Bone mineral density and lifetime physical activity in South African women

We investigated the relation between lifetime physical activity and bone mineral density (BMD) in South African women using data collected in a case-control study of breast cancer in relation to BMD. Subjects (n = 144) were of black African or mixed ancestral origin, and <60 years of age (mean age 42.6 +/- 8.9 years). Cases had newly diagnosed breast cancer (n = 62) and controls were referred for conditions unrelated to BMD or breast cancer (n = 82). Physical activity data consisting of household, occupational and leisure-time activity, and activity for transport, were collected via questionnaire at 4 life stages (epochs), viz. 14-21, 22-34, 35-50, and 50+ years of age. Total energy (MET hrs) and peak strain scores were calculated. Lumbar spine and total proximal femur BMD were measured using dual-energy x-ray absorptiometry. BMD measures were similar between groups, therefore data were combined. BMD measures were unrelated to total lifetime physical activity. However, the major determinants of total proximal femur BMD included age, transport activity including walking and bicycling between the ages of 14 and 21 years, and current weight (adjusted r2 = 0.33, P < 0.0001). The major determinants of lumbar spine BMD included age, household energy expenditure between the ages of 14 and 21 years, and current weight (adjusted r2 = 0.23, P < 0.0001). Total peak bone strain score for activities between 14-21 years of age was also significantly correlated with lumbar spine BMD ( r = 0.18, P < 0.05). Intraclass correlation coefficients to assess tracking of activity through epochs 1, 2, and 3 were high for total energy expenditure (0.96; 95%CI: 0.94-0.97), household (0.98; 95%CI: 0.97-0.99) and occupational activity (0.78; 95%CI: 0.71-0.84) and activity for transport (0.92; 95%CI: 0.89-0.94). These data suggest that walking or activities resulting in impact loading at a young age are associated with higher BMD in later years. In addition, our findings suggest tracking of physical activity over time.