Geospatial Distribution of Prenatally and Postnatally Diagnosed Congenital Heart Disease: Implications for Equitable Care from a Fetal Heart Society Research Collaborative Study

Abstract

To characterize patterns in the geospatial distribution of pre- and post-natally diagnosed congenital heart disease (CHD) across six surgical centers. A retrospective , multicenter case series from the Fetal Heart Society identified patients at six centers from 2012 through 2016 with prenatally (PrND) or postnatally (PoND) diagnosed hypoplastic left heart syndrome (HLHS) or d-transposition of the great arteries (TGA). Geospatial analysis for clustering was done by the average nearest neighbor (ANN) tool or optimized hot spot tool, depending on spatial unit and data type. Both point location and county case rate per 10,000 live births were assessed for geographic clustering or dispersion. Of the 453 CHD cases, 26% were PoND (n =117), and 74% were PrND (n=336). PrND cases, in all but one center, displayed significant geographic clustering by ANN. Conversely, PoND cases tended towards geographic dispersion. Dispersion of PoND HLHS occurred in two centers (ANN =1.59, p <0.001; and 1.47, p = 0.016), and PoND TGA occurred in two centers (ANN = 1.22, p < 0.05; and ANN =1.73, p<0.001). Hot spot analysis of all CHD cases (TGA and HLHS combined) revealed clustering near areas of high population density and the tertiary surgical center. Hot spot analysis of county-level case rate, accounting for population density, found variable clustering patterns. Geographic dispersion among postnatally detected CHD highlights the need for a wider reach of prenatal cardiac diagnosis tailored to the specific needs of a community. Geospatial analysis can support centers in improving the equitable delivery of prenatal care.