Longitudinal assessment of lifestyle factors associated with nonrestorative sleep in Japan

Objective: Nonrestorative sleep (NRS) is common in the general global population. This study surveyed the incident/remission rate of NRS and identified related lifestyle factors in a general Japanese adult cohort.Methods: Data from a Japanese health check-based cohort between the fiscal years 2011 and 2018 were obtained. NRS was assessed through a single item question, and socio-demographic and clinical data were assessed through self-reports. To identify potential correlates of NRS incidence/remission, Cox regression analysis (or survival analysis) was performed.Results: Among the 135 609 individuals surveyed at baseline, data from 93 548 were considered for analysis. NRS exhibited incidence and remission rates of 11.3% and 15.4%, respectively. New NRS incidence was predicted by eating fast (hazard ratio [HR]: 1.07, 95% confidence interval [CI]: 1.04-1.10), dining close to bedtime (HR: 1.33, 95% CI: 1.27-1.38), midnight snacking (HR: 1.09, 95% CI: 1.04-1.13), skipping breakfast (HR: 1.16, 95% CI: 1.10-1.22), non-regular exercise (HR: 1.12, 95% CI: 1.07-1.17), lower levels of physical activity (HR: 1.19, 95% CI: 1.14-1.24), and slow walking speed (HR: 1.34, 95% CI: 1.29 -1.39), while heavy drinking was less predictive of NRS. However, the same factors (except for heavy drinking) were found to predict NRS remission.Conclusions: We found that unhealthy lifestyle behaviors, particularly dining close to bedtime and slow walking speed, were positively associated with NRS incidence, while they were negatively associated with NRS remission. These results can be useful in developing more effective health promotion policies to achieve good sleep.