Background:​ Physical activity of high and rapid impact is recommended to prevent and manage osteoporosis. However, current understanding of the associations of habitual impact activity and bone health may be influenced by the differences in objective and subjective physical activity assessment. It is also unclear how impact exercise affects bone structure when bone mechanosensitivity changes across the lifespan. Self-managed, home-based interventions may reduce barriers to impact exercise, but their feasibility and efficacy are unclear. This thesis aimed to further understand the relationship between impact physical activity and skeletal health across the lifespan.

Methods:​ This thesis included several observational studies, a systematic review and meta-analysis, and an interventional study. The Bone-specific Physical Activity Questionnaire (BPAQ) was self-administered in a study of 50 postmenopausal women with low bone density, and load ratings associated with the BPAQ were used to obtain loading scores from questionnaires which assessed energy expenditure in the Raine study (2868 young adults assessed at ages 17 and 20 years) and the Concord Health and Ageing in Men Project (CHAMP;​ 1705 older men aged ≥70 years) in order to determine associations between mechanical loading of self-reported physical activity and bone health. The systematic review included 28 randomised controlled impact exercise trials which examined changes in 3D-imaged bone structure at clinically relevant sites across the lifespan. A 16-week home-based impact exercise intervention recruited 50 postmenopausal women with low bone mass to assess the feasibility, safety and efficacy of the program.

Results:​ Current subjectively and objectively assessed physical activity was positively associated with femoral neck areal bone mineral density (BMD) and cortical bone structure in postmenopausal women with low bone mass, and historical BPAQ scores were positively associated with trabecular microarchitecture (assessed by high-resolution peripheral quantitative computed tomography). In young adults in the Raine study, physical activity of higher loading had stronger positive associations with whole-body and leg bone mass, while higher energy expenditure was more strongly associated with arm bone mass. Older men in the CHAMP study who engaged in physical activity of higher loading maintained higher hip and spine areal BMD, while higher energy expenditure was associated with reduced falls risk. From the systematic review and meta-analysis of impact exercise trials, positive site-specific bone adaptation in children and adolescents was limited to cortical bone at the proximal radius. Comparatively, impact exercise in postmenopausal women improved total and trabecular volumetric BMD at the distal tibia. Lastly, a self-managed, home-based impact exercise intervention in postmenopausal women with low bone mass was feasible, safe and effective in improving femoral neck areal BMD, cortical structure, and physical function.

Conclusion:​ These findings serve as an impetus for investigators to utilise estimates of loading scores, which can be obtained via the validated BPAQ or adapted from existing energy expenditure questionnaires, in future studies where associations between physical activity and bone health are of interest. Rigorous methodological designs of impact exercise trials are necessary when evaluating its effects on bone structure, especially with the ascendency of 3D bone imaging in musculoskeletal research. Given that impact activity can feasibly be remotely prescribed and performed, future trials should also explore the effectiveness of remotely-delivered exercise interventions for maintaining and improving bone health across the lifespan.