||Our research includes some strengths and limitations. The most important strength is objective measurement of light at night (LAN) intensity using a bedside light meter. Most previous studies evaluating the association between LAN and health outcomes have assessed indoor LAN levels using a self-reported questionnaire or outdoor LAN levels using satellite data; however, self-reported indoor LAN levels has not yet been validated with objective measurement and outdoor LAN levels are surrogates for an individual LAN exposure. The second strength of our study includes its longitudinal design using multivariable methods to adjust for confounders, which indicated LAN exposure may be a cause of the incidence of depressive symptoms. Indeed, the depressive score evaluated by questionnaires may be above or below the cut-off value over the short term; therefore, a long-term study considering such unstable outcomes should be conducted. In the current study, LAN exposure was measured for only two nights; thus, an amplitude of LAN intensity has been focused. However, multiple measurements over time in the future study would allow an analysis of fluctuations in LAN exposure, which might be important for circadian physiology.