Green space as preventive medicine : investigating the relationship between green space and life-threatening diseases using spatial statistics

Abstract

The leading causes of death, including non-communicable diseases (NCDs) and communicable diseases (CDs), have caused huge losses to human life globally. Extensive studies have shown that green space benefits NCDs such as cardiovascular and chronic respiratory diseases. However, few studies examined the relationships between various types of green space and NCDs across urbanicity at a national scale using spatially explicit green space measures. And yet, the relationships between green space and CDs, such as COVID-19, remain unclear. Moreover, most studies have used non-spatial models to evaluate green space-health associations. Understanding these associations using more advanced exposure assessment methods and spatial statistics at a national scale is necessary for reducing the global disease burden, improving planning and design practices, and promoting the development of healthy cities and communities. This dissertation aims to: 1) investigate the associations between various types of green space and the prevalence of cardiovascular diseases across urbanicity using a nationwide tract-level study; (2) investigate the nationwide associations between various types of green space and the prevalence of respiratory disease across urbanicity using a nationwide tract-level study; (3) investigate the associations between various types of green space and COVID-19 outcomes using a nationwide county-level study. Data of this dissertation are collected from the Centers for Disease Control and Prevention (CDC) and state- and local-level public health agencies for (1) the tract-level total number of respondents aged ≥18 years living with coronary heart disease (CHD), stroke, chronic obstructive pulmonary disease (COPD) and asthma diagnosed by a doctor, nurse, or other health professionals per 100,000 people in the contiguous United States in 2019; (2) the county-level total number of cases of COVID-19 infection and mortality rates in the contiguous United States in 2020. The negative binomial mixed-effect models and spatial autoregressive models were used to investigate the aims of this dissertation. The main findings are: (1) green spaces had different associations with NCDs and CDs due to different underlying mechanisms; (2) open space was most beneficial for cardiovascular and chronic respiratory diseases, while forest was most beneficial for COVID-19; (3) the beneficial association with NCDs remained in highly urbanized areas but disappeared in less urbanized areas; (4) exposure to forest within a moderate walking distance (1.2 km) is most effective at lowering the risk of COVID-19. Implications for policy, practice, and future research are discussed.