Duncan, Glen EMiller, Isabel Catherine2013-11-142013-11-142013-11-142013Miller_washington_0250O_12327.pdfhttp://hdl.handle.net/1773/24125Thesis (Master's)--University of Washington, 2013Background: The health benefits of regular physical activity are well established. However, the contribution of natural and "built" environmental factors, including access to green space, to physical activity levels remains uncertain. In this study we use walking, moderate-to-vigorous physical activity (MVPA), public transit use, and neighborhood green space data from twin pairs in Washington State to elucidate this relationship. Twin pairs were used in this analysis to control for the genetic contribution to physical activity participation and self-selection into environments with more (or less) green space. We hypothesized that an increased density of green space around a subject's home would be associated with an increase in overall physical activity, controlling for demographic factors and public transit use. Methods: Data were obtained from 2,244 same-sex twin pairs from the community-based University of Washington Twin Registry. Green space around each twin's home was measured using the Normalized Difference Vegetation Index (NDVI), which ranges from -1 (barren areas of rock, sand, or snow) to +1 (temperate and tropical rainforests). Unadjusted and adjusted GEE and twin-difference regression models were constructed to determine the association between NDVI and physical activity levels. Results: There was a significant association between NDVI and both total walking levels in the neighborhood and total MVPA in unadjusted models (all p < 0.001). The associations were attenuated, but still significant (all p < 0.001), when controlling for age, income, education and ethnicity/race with no further changes when additionally controlling for transit use. However, there were no significant associations between NDVI and each of the activity outcomes in the twin-difference models (all p > 0.05). The results were unchanged when stratified by zygosity. Conclusion: The findings of this study demonstrate that higher levels of objectively measured green space are associated with more walking and MVPA overall. However, this relationship is confounded by genetic and shared environmental factors, suggesting that physical activity levels and access to green space are not causally related.application/pdfen-USCopyright is held by the individual authors.green space; ndvi; physical activityPublic healthnutritional sciencesThesis