0821 GPS-derived Green Space Exposure and Sleep Among Elementary School Children

Exposure to greenspace has been linked to improved well-being among children such as cognitive restoration and reduced stress. Greenspace is also associated with increased exposure to light, a zeitgeber of the circadian system. Studies examining the association between greenspace exposure and sleep, however, have been equivocal, largely due to methodology related to static measures of environmental context. We address this issue by using GPS-derived measures of greenspace that move beyond static residential contexts, and examine multiple measures of access to greenspace.

Schoolchildren in grades 1-3 in the ongoing Project Greenspace, Sleep, and Mental Health (G-SPACE) study in Rhode Island were recruited to wear a GPS and an accelerometer for seven days. GPS was used to identify stay points if participants spent 15 minutes in a space. We extracted and averaged Normalized Difference Vegetation Index (NDVI) values in 100m, 200m, 300m, 500m, and 1,000m circular buffers surrounding each participant's stay points to calculate greenspace exposure. Sleep measures were derived using actigraphy, and potential association with greenspace exposure was analyzed using a series of linear regression models.

Forty-nine participants had 398 valid days of wear-time. On average, participants wore devices for 6.2 days (SD=1.72). The mean sleep values included: Time in Bed, 9.63 hours; Total Sleep Time, 9.17 hours; Sleep Efficiency, 88.78%, and Wake After Sleep Onset (WASO), 37 minutes). Both TIB (hours) (100m β:1.98, p=0.01; 200m β: 1.87, p=0.01; and 300m β: 1.65, p=0.05) and TST (hours) (100m β: 1.94, p=0.01; 200m β:1.67, p=0.01; and 300m β: 1.37, p=0.05) showed positive associations with mean NDVI across smaller buffer distances. However, associations between Sleep Efficiency, WASO and greenspace exposure were not significant.

GPS-derived greenspace exposures in close proximity to daily activities are associated with total sleep time among elementary school children, with the strength of the association diminishing with each widening radius from the stay point. Our results have implications for urban planning and greening of neighborhoods, but additional analyses are needed with larger sample sizes and control for sociodemographic factors.

Project G-SPACE is supported by the National Institutes of Health/National Institute on Minority Health and Health Disparities (R01MD016241) and NIGMS 5P20GM139743