Objectives: This paper describes the use of neighborhood mapping as a key element in an ecological study of a community-based urban infant mortality prevention program. We propose the use of neighborhood mapping in evaluation research to more fully examine the local context of community health programs. Mapping can be used to study community change and to describe community assets and structural, epidemiological, and social features of neighborhoods that may influence program implementation and outcomes.
Methods: Data on physical features were collected by community residents during street-by-street neighborhood walkthroughs. Other data sources included program records, Census, birth certificate, and state and city data. Analytic methods included geo-coding, exploratory factor analysis to create spatial density indicators of neighborhood features at the Census block group level, and analysis of associations between neighborhood features and outcomes.
Results: Point and chloropleth maps provide a powerful illustration of neighborhood features (e.g., vacant buildings), client distribution and participation, health outcomes, and change over time. Factor analysis indicated two salient clusters of non-residential land use: (1) legitimate daily usage (liquor stores and other businesses) and (2) non-legitimate daily use (houses of worship and vacant buildings). A composite scale was created to indicate overall risk related to physical neighborhood features.
Conclusions: Neighborhood mapping is a powerful tool that brings participants and residents into the research process. Moreover, it can improve understanding of the role of neighborhood ecology in program implementation and outcomes.