W. Kwok1,2, S. Khalatbari-Soltani2,3, X. Dolja-Gore4, J. Byles4, J. Oliveira1,2, M. Pinheiro1,2, C. Sherrington1,2
1Institute for Musculoskeletal Health, The University of Sydney and Sydney Local Health District, Sydney, Australia, 2School of Public Health, Faculty of Medicine and Health, The University of Sydney, Sydney, Australia, 3ARC Centre of Excellence in Population Aging Research (CEPAR), The University of Sydney, Sydney, Australia, 4School of Medicine and Public Health, The University of Newcastle, Newcastle, Australia
Background: Physical activity (PA) positively affects bone health which could be a protective factor for fall-related fractures. However, PA may be a risk factor for falls due to increased exposure to situations where falls occur. Of note, falls and fall-related injuries could also impact PA. The bi-directional relationship between PA and falls and relationship between longer-term PA participation and falls are yet to be examined.
Purpose: 1) To examinebi-directional prospective relationship between PA and non-injurious and injurious falls; and
2) to identify the long-term PA participation patterns and their associations with subsequent non-injurious and injurious falls.
2) to identify the long-term PA participation patterns and their associations with subsequent non-injurious and injurious falls.
Methods: Data of the 1946-51 born women from the Australian Longitudinal Study on Women's Health; self-reported PA and falls were collected every 3-year-period between 1998 and 2019. Bi-directional prospective associations between PA and non-injurious and injurious falls were examined using generalised estimating equation models from 2001 survey (54 (1) years) to 2019 survey (71(1) years) (n=11,780). Latent class analysis was used to describe distinct profiles of PA participation patterns from mid-life (51(1) years, 1998 survey) to older age (68 (1) years, 2016 survey) (n=11,796) and their associations with non-injurious and injurious falls at 2019 survey, using multinominal logistic regression. All associations were adjusted for directed-acyclic graph-informed potential confounders (i.e., Accessibility Remoteness Index of Australia, education and ability to manage income).
Results: Compared to no PA, recommended amount of PA by World Health Organization (150-300 minutes/week) was associated with lower risk of prospective non-injurious (adjusted odds ratio (OR)0.87, 95% CI 0.80 to 0.96) and injurious falls (0.80, 0.73 to 0.87). Considering the impact of having falls on PA, compared to women without falls, women who had non-injurious falls and injurious falls were 16% less (95% CI 8% to 23% less) and 26% less (19% to 32% less) likely to be involved in prospective recommended amount of PA.
Five profiles of PA participation pattern were identified. Compared to profile “always active” (n=2621, 22%), “always inactive” (n=2143, 18%), “some activities but <150 minutes” (n=2147, 18%) and “increasing inactivity” (n=3540, 30%) were all associated with higher risk of non-injurious and injurious falls. After adjusted for potential confounders, profiles “always inactive” (1.56, 1.25 to 1.96) and “some activities but <150 minutes” (1.24, 1.00 to 1.53) remained associated with higher risk of non-injurious falls; and “increasing inactivity” remained associated with higher risk of injurious falls (1.39, 1.11 to 1.75). No associations were found between profile “decreasing inactivity” (n=1345, 11%) and non-injurious and injurious falls.
Five profiles of PA participation pattern were identified. Compared to profile “always active” (n=2621, 22%), “always inactive” (n=2143, 18%), “some activities but <150 minutes” (n=2147, 18%) and “increasing inactivity” (n=3540, 30%) were all associated with higher risk of non-injurious and injurious falls. After adjusted for potential confounders, profiles “always inactive” (1.56, 1.25 to 1.96) and “some activities but <150 minutes” (1.24, 1.00 to 1.53) remained associated with higher risk of non-injurious falls; and “increasing inactivity” remained associated with higher risk of injurious falls (1.39, 1.11 to 1.75). No associations were found between profile “decreasing inactivity” (n=1345, 11%) and non-injurious and injurious falls.
Conclusions: PA was prospectively associated with lower risk of non-injurious and injurious falls. Both non-injurious and injurious falls decreased the likelihood of subsequently reporting PA. Being constantly inactive or having increasing inactivity from mid-life to older age increased risk of subsequent non-injurious and injurious falls.
Implications: Bi-directional prospective associations were found between PA, and non-injurious and injurious falls. Being constantly active was associated with lower risk of non-injurious and injurious falls. PA promotion and fall prevention programs could promote healthy ageing in older adults and stop the “vicious cycle” of falls and inactivity.
Funding acknowledgements: WK was supported by the National Health and Medical Research Council Center of Research Excellence Prevention of Falls injuries
Keywords:
Accidental Falls
Physical activity
Older adults
Accidental Falls
Physical activity
Older adults
Topics:
Health promotion & wellbeing/healthy ageing/physical activity
Older people
Non-communicable diseases (NCDs) & risk factors
Health promotion & wellbeing/healthy ageing/physical activity
Older people
Non-communicable diseases (NCDs) & risk factors
Did this work require ethics approval? Yes
Institution: University of Newcastle and University of Queensland
Committee: Human Research Ethics Committees (HRECs)
Ethics number: Approval numbers H-076-0795 and 2004000224, respectively
All authors, affiliations and abstracts have been published as submitted.
