GeoSpatial Analysis of Stroke Hospitalizations in Virginia 1994-2008: A Methodological, Analytical and Policy Perspective

Background. Geographic patterns of stroke hospitalization are of concern in Virginia because of the high rates of hospitalization from rural areas that are least accessible to emergency services and are most distant from comprehensive stroke hospital care. The lack of stroke stabilization capabilities in remote rural hospitals and the shortage of neurologists in these areas have been seen as a major causal factor in the relatively high stroke mortality rates in these areas. Objective/purpose. The study analyzes the spatial distribution of Virginia's stroke hospitalizations between 1994-2008 to better understand historically changing patterns of care within rural hospitals and the accessibility of the state's Comprehensive Stroke Centers. Fifteen years of hospital discharge data, which have been augmented with travel time measures for each hospitalization, are used to accurately assess “time to care. Methods. Standard geospatial cluster techniques, including, SatScan, Geoda software and ArcGIS, were used to determine the Moran Index, Kriging, Kernel Density and Hot Spots with Rendering. Network Analyst was used to calculate drive time distances with impedance for each stroke hospitalization using the patients ZIP Code population weighted centroid. Results. The research has demonstrated the viability of using spatial analysis techniques and methodologies to detect significant clusters in longitudinal data and the capability of better understanding the dynamics of changing patterns. A convergence of hospitalization clusters from different stroke types was also noted at all levels of aggregations. Discussion/conclusions. The study has reframed future inquiry, provided guidance for public health interventions, and supported community involvement in new models of care. Public health.