We conducted a population-based case-control study of 1,237 incident breast cancer cases and 1,241 controls in Alberta between 1995 and 1997 to examine the effect of physical activity performed at different ages and life periods on breast cancer risk. In this study, we measured all types of physical activity done throughout life with a questionnaire developed and tested specifically for this study. We found that breast cancer risk was most associated with a risk reduction for activity done later in life, particularly between menopause and the reference year, for which we observed an odds ratio of 0.70 (95% confidence interval = 0.52-0.95). We also stratified the study participants into four categories according to their patterns of physical activity performed before and after menopause. For the women who sustained physical activity throughout life vs those who were never active, we found an odds ratio of 0.58 (95% confidence interval = 0.41-0.83). This study suggests that sustained activity throughout life and particularly activity done later in life may have the most benefit in reducing breast cancer risk.