Differences in twenty-four-hour profiles of blue-light exposure between day and night shifts in female medical staff

• Published in: The Science of the total environment 2019-02, Vol.653, p.1025-1033
• Main Authors: Rabstein, Sylvia, Burek, Katarzyna, Lehnert, Martin, Beine, Alexandra, Vetter, Céline, Harth, Volker, Putzke, Simone, Kantermann, Thomas, Walther, Jörg, Wang-Sattler, Rui, Pallapies, Dirk, Brüning, Thomas, Behrens, Thomas
• Format: Article
• Language: English
• Subjects: Artificial light at night (ALAN); Circadian disruption; Field study; Lifestyle; Light at night (LAN); Work exposure
• ISSN: 0048-9697; 1879-1026
• DOI: 10.1016/j.scitotenv.2018.10.293

Light is the strongest zeitgeber currently known for the synchronization of the human circadian timing system. Especially shift workers are exposed to altered daily light profiles. Our objective is the characterization of differences in blue-light exposures between day and night shift taking into consideration modifying factors such as chronotype. We describe 24-hour blue-light profiles as measured with ambient light data loggers (LightWatcher) during up to three consecutive days with either day or night shifts in 100 female hospital staff including 511 observations. Linear mixed models were applied to analyze light profiles and to select time-windows for the analysis of associations between shift work, individual factors, and log mean light exposures as well as the duration of darkness per day. Blue-light profiles reflected different daily activities and were mainly influenced by work time. Except for evening (7–9 p.m.), all time windows showed large differences in blue-light exposures between day and night shifts. Night work reduced the duration of darkness per day by almost 4 h (β^ = −3:48 hh:mm, 95% CI (−4:27; −3.09)). Late chronotypes had higher light exposures in the morning and evening compared to women with intermediate chronotype (e.g. morning β^ = 0.50 log(mW/m2/nm), 95% CI (0.08; 0.93)). Women with children had slightly higher light exposures in the afternoon than women without children (β^ = 0.48, 95% CI (−0.10; 1,06)). Time windows for the description of light should be chosen carefully with regard to timing of shifts. Our results are helpful for future studies to capture relevant light exposure differences and potential collinearities with individual factors. Improvement of well-being of shift workers with altered light profiles may therefore require consideration of both – light at the workplace and outside working hours. [Display omitted] Circular plots of 24-hour blue light profiles during day shifts (A) and night shifts (B) (~460 nm in mW/m2/nm): least square mean estimates of logarithmized light exposures per hour from multivariable statistical mixed models adjusted for season and LightWatcher device, back transposed to the original scale in mW/m2/nm, presented are exp(μ^) with 95% confidence limits. Please, note the different scale of the plots for day and night shift. •Twenty-four-hour blue light exposure profiles during periods with either day or night are described.•A new approach to select time-windows of the day for the description