ABSTRACT
Geographic information systems (GIS) provide powerful methodological tools that have been used to confirm the neighborhood impact on adverse birth outcomes (Li et al. 2010). Neighborhoods that suffer high burdens of birth morbidity (e.g., low birth weight, birth defects) may have more in common than just individual maternal risk factors. Susceptibility to morbidity may also be a function of maternal environment. The literature has strongly advocated the use of GIS in surveillance of the maternal environment and its impact on birth morbidity (Stallones et al. 1992; Reif et al. 1996; Richards et al. 1999; Ali et al. 2002; Dangendorf et al. 2002; Cromley 2003; Croner 2003; Elgethun et al. 2003; Cockings et al. 2004). Spatial analysis of birth morbidity provides a way to identify inadequate health-care access as well as potential environmental and behavioral causal factors. Cluster analysis is frequently used to identify an unusually high occurrence of birth morbidity that is clustered in space and time (Alexander and Cuzick 1992). The most commonly used spatial analytical methods in birth morbidity studies are probability mapping (Waller and Gotway 2004), spatial filtering (Talbot et al. 2000; Rushton 2003), Bayesian smoothing (Johnson 2004), and cluster detection methods such as SaTScan (Forand et al. 2002; Ozdenerol et al. 2005). These methods not only map spatial variability of birth health but also attempt to correct for rate instability and preserve case confidentiality (National Vital Statistics Reports 2007). In probability mapping, the statistical significance of rates is mapped (Waller and Gotway 2004). Spatial filtering generates simulated rates in comparison to the actual rate to reveal significantly high rates (Rushton 2003). The Bayesian smoothing process pulls rates toward a regional or a national rate, making rates more stable (Langford 1994). Cluster detection methods evaluate the overall tendency toward clustering (Forand et al. 2002). Other geographic studies explore neighborhood risk characteristics and evaluate neighborhood effects on birth health (Diez Roux 2001, 2004; Picket and Pearl 2001; Macintyre et al. 2002; Kawachi and Berkman 2003; Oaks 2004; Grady 2006).
