Cardiovascular disease, muscle function, and long-term falls risk: The Perth longitudinal study of ageing women

Background: A few cross-sectional studies have highlighted inconsistent associations between cardiovascular disease (CVD) and musculoskeletal conditions. We sought to investigate the relationship between clinical CVD including subtypes, compromised muscle function, as well as incident self-reported and injurious falls in older women. Materials and methods: 1431 community-dwelling older women (mean age ± SD; 75.2 ± 2.7 years) were included in over 14.5 years of a prospective study, the Perth Longitudinal Study of Ageing in Women. CVD (up to 18-years prior to the baseline visit) and injurious fall hospitalizations over 14.5 years were obtained from linked health records. Self-reported falls for five years were obtained via a written adverse event diary posted every four months. Timed-Up-and-Go (TUG) test and hand grip strength were used to assess mobility and muscle strength, respectively. Mobility impairment was defined as TUG performance >10.2 sec and muscle weakness characterized as grip strength <22 kg. Results: Over 5-years, 411 (28.7%) women reported a falls, while 567 (39.6%) were hospitalized due to an injurious fall over 14.5 years. Prior CVD events were associated with 32% (HR 1.32 95%CI, 1.06–1.64) and 29% (HR 1.29 95%CI, 1.07–1.56) increased risk of self-reported and injurious falls, respectively, in multivariable-adjusted models. When considering subtypes of CVD, only cerebrovascular disease was related to self-reported (HR 1.77; 95%CI, 1.15–2.72) and injurious falls requiring hospitalization (HR 1.51; 95%CI, 1.00–2.27). CVD was also associated with cross-sectional and prospective mobility impairments. However, no evidence for such relationships was observed for muscle weakness. Conclusions: Prevalent CVD events, particularly cerebrovascular disease, are related to an increased risk of long-term falls. These findings highlight the need to recognize increased falls risk in patients with CVD. Further, there is a need to understand whether incorporating prevalent CVD into falls screening tools improves risk stratification or affects model calibration.