RT Journal Article SR Electronic T1 OP42 Association of physical activity with body size and composition in pre- and postmenopausal UK Biobank women: a cross-sectional study JF Journal of Epidemiology and Community Health JO J Epidemiol Community Health FD BMJ Publishing Group Ltd SP A23 OP A23 DO 10.1136/jech-2014-204726.45 VO 68 IS Suppl 1 A1 W Guo A1 KE Bradbury A1 V Beral A1 GK Reeves A1 TJ Key YR 2014 UL http://jech.bmj.com/content/68/Suppl_1/A23.1.abstract AB Background UK Biobank provides a unique opportunity to study the association between physical activity and various body composition measures. We examined the relationship between physical activity and both total and central body fat separately for premenopausal and postmenopausal women, since previous studies suggest menopause-related changes in fat accumulation and distribution. We also assessed whether higher overall levels of physical activity were associated with lower fat percentage and absolute mass, for a given body mass index category. Methods A total of 40,006 premenopausal and 81,407 postmenopausal generally healthy women aged 39–70 in UK Biobank were included in this cross-sectional study. Self-reported information on frequency and duration of low, moderate, and vigorous physical activity was collected by touchscreen questionnaire. Physical activity was calculated as metabolic equivalent task hours per week (MET-hours/week) and participants were categorised as having low (<10 MET-hours/week), moderate (10–49.9 MET-hours/week), and high (>50 MET-hours/week) levels of physical activity based on International Physical Activity Questionnaire guidelines. Body size was measured by trained technicians. Overall adiposity was assessed by body mass index (BMI) as well as body fat percentage and mass (bioelectrical impedance). Central adiposity was represented using waist circumference, waist-hip ratio, trunk fat mass, and trunk fat percentage. Associations between body composition measures were analysed using multiple linear regression. We adjusted for age at recruitment, educational qualifications, quintiles of Townsend deprivation index, parity, hormone replacement therapy use, and smoking status. Stata 13 (StataCorp LP, Texas, USA) was used for all statistical analyses. Results The median (IQR) physical activity was 29.1 MET-hours/week (13.9–54.2) for premenopausal and 31.8 MET-hours/week (15.5–62.0) for postmenopausal women. All measures of total and central adiposity decreased with increasing physical activity. An increase in physical activity from low to moderate or moderate to high was associated with an average 0.94 unit lower BMI (95% CI 0.88–1.01) for premenopausal and 0.82 unit lower BMI (95% CI 0.77–0.86) for postmenopausal women. For a given BMI category, those with a moderate level of physical activity had a lower body fat percentage than those with a low level and the group with the highest level of physical activity had the lowest body fat percentage for a given BMI category. Conclusion Both total and central adiposity decrease with increasing levels of physical activity. For a given BMI category, greater physical activity is associated with decreased body and trunk fat mass and percentage.