Researchers are increasingly seeking to understand the potential impacts of local neighborhoods on public health (Sallis et al. 2006, Handy 2005). Indicators for which measurements are being developed include population and employment densities; local exposure to hazards (e.g., pollutants generated by road traffic); the availability, quantity, quality, and accessibility of physical resource activities within a neighborhood; the availability and accessibility of transportation; the integration of residential and commercial land uses; the availability and quality of food resources (e.g., groceries, convenience stores, fast food); and the availability and accessibility of health services. Examining spatial relationships at this scale requires a level of geographical detail that can be acquired either by field surveys, which are expensive, and/or by using locally available geospatial data for both inventorying neighborhoods (e.g., parks, schools, land use) and for geocoding businesses, services, health records, and research participants (Brennan Ramirez et al. 2006).
Most geospatial data that allows analysis at the urban/ suburban neighborhood level tends to be locally produced data developed by cities and counties for purposes other than health research, including infrastructure management, land-use planning, or tax assessment. The structure and content of local geospatial data can vary widely by jurisdiction. It would be ideal to use nationally available geospatial data to support health research at the neighborhood level to easily enable comparative studies across cities, regions, and states. However, there are many instances in which national data does not exist for the indicators needed (e.g., land use), or the data that exists (e.g., roads from the Census TIGER/Line file) is not accurate enough to support the measurements of interest. Developing a geospatial database to support health research at the neighborhood scale, therefore, requires extensive knowledge of both national and local geographic information system (GIS) data sets, their accuracy, content, and quirks.
THE HEALTH IS POWER (HIP) PROJECT
This paper discusses the development of a geospatial database to support the Health is Power (HIP) project, a study funded by a National Institute of Health R01 grant (1R01CA109403). HIP is a multisite intervention study examining the effect of a social cohesion intervention on physical activity and nutrition behavior of African-American and Hispanic women. A key research question in this study is whether the effectiveness of the intervention varies by characteristics of a participant's neighborhood environment. The study is ongoing as of June 2007 and is being conducted in Houston (Harris County) and Austin (Travis County), Texas. The goal is to recruit 240 women between the ages of 25 and 60 years of age in each county (African-Americans in Harris County and Latinas in Travis County), using community partners (primarily churches). Participants in each county are randomized into two groups--one group forming teams for the physical activity social cohesion intervention (the PA group) and a second control group focusing on nutritional practices. Participants take a set of surveys and undergo physical assessments, and in the PA group, they wear accelerometers for short time periods to measure their walking. The PA group forms teams that set physical activity goals and meet periodically to monitor progress. Researchers will assess participants over a two-year period to gauge the effectiveness of the social cohesion intervention and the role of neighborhoods in supporting or obstructing physical activity. GIS is playing an important role in recruitment, participant mapping, field survey preparation and management, and environmental analysis.
Geocoding for Recruitment and Neighborhood Proximity Analysis
The research team defined neighborhood for purposes of this study at two scales--a 400-meter and 800-meter buffer around each participant's residence. …