Pre-awake light exposure and sleep disturbances: findings from the HEIJO-KYO cohort

Abstract

Objectives: To evaluate the associations between pre-awake light (PAL) exposure and subjective and objective measures of sleep quality.

Methods: In this cross-sectional study of 1108 elderly participants (mean age, 71.9 years), we measured bedroom light intensity using a bedside light meter for two nights and sleep quality using actigraphy and a questionnaire. PAL was determined as the 2h-average light intensity before rise time, and sleep disturbance was defined as the Pittsburgh sleep quality index score ≥6.

Results: Sleep disturbance prevalence increased with increased PAL exposure (P = 0.002). In multivariable models adjusted for potential confounders, the odds ratio (OR) for sleep disturbances was significantly higher in the highest quartile PAL group (Q4) than in the lowest quartile group (Q1) (OR, 1.65; 95% CI, 1.16-2.34). This association occurred independent of post-bedtime light exposure; and was stronger in the later chronotype group (n = 556) (OR, 1.80; 95% CI, 1.05-3.09) than in the earlier chronotype group (n = 552) (OR, 1.64; 95% CI, 1.01-2.64). Actigraphic sleep efficiency in the Q4 group was significantly lower by 2.6% (95% CI, 1.3-3.8) than that in the Q1 group. Moreover, longer wake after sleep onset by 7.5 min (95% CI, 2.8-12.2) and sleep onset latency by 0.2 log min (95% CI, 0.1-0.4) were observed in the Q4 group than the Q1 group.

Conclusions: Higher PAL exposure was significantly associated with a higher prevalence of sleep disturbances, independent of post-bedtime light exposure. Consistent results were observed in the actigraphy analysis.

Keywords: Actigraphy; Circadian rhythm; Light at night; Pre-awake; Sleep.