Light Exposure and Melatonin among Rotating Shift Nurses—Response

We thank both Drs. Erren and Rabstein and colleagues for their interest in our research (1). We agree with Dr. Erren about the importance of including chronotype in the conceptual model of the shift work–light at night–melatonin relationship and that this variable should be considered as an important confounder or effect modifier in future analysis. He raises an interesting point about the potential for selection effects among rotating shift workers, where extreme chronotypes may be underrepresented because of the requirement for work at all points in the 24-hour day at some point during a rotating shift schedule. To appropriately consider this result with respect to other more extreme shift schedules, as he suggests, it will be necessary to ensure accurate classification of chronotype among shift workers. To this end, consideration of alternative tools for assessing chronotype, such as the Munich Chronotype Questionnaire (MCTQ) which considers sleep patterns on both work and free days when determining chronotype (2), may be worthwhile. Furthermore, a modified version of this questionnaire has been developed for shift workers due to the fact that sleep times on work and free days will vary substantially based on shift time for rotating shift workers (3). Accurate classification of chronotype for shift workers will be imperative in considering a role for chronotype in the shift work–melatonin pathway.

Dr. Rabstein and colleagues raise an important point concerning appropriate exposure metrics for light in the light at night–melatonin relationship and reasonably suggest that peak light levels may also influence melatonin levels, in addition to the influence of light captured in the average light measures used in our analysis (1). Our data did allow us to explore the influence on melatonin of peak light levels from 12 midnight to 5:00 am, the results of which are shown in the?Table 1. As in our previous study (4), no significant association between peak light exposure between 12 midnight and 5:00 am and melatonin levels was observed. However, as in our analysis of average light intensity (1), peak light levels in our study were relatively dim, so if there exists a threshold for the influence of light on melatonin, perhaps lighting conditions in our study were not sufficiently bright enough. Consideration of both these exposure metrics in future studies among individuals with wider ranges of light intensity exposure will be important in determining that which is the most appropriate characterization for light with respect to its influence on melatonin.

See original Letter to the Editor, p. 557