UC Berkeley

China’s cities have been growing both in size and population at an unprecedented rate over the last three decades. The evolving urban landscape has important consequences for public health. However, the relationships among the physical environment, human behaviors, environmental exposures, and health are understudied in Chinese populations. Furthermore, more evidence from Chinese studies is needed to inform the design of urban environments and public health programs that promote and improve both mental and physical health.

This dissertation examines how urban development trends in China affect health and quality of life. I approached this question by conducting a cross-sectional socio-behavioral and health survey of 1608 adults in 20 neighborhoods in Xi’an, China in 2013. This cross-sectional study includes residents of four types of neighborhoods that represent different stages of China’s urbanization: work-units, lane and courtyard housing, and two forms of commodity housing (high-density high rises and low-density high rises) neighborhoods. Although cross-sectional in design, this dissertation leverages the temporal history of the neighborhoods present in Xi’an to explore the relationships of development trends with behaviors and health. In particular, I examine the relationships between the natural and built environments and urban health. In addition, I identify neighborhood-specific factors that public health practitioners and urban planners might target to improve health.

First, I apply land use regression (LUR) methodology and the deletion/substitution/addition (DSA) algorithm to select predictive models and create concentration surfaces for four pollutants: PM2.5, NO2, SO2, and O3. The LUR models identified substantial areas of Xi’an that had annual PM2.5, SO2, and NO2 concentrations exceeding current health standards set by the World Health Organization (WHO), providing more evidence for the potential health risks from ambient air pollution in Chinese cities.

Because consistent and reliable air quality monitoring networks are rarely able to keep pace with urbanization in China, new technologies are needed to complement the existing methods of environmental management in cities. Thus, I also test the validity of a new low cost particulate matter sensor (PUWP) for use in high concentration areas like Xi’an. The PUWP sensor performed well as compared to mature PM monitors and could be used to rapidly screen for air pollution “hotspots” in large areas where setting up extensive monitoring stations is challenging. The analysis also observed a sinusoidal relationship between sensor response and PM2.5 concentrations, indicating gradual saturation in the optical sensor’s ability to detect ambient concentrations in high PM environments above 300 µg/m3.

In addition, I present the results of the cross-sectional socio-behavioral and health survey where I examine the associations between self-reported perceptions of the built environment and quality of life, and assess whether these associations differ across the four types of neighborhoods. Neighborhood built environment was strongly associated with both mental and physical-health related quality of life in the commodity housing neighborhoods (high and low-density). In particular, pedestrian infrastructure, diversity of resources, access to and from the neighborhood, and neighborhood safety had the highest positive associations with increased mental health in the high-density high-rise neighborhoods. In the work-unit neighborhoods, increased access to and from the neighborhood was found to be a significantly associated with both mental and physical health. Pedestrian infrastructure, diversity of neighborhood resources, and esthetics were found to be positively associated with mental health in lane/courtyard neighborhoods.

Finally, results from the LUR analysis are also used in an exposure assessment of ambient air pollution for the 20 surveyed neighborhoods. I examine the role of neighborhood air pollution in modifying the associations between leisure-time physical activity (LTPA) and adverse health impact and quality of life. Neighborhood ambient air pollution is included in health effects models in two ways: 1) categorical single pollutant and 2) categorical mixtures models. Increasing LTPA levels are associated with lower odds of adverse health impacts and higher reported quality of life. However, the health and quality of life benefits of physical activity are potentially lower in areas where ambient PM2.5 and O3 are elevated. In addition, single pollutant models are poor proxies of mixtures of pollutants, which indicate a need for considering multi-pollutant exposures in epidemiological studies.

Collectively, these results suggest the built, natural, and social environments should be considered simultaneously as potential targets of intervention to improve quality of life and health in Chinese cities.