This study examined relationships between physical activity and sleep relative to leisure-time, household, and occupational physical activity in community-dwelling, older adults. From 3,000 randomly chosen community-dwelling, adults, aged 65-85 years, we recruited 509 eligible subjects (mean age 73.2 ± 5.1 years). We assessed nocturnal sleep duration, sleep onset latency and subjective sleep quality over the previous month through a questionnaire. Physical Activity Scale for the Elderly was used to assess leisure-time, household, and occupational physical activity. Items pertaining to sleep were expressed as dichotomous variables (good/poor), and each physical activity score was divided into two categories based on activity level. To investigate the relationship between sleep (dependent variable) and physical activity (independent variable), we used a logistic regression analysis, controlling for age, gender, living arrangement, depressive symptoms, and cognitive function. Prolonged sleep latency was significantly related to no participation in low intensity exercise (OR 2.14; 95% CI 1.42-3.21) and muscle strength exercise (OR 1.99; 95% CI 1.06-3.74). Our data suggest that not participating in low intensity exercise or muscle strength exercise may be associated with difficulty initiating sleep in older adults.