Gender-disaggregated longitudinal associations between health behaviours and life satisfaction in community-dwelling older adults: a six-year growth curve analysis of the CLSA

BACKGROUND: Health behaviours such as physical activity, sedentary behaviour, nutrition, sleep, alcohol consumption, and smoking are modifiable determinants of life satisfaction - an essential outcome for healthy ageing and well-being. We explored the longitudinal association between these health behaviours and life satisfaction among older adults in Canada. METHODS: The study was a secondary analysis of data from the Canadian Longitudinal Study on Aging, focusing on 11,215 adults aged 65 years and older at baseline, with follow-ups 1 and 2 at three-year intervals. Data were analysed using linear mixed-effects growth curve models, stratified by gender, with health behaviours decomposed into within- and between-person components. RESULTS: At baseline, the participants' mean (SD) age was 72.94 (5.64) years, and 54.2% were women. Women had lower life satisfaction than men across waves, but the difference was trivial (B=-0.24, p = 0.035, predicted difference [pd]=-0.24 points, effect size [d]=-0.04). Men showed a decline in life satisfaction over time, though the effect size was small (B=-0.09, p = 0.004, pd=-0.54 points, d=-0.09). Within-person analyses showed that higher alcohol use relative to one's usual pattern was associated with lower life satisfaction (B=-0.23, p = 0.037, pd=-0.92 points, d=-0.15), especially among men, with an appreciable effect size (B=-0.35, p = 0.024, pd=-1.40 points, d=-0.24). In contrast, improvements in nutritional risk (B = 0.04, p < 0.001, pd = 0.24 points, d = 0.04) and sleep duration (B = 0.22, p < 0.001, pd = 0.32 points, d = 0.05) were associated with negligible concurrent increases in life satisfaction. At the between-person level, higher average physical activity across the waves was associated with slightly higher life satisfaction (B = 0.01, p = 0.004, pd = 0.58 points, d = 0.10), particularly among men (B = 0.01, p < 0.001, pd = 0.58 points, d = 0.10), whereas higher average sedentary time was weakly associated with lower life satisfaction (B=-0.04, p < 0.001, pd=-0.33 points, d=-0.06). Lower average nutritional risk (B = 0.20, p < 0.001, pd = 1.22 points, d = 0.21) and higher average sleep duration (B = 0.35, p < 0.001, pd = 0.51 points, d = 0.09) were associated with higher life satisfaction, with nutrition showing a modest effect size. CONCLUSIONS: The findings demonstrate that behavioural changes and sustained health behaviours are associated with differences in older adults' life satisfaction, with gender-specific patterns, providing actionable evidence for individuals, healthcare practitioners, and policymakers to promote healthy behaviours.