Short-wavelength light exposure at night and sleep disturbances accompanied by decreased melatonin secretion in real-life settings: a cross-sectional study of the HEIJO-KYO cohort

To determine the association between short-wavelength light exposure at night (LAN) power and sleep quality or melatonin levels in real-life settings.In this cross-sectional study of 580 older adults (mean age, 71.0 years), we measured short-wavelength LAN power at cornea level using an originally developed light loggers over two nights. Sleep quality and physiological melatonin levels were measured using the Pittsburgh sleep quality index (PSQI) questionnaire and overnight urinary 6-sulfatoxymelatonin excretion (UME), respectively.The first and second tertile short-wavelength LAN power values obtained were 0.77 × 10-2 μW/cm2 and 7.0 × 10-2 μW/cm2, respectively, and the overall prevalence of sleep disturbances was 34.7%. The mean UME was 1.88 ± 0.70 log μg. The mean global PSQI score and the prevalence of sleep disturbances significantly increased (P = 0.004 and 0.006, respectively) with increasing tertile groups of short-wavelength LAN power. In multivariable analysis adjusted for potential confounders, the odds ratio (OR) for sleep disturbances was significantly higher in the highest tertile group of short-wavelength LAN power compared with that in the lowest tertile group (adjusted OR, 1.90; 95% confidence interval [CI]: 1.20, 3.00; P = 0.006). In addition, UME was significantly lower in the highest tertile group of short-wavelength LAN power than that in the lowest tertile group (adjusted mean difference, -0.14 log μg; 95% CI: -0.28, -0.007; P = 0.039).Although short-wavelength LAN power measured at cornea level in real-life settings seemed to be significantly low, our findings suggest that short-wavelength LAN power is significantly associated with both melatonin levels and sleep disturbance.