Abstract
The 36-item Short-Form Health Survey (SF-36) is widely used to assess health-related quality of life. However, only few studies have examined longitudinal SF-36 trajectories in general populations, and even fewer have explored their relationships with lifestyle factors, including sleep quality. We conducted a longitudinal analysis of SF-36 subscale trajectories among 910 Japanese individuals aged ≥ 60 years from 2007 to 2018 (4,799 records). Using latent class mixed models, we identified distinct trajectory patterns for each subscale. Notably, the role-physical (RP) and role-emotional (RE) subscales exhibited multiple patterns, including rapid decline and non-decline trajectories. We further examined predictors that differentiated between these two trajectory patterns-rapid decline and non-decline-with similar baseline scores. Consequently, poor performance on the open-eye one-leg standing test and higher depression scores on the Centre for Epidemiologic Studies Depression Scale predicted RP and RE declines, respectively. Importantly, poor sleep quality, as measured by the Pittsburgh Sleep Quality Index-particularly the total score and component 7 (daytime dysfunction)-was associated with subsequent RP and RE declines. In contrast, sleep timing variables (bedtime, sleep onset, and wake-up time) were not predictive. These findings suggest that preserving good sleep quality may help prevent future limitations in daily physical and emotional roles.