A case-control study design was used to examine whether habitual physical activity prior to the final menstrual period (FMP) was associated with reduced risk of vasomotor and other symptoms during the perimenopausal period. Both cases and controls were identified through a screening interview with randomly selected women members, ages 48-52, of a large health maintenance organization. Cases (n = 82) were defined as women 3-12 months past their FMP who reported regularly having hot flashes or night sweats at least once a day or night during the 3 months following their FMP. Controls (n = 89) were of the same biologic age with respect to the FMP but reported vasomotor symptoms less than once a week during the reference time period. Neither cases nor controls had a history of hormone replacement therapy (HRT), hysterectomy, or bilateral oophorectomy. Case-control status, habitual physical activity (including recreational, housework, child care, and occupational activity), and psychological and somatic symptoms were assessed by self-report. Participation in vigorous recreational activity during the year prior to the FMP was not associated with reduced risk of frequent vasomotor symptoms after the FMP (odds ratio [OR] = 1.03 for a 50-unit increase in activity score, 95% confidence interval [CI] = 0.97-1.1). This lack of relationship was observed in all domains of activity. Factors that were associated with decreased risk included higher body mass index (BMI) (weight in kg/(height in meters)2) (OR = 0.95 per 1 unit increase in BMI, 95% CI = 0.90-1.00) and higher education (having a college degree relative to less education) (OR = 0.56, 95% CI = 0.40-0.80). Physical activity was also unassociated with reduced risk of psychologic distress, depressive feelings, or somatic symptoms, but, relative to controls, having vasomotor symptoms (being a case) was strongly associated with increased risk of experiencing those symptoms (OR ranging from 1.83 for psychologic distress to 2.84 for depressive feelings). These findings suggest that regular physical activity before the FMP may not reduce the likelihood of experiencing symptoms during the perimenopause, although the small sample size may limit the inferences that can be drawn.