Unpacking the temporal dynamics of physical activity and loneliness: a 1-year interval analysis

This data-driven commentary expands on Surkalim et al. by analyzing the bidirectional relationship between physical activity (PA) and loneliness using 22 years of data from the Household, Income, and Labor Dynamics in Australia (HILDA) Survey. Employing random intercept cross-lagged panel modeling (RI-CLPM) with annual intervals and a more nuanced measure of loneliness, our findings suggest significant within-person bidirectional effects: increases in PA were associated with reductions in loneliness 1 year later. In contrast, higher loneliness predicted declines in PA. Generally, the former effect was stronger, whereas both effects are rather small. However, they can build up to large effects over time. The significant negative between-person correlation between habitual PA and loneliness further supports the negative association at a general between-person trait level. Considering both the results of Surkalim et al. and the findings presented here highlights the crucial role of the time interval under examination. These findings suggest that promoting regular PA can be an effective strategy for reducing loneliness, and that reducing loneliness has a positive effect on PA. Although both effects are generally small, their bidirectional nature suggests they may accumulate over time, reinforcing improvements in both health behavior and well-being.