Night work, light exposure and melatonin on work days and days off

ABSTRACT

We aimed to examine the effects of night work on salivary melatonin concentration during and subsequent to night work and the mediating role of light. We included 254 day workers and 87 night workers who were followed during 322 work days and 301 days off work. Each day was defined as the 24 hour period starting from the beginning of a night shift or from waking in the mornings with day work and days off. Light levels were recorded and synchronized with diary information (start and end of sleep and work). On average, participants provided four saliva samples per day, and these were analyzed for melatonin concentration by liquid chromatography tandem mass spectrometry (LC-MS/MS). Differences between day and night workers on work days and days off were assessed with multilevel regression models with melatonin concentration as the primary outcome. All models were stratified or adjusted by time of day. For light exposure, we estimated the total, direct and indirect effects of night work on melatonin concentrations obtaining 95% confidence intervals through bootstrapping. On work days, night workers showed 15% lower salivary melatonin concentrations compared with day workers (−15.0%; 95% CI: −31.4%; 5.2%). During the night, light exposure mediated a melatonin suppression of approximately 6% (−5.9%, 95% CI: −10.2%; −1.5%). No mediating effect of light was seen during the day time. On days off, we observed no difference in melatonin concentrations between day and night workers. These findings are in accordance with a transient and partly light-mediated effect of night work on melatonin production.

KEYWORDS:

• Night Work
• melatonin suppression
• light exposure
• field study

Acknowledgments

We would like to thank all the participants in the study; Marie Aarrebo Jensen, Anne Abiltrup, Inge Christensen, Dorrit Meincke and Ulla Tegner for analysis of melatonin in in saliva; Anja Jørgensen, Louise Brus Hesselvang, Anne Abiltrup, Inge Christensen, Dorrit Meincke and Ulla Tegner for collection of data. Jesper Medom Vestergaard and Morten Frydenberg are thanked for their skilful help with data management and analysis.

Funding

The study was funded by the Danish Working Environment Research Fund (02-2010-09).

Additional information

Funding

The study was funded by the Danish Working Environment Research Fund (02-2010-09).