Can Special Light Glasses Reduce Sleepiness and Improve Sleep of Nightshift Workers? A Placebo-Controlled Explorative Field Study

Abstract

Nightshift workers go against the natural sleep-wake rhythm. Light can shift the circadian clock but can also induce acute alertness. This placebo-controlled exploratory field study examined the effectiveness of light glasses to improve alertness while reducing the sleep complaints of hospital nurses working nightshifts. In a crossover within-subjects design, 23 nurses participated, using treatment glasses and placebo glasses. Sleepiness and sleep parameters were measured. A linear mixed model analysis on sleepiness revealed no significant main effect of the light intervention. An interaction effect was found indicating that under the placebo condition, sleepiness was significantly higher on the first nightshift than on the last night, while under the treatment condition, sleepiness remained stable across nightshift sessions. Sleepiness during the commute home also showed a significant interaction effect, demonstrating that after the first nightshift, driver sleepiness was higher for placebo than for treatment. Subjective sleep quality showed a negative main effect of treatment vs. placebo, particularly after the first nightshift. In retrospect, both types of light glasses were self-rated as effective. The use of light glasses during the nightshift may help to reduce driver sleepiness during the commute home, which is relevant, as all participants drove home by car or (motor) bike.

Keywords: alertness; care professionals; rapidly rotating; shift work; short-wavelength light; sleep.

Figure 1

Estimated marginal means for the Karolinska Sleepiness Scale (KSS) in 2-h intervals, (A) during nightshift work, (B) after nightshift work, and (C) on the first two days and recovery days for the treatment and placebo condition. Higher values indicate higher levels of sleepiness. Error bars indicate the 95% confidence intervals of the estimated marginal means. D1 = first day before nightshift, D2 = second day before nightshift, FN = first night of nightshift, N = nightshift, LN = last night of nightshift, R1 = first recovery day, and R2 = second recovery day. For all time periods, Conditions did not significantly differ per day. Note that significant effects at the level of Time of day are not displayed. ** = p < 0.01.

Figure 2

Estimated marginal means for Driver Sleepiness (DSS) and Sleepiness (KSS) for the commute home for the treatment and placebo condition across nightshift days. Higher values indicate higher levels of sleepiness on both scales. Error bars indicate the 95% confidence intervals of the estimated marginal means. * = p < 0.05; ** = p < 0.01, *** = p < 0.001. FN = first night of nightshift, N = nightshift, LN = last night of nightshift.

Figure 3

Estimated marginal means (EMM) of bed time (BT), get-up time (GUT), time in bed (TIB), total sleep time (TST), sleep latency (SL), sleep efficiency (SE), fragmentation index (FI), and Groningen Sleep Quality Scale (GSQS) score across Daytype for the placebo and treatment condition. Error bars indicate the 95% confidence intervals of the estimated marginal means. D2 = night before nightshift, FN = first night of nightshift, N = nightshift, LN = last night of nightshift, R1 = first recovery day, and R2 = second recovery day. * = p < 0.05, ** = p < 0.01.

Figure 4

Mean hourly illuminance (log-transformed) across Daytype. Grey areas indicate nightshift work (~23:15–07:45). Note that the data presented in this plot were not adjusted for the transmission of the blue-blocking goggles and do not include light exposure from the light glasses. D1 = first day before nightshift, D2 = second day before nightshift, FN = first night of nightshift, N = nightshift, LN = last night of nightshift, R = first recovery day, and R2 = second recovery day.

Figure 5

Top: light glasses (treatment, placebo) and the blue-blocking goggles. Bottom left: spectral power distribution of the treatment light glasses, melanopic irradiance, and the spectral transmission of the blue-blocking goggles. Light measurements were performed at corneal position, using a realistic head model. Bottom right: radiometric and photometric properties of the treatment light glasses. Calculations through the CIE S 026 Toolbox—beta version E1.05 [39].

Figure 6

Top: overview of the study design, Bottom: exemplary light exposure protocol and measurement timings for a three-day measurement period. Participants started at 12:00 the day before (D1) the first nightshift, slept (black areas, D2) as normal and started their first nightshift (gray areas, end of D2). Light exposure (blue areas) was administered 4 × 15 min during nightshifts (FN, N, and LN) and for 30 min within 2 h after awakening. During the morning commute home, blue-blocking goggles were worn (orange areas). The measurement period ended at 12:00 two days after the last nightshift (R2). KSS (K) was measured every two hours during the wake period. The driver sleepiness questionnaire (D) was filled in before going to bed after nightshifts. The GSQS (G) and sleep diary were filled in directly after sleep. Objective sleep and person-bound light exposure were continuously monitored during the entire measurement period.