Greenspace has been shown to have salutogenic effects on physical and mental health in observational and experimental studies. However, the pathways through which public and private greenspace influence health may differ. This study investigates the relationship between public and private greenspace and mortality in the 20-county Metropolitan Atlanta region. Generalized estimating equations with a negative binomial distribution and a log link function were used to model the association between counts of deaths at the census tract level (stratified by age, sex, race and ethnicity and standardized by population size) and the percentage of that census tract covered by either total greenspace or public greenspace. Strata with 1-4 deaths were censored to protect confidentiality. Multiple imputations were used to account for censored data. When controlling for relevant census-tract level confounders, a marginally insignificant association was observed between percent greenspace coverage and mortality rate (IRR=0.97 per 10 percentage point increase in greenspace, 95% CI: 0.94, 1.00) and a marginally significant trend of decreasing IRR with increasing quintile of greenspace exposure (p=0.03). No association was detected between amount of public greenspace and mortality rates. Greenspace may be a tool for regional planners to positively influence population health. Future studies should examine the particular association between public greenspace and population health to inform planning practice and public policy.
Table of Contents
This table of contents is under embargo until 21 May 2020
About this Master's Thesis
|Committee Chair / Thesis Advisor|
|File download under embargo until 21 May 2020||2018-04-25||File download under embargo until 21 May 2020|