Background
Humans have an innate connection and attraction to nature, often referred to as ‘biophilia.’ Exposure to nature has been reported to have a positive impact on health, and several studies have documented that engaging in physical activity outdoors improves mental health and well being. Exact mechanisms remain unknown but evidence suggests that greater enjoyment and opportunities for more social interaction may contribute to the outdoor activity experience. No studies to date, however, have used objective measures of physical activity and location in order to better understand possible relationships. The relationship between physical activity location and health has not been studied extensively in older adults. Older adults are the least physically active segment of the US population with less than 3% meeting the recommendations of 150 minutes per week of activity. Additionally, they spend the most time engaged in indoor sedentary behaviors, such as watching TV. Older adults can benefit greatly from physical activity; improving physical, emotional and cognitive functioning. However, limitations in physical functioning, fear of falling and neighborhood design may prevent older adults from being active outdoors. Studies have shown that going outdoors can have long term health benefits for older adults. For example, older adults often have low levels of Vitamin D which is inversely related to time spent outdoors. Vitamin D deficiency is related to many chronic conditions including cancer, heart disease and bone health.

Objectives
To objectively assess the amount of outdoor physical activity in a sample of older adults living in retirement communities and to evaluate associations between time spent outdoors and physical, emotional and cognitive functioning.

Methods
Participants were sedentary older adults (n = 64; mean age 84 years; 72% female) living in three retirement communities in San Diego County who had not had a serious fall in the previous 12 months. Participants were asked to wear a hip-mounted ActiGraph GT3X+ accelerometer and Qstarz GPS device for 12 hours a day on 7 consecutive days. Data were recorded every 15 seconds. Participants also completed self report measures of quality of life, depression, and fear of falling. Participants underwent objective measures of physical functioning (SPPB) and cognitive functioning (Trails A & B).

Accelerometer and GPS data were merged using PALMS software, filtered for spurious data points. Indoor/outdoor time was calculated based on an algorithm that uses the ratio of GPS-detected satellites to the total number of orbiting satellites. Data were aggregated for each participant into total accelerometer counts (higher counts indicating more activity), total accelerometer counts outdoors, and total minutes outdoors. Data are expressed as means per day. Spearman rho correlations were used to assess the relationship between physical activity location and the health indicators.

Results
Minutes per day spent outdoors were not significantly related to any of the health indicators and the relationships were not in the direction that was expected. Total physical activity counts were significantly related to quality of life, cognitive function and fear of falling. Daily physical activity outdoors was related to the all the health indicators in the expected direction, but none of the relationships reached statistical significance.

Conclusions
This study confirms previous data that physical activity is related to health in older adults. The findings that outdoor activity or time spent outdoors were not related to health may indicate that location of activity may not be important. It is possible that some outdoor activities may expose older adults to pollutants that can adversely affect health. This study employed new techniques to objectively assess outdoor physical activity, and demonstrates the utility of such measures. Further analyses in large samples with validated algorithms may clarify these findings.

Support/Funding
This study was funded by the National Heart Lung and Blood Institute, NIH (R01 10152583).