Geospatial variation in caesarean delivery

AIM: The purpose of this study was to evaluate the variation in caesarean delivery rates across counties in Georgia and to determine whether county‐level characteristics were associated with clusters. DESIGN: This was a retrospective, observational study. METHODS: Rates of primary and repeat caesarean by maternal county of residence were calculated for 2008 through 2012. Global Moran's I (Spatial Autocorrelation) was used to identify geographic clustering. Characteristics of high and low‐rate counties were compared using student's t test and chi‐squared test. RESULTS: Spatial analysis of both primary and repeat caesarean rate identified the presence of clusters (Moran's I = 0.375; p < .001). Counties in high‐rate clusters had significantly lower access to midwives, more deliveries paid by Medicaid, higher proportion of births for women belonging to racial/ethnic minority groups and were more likely to be rural.