Background
The aim is to investigate whether social isolation and loneliness are associated with changes in grip strength, gait speed, BMD, and fractures.
Methods
Canadian Longitudinal Study on Aging (CLSA) Comprehensive Cohort participants aged 65 years and older at baseline (2012–2015) who completed the three-year follow-up interview (2015–2018) were included in this analysis (n = 11,344). Social isolation and loneliness were measured using the CLSA social isolation index (CLSA-SII, range 0–10). We calculated absolute and percent change in grip strength (kg) and gait speed (m/s) and annualized absolute (g/cm2) and percent change in femoral neck and total hip BMD during the three-year follow-up. Self-reported incident fractures of all skeletal sites in the previous 12 months were measured at three-year follow-up. Multivariable analyses were conducted. Odd ratio (OR) and 95% confidence interval (CI) are reported.
Results
The mean age (standard deviation [SD]) was 72.9 (5.6) years and 49.9% were female. The mean (SD) of CLSA-SII at baseline was 3.5 (1.4). Mean absolute and percentage change (SD) in grip strength (kg) and gait speed (m/s) were -1.33 (4.60), -3.02% (16.65), and -0.05 (0.17), -3.06% (19.28) during the three-year follow-up, respectively. Mean annualized absolute (g/cm2) and percentage change (SD) in femoral neck and total hip BMD were -0.004 (0.010), -0.47% (1.43) and -0.005 (0.009), -0.57% (1.09), respectively. 345 (3.1%) participants had incident fractures. As CLSA-SII increased (per one unit change), participants had 1.13 (adjusted OR 1.13, 95% CI 1.01–1.27) times greater odds for incident fractures. The interaction term between the CLSA-SII and centre for epidemiology studies depression 9 scale (CES-D 9) for self-reported incident fractures was shown (interaction OR 1.02, 95% CI 1.00–1.04).
Conclusions
Socially isolated and lonely older adults were more likely to have had incident fractures, but social isolation was not associated with the three-year changes in grip strength, gait speed, or BMD.