The relationship between the spatial structure of health systems and sub-regional variations in health

The purpose of this research was to examine the effects of a change in the spatial structure of the medical care system of St. Louis on the health of the local community. A spatial analysis of the relationship between the geographic location of incidents of infant deaths and inadequate prenatal care and the location of medical facilities was used to accomplish this task.;It was hypothesized that (a) cases of infant risk and loss were dispersed in a clustered pattern throughout the study area, controlling for the race and educational level of the mother, (b) the location and dispersion of cases of infant risk and loss changed between 1984 and 1986, with cases of infant risk less likely to change than cases of infant loss, controlling for the race and educational level of the mother, and (c) the distance between cases of infant risk and loss events and the health facilities of interest was expected to change significantly between 1984 and 1986.;The analytic approaches taken were fairly straightforward. A Quadrat Analysis was used to assess the clustering of cases of infant risk and loss. Analysis of variance was used to detect significant differences in distances between cases of infant risk and loss and health facilities of interest.;The results indicate that, in general and only for the study area and period examined; (1) infant risk and loss events were dispersed in a clustered pattern throughout the study area, (2) infant risk events were more likely to be clustered than infant loss events, (3) race was significantly related to geographic access to medical services, (4) Caucasians had better geographic access than African-Americans, (5) after 1987 this was reversed, (6) education as a proxy variable for socioeconomic status was not related to geographic access to medical services, and (7) African-American populations at risk, as indicated by cases of infant risk and loss, were migrating South.;The results not only point to definite relationships within the populations under study but also between populations and the structural configuration of the set of medical care facilities. Finally, this study underscores the utility of the application of geographic information and spatial analytic technologies towards relevant health services research issues.