Is self-reported daylight exposure reliable? A validation study with actigraphy in urban environments

Daylight exposure is an environmental determinant of human health, yet measuring real-world exposure at scale remains a major challenge. Large population studies frequently rely on self-reported time spent outdoors as a proxy for daylight exposure, although the validity of this measure remains uncertain. We evaluated the agreement between daily self-reported outdoor daylight exposure and objective light measurements obtained from wearable sensors and investigated whether depressive symptoms influence this agreement. A total of 172 adults living in two Brazilian metropolitan regions were monitored for up to 44 consecutive days, yielding over 5,800 matched days of subjective and objective exposure data. Objective measures included device output illuminance, photopic illuminance, and melanopic equivalent daylight illuminance (EDI). Agreement was assessed across the light exposure thresholds and durations using Lin's concordance correlation coefficient and ROC analyses. Moderate agreement was observed between self-reported and actigraphy-based daylight exposure. The highest concordance occurred at light levels consistent with outdoor daylight (>1000 l× for photopic illuminance and melanopic EDI; >750 l× for device illuminance). A minimum of 30 min of outdoor daylight exposure was the duration most accurately self-reported. Depressive symptoms did not affect these agreement patterns. These findings support the use of simple self-report questionnaires to estimate outdoor daylight exposure in clinical applications and in studies examining how environmental factors shape health.