting: Republic of Ireland and Northern Ireland, 1997-1999. Subjects: Random sample of 1379 adults aged 18-64 years. Results: Men were approximately twice as active in work and recreational activity (139.7 +/- 83.9 METS) as women (68.5 +/- 49.8 METS; P < 0.001), but women were three times more active in household tasks (65.9 +/- 58.7 METS vs. 22.6 +/- 24.6 METS; P < 0.001). Overall levels of physical activity declined with increasing age, particularly leisure activity in men. In women the decline in work activity was offset by spending more time in household pursuits. Twenty-five per cent of the subjects were extremely overweight (body mass index (BMI) > 28 kg m(-2)) or obese (BMI > 30 kg m(-2)). Fewer obese subjects reported higher levels of work and leisure activities. However, a higher percentage of obese women reported participation in the higher levels of household activities. Participation rates in recreational activities were low. Walking was the most important leisure activity of both men (41%) and women (60%). In terms of hours per week spent in vigorous physical activity, men were more active than women, professional and skilled non-manual women were more active than women in other social classes, and younger subjects (aged 18-35 years) were more active than older subjects. Conclusions: The holistic approach used in the assessment of physical activity in this study has revealed important and subtle differences in the activity patterns of men and women. Failure to fully characterise the respective activity patterns of men and women could lead to ill-informed public health policy aimed at promoting and sustaining lifetime habits of physical activity. The results suggest that simple population-focused programmes to promote physical activity are unlikely to offer the same chance of long-term success as more sensitive and individualised strategies.