460 Journal of Physical Activity and Health, 2010, 7, 460-464 © 2010 Human Kinetics, Inc. Leisure-Time Physical Activity: Association With Activity Levels in Other Domains Airton J. Rombaldi, Ana M.B. Menezes, Mario Renato Azevedo, and Pedro C. Hallal Objectives: To explore whether participation in leisure-time physical activity (LTPA) is associated with partici- pation in occupational, housework, and transport-related physical activity. Methods: Population-based cross- sectional study covering a multistage sample of 972 subjects age 20 to 69 years. Physical activity was measured using the long version of the International Physical Activity Questionnaire. A LTPA score was calculated as follows: min/wk of walking + min/wk of moderate-intensity physical activity + (min/wk of vigorous-intensity physical activity × 2). Similar scores were generated for each domain. For categorical analyses, the scores were divided into 3 categories: 0 min/wk, 10–149 min/wk, and ≥150 min/wk. Results: The proportion of subjects practicing less than 150 min/wk of physical activity in each domain was: leisure-time (69.8%), occupational (58.3%), housework (35.0%), transportation (51.9%). Subjects with a transport-related physical activity score equal to or above 150 min/wk were 40% less likely to be sedentary in leisure-time in comparison with those who did not practice transport-related physical activity. Housework and occupational physical activity were not related to participation in LTPA. Conclusions: Future physical activity campaigns should focus on other domains instead of LTPA alone, particularly supporting transport-related physical activity as a strategy of health promotion. Keywords: motor activity, public health, epidemiology, developing countries Rombaldi is with the Dept of Sports, Federal University of Pelotas, Brazil. Menezes is with the Dept of Clinical Medicine, Federal University of Pelotas, Brazil. Azevedo and Hallal are with the Post-graduate Program in Epidemiology, Federal University of Pelotas, Brazil. Current physical activity guidelines state that every adult should perform, at least, 30 minutes per day of moderate-intensity physical activity 5 days or more per week, or 20 minutes of vigorous-intensity physical activity 3 or more days per week. 1 Such level of activ- ity may be obtained by means of leisure-time physical activity (LTPA), but also by occupational, housework, and transport-related physical activity. Because recent studies have highlighted the impor- tance of addressing all domains of physical activity, 2–4 it is important to fully understand how activities practiced in one domain infuence activities performed in other domains. This literature gap is particularly important in low and mid- dle-income countries, in which activities practiced in other domains are more frequent than in developed countries. 4 The aim of the current study was to explore whether participation in LTPA is associated with participation in occupational, housework, and transport-related physical activity. Methods A population-based cross-sectional study was carried out in Pelotas, southern Brazil in the second semester of 2006. The target population included adults age 20 to 69 years living in the urban area of the city; 93% of the city’s population lives in urban areas. A multistage sampling strategy was used, to obtain a representative sample of households in the city. Brazil- ian cities are divided into delimited geographical areas named census tracts. Each tract comprises approximately 300 households. All the 404 census tracts in Pelotas were listed. A random sample of 40 tracts was selected. In each of the 40 sampled tracts, 15 households were systemati- cally selected. Every resident age 20 to 69 years in the sampled households was eligible for the study, except those with severe mental or physical impairment. Three separate sample size calculations were carried out before data collection. In the frst, we estimated the sample size necessary for evaluating the prevalence of insuffcient physical activity. The estimated prevalence was 35%, the margin of error accepted was 3 percent- age points, and a 95% confdence level was used. This calculation required a sample size of 576 individuals after accounting for an estimated 10% nonresponse rate and an estimated design effect of 1.5. To explore the association between insuffcient physical activity and the independent variables, the following parameters were used: confdence level of 95%, power of 80%, prevalence of exposures ranging from 10% to 90%, and odds ratios of 1.7 or greater. After accounting for nonresponse, design effect, and multivariable analyses (extra 15%), the sample size required was 884 subjects. Finally, to estimate the proportion of total physical activity explained by each of the domains (leisure-time, occupational, housework, and transport-related), we estimated a minimum proportion of