Associations between physical activity and bone structure in older adults: does the use of self-reported versus objective assessments of physical activity influence the relationship?

Associations of current and previous physical activity (PA) with bone health are unclear. In postmenopausal women with low bone mineral density (BMD), current PA was positively associated with femoral neck BMD and microarchitecture. Past PA was positively associated with tibial microarchitecture. PA appears beneficial for bone health throughout the lifespan. Introduction To compare associations of current and past self-reported bone-specific physical activity, and current accelerometer-determined physical activity (PA), with bone structure (bone mineral density [BMD] and microarchitecture) in postmenopausal women with osteopenia or osteoporosis. Methods Fifty community-dwelling postmenopausal women (mean age 64.4 +/- 7.7) with hip or spine BMD T-score < - 1.0 SD were recruited for an exercise intervention. At baseline, current, past and total Bone-specific Physical Questionnaire (BPAQ) scores were self-reported, and percentages of sedentary, light and moderate to vigorous PA (MVPA) were objectively determined by accelerometer measurements. Bone structure was assessed by lumbar spine and hip dual-energy X-ray absorptiometry (DXA), 3D modelling algorithms (3D-SHAPER) of hip DXA scans and distal tibial high-resolution peripheral quantitative computed tomography (HR-pQCT) scans. Results Current BPAQ scores and MVPA were significantly positively associated with femoral neck areal BMD (beta = 0.315, p = 0.031 and beta = 0.311, p = 0.042, respectively) following multivariable adjustments. MVPA was also positively associated with femoral cortical surface BMD (beta = 0.333, p = 0.028) and mean cortical thickness (beta = 0.374, p = 0.013). Past and total BPAQ scores demonstrated positive associations with tibial trabecular number (beta = 0.391, p = 0.008 and beta = 0.381, p = 0.010, respectively), and negative associations with trabecular separation (beta = - 0.396, p = 0.006 and beta = - 0.380, p = 0.009, respectively) and distribution (beta = - 0.411, p = 0.004 and beta = - 0.396, p = 0.006, respectively). Current BPAQ score was positively associated with tibial cortical periosteal perimeter (beta = 0.278, p = 0.014). Conclusion BPAQ scores were most consistently associated with tibial bone parameters in older women, with past PA having lasting benefits for trabecular microarchitecture, and current PA positively associated with cortical bone.