Penn Researchers Use Geospatial Data to Map Stroke Deaths

Stroke researchers from the Perelman School of Medicine at the University of Pennsylvania unveiled a map demonstrating geographic hotspots of increased stroke mortality across the U.S., among a series of stroke studies presented at the American Academy of Neurology’s 66th Annual Meeting in Philadelphia.

The research, published in Neurology, used spatial modeling to identify well-defined areas of the U.S. where stroke-related death rates were disproportionately high (hot spots) or low (cold spots). County death data from the 2010 Multiple Cause of Death data files (National Center for Health Statistics) were combined with 2010 Area Resource File demographic data. Stroke deaths were identified using previously described ICD-10 (I60-69) codes. Local tests of spatial autocorrelation (LISA statistic) were conducted in ArcGIS to identify locations where disproportionately high or low stroke death rates were clustered.

The researchers found that in 284 counties, mostly clustered in the southeastern U.S., there were 103 stroke-related deaths per 100,000 people, compared to the nationwide median of 59.9 in 3133 counties. Cool spots were identified as 152 counties with median stroke death rate of 26.1 per 100,000.

"Geographic analysis can be used to identify clusters of high stroke mortality rates," the researchers concluded. "These techniques may help identify geographic areas for targeted interventions to improve population health. Further study should assess the stability of these findings over time and determine reasons for the clustering."

For reprint and licensing requests for this article, click here.