Describing Geographic Access to Physicians in Rural America Using Statistical Applications in GIS

Funder:	Office of Rural Health Policy (ORHP)
Research center:	North Carolina Rural Health Research and Policy Analysis Center Phone: 919.966.5541
Lead researcher:	Rebecca T. Slifkin, PhD
Contact:	Thomas C. Ricketts, PhD, 919.966.5541, tom_ricketts@unc.edu
Project funded:	September 2004
Project completed:	May 2007
Topics:	Health services Physicians

This study is testing the use of geographically weighted regression to assess the influence of distance and travel time on the distribution of physicians in rural America. The Medicare Modernization Act contains financial provisions aimed at changing the distribution of physicians, and called for a revision in the determination of areas eligible for Medicare bonus payment support. There continues to be a need to accurately characterize primary care distribution and measure access to care for rural places. The goal of this work is to improve our measures of access by identifying the extent to which cross border resources can be considered in indices of access.

There are four specific goals for the proposed research:

• To identify more accurately how supply in adjacent areas can be used to adjust ratios or to characterize overall geographic access
• To analyze county boundaries to determine where they are effective proxies for service areas for shortage determination and how to make best use of the Primary Care Service Areas (PCSAs) developed by the Dartmouth and the Medical College of Virginia team in the assessment of geographic access tocare
• To repeat analyses that characterize the distribution of physicians as more adequate using more current data and optional assumptions that capture the range of possible policy options
• To develop summary indices of supply for small areas, regional areas, and guidance for using smoothed supply estimate.

This project has used GWR to "smooth" the effects of the distribution of physicians beyond county boundaries in assessing the effect of primary care supply on mortality rates. This produces a strong regional effect that runs counter to other published research. We have contacted the research team at Johns Hopkins University (Barbara Starfield and Leiyu Shi) who developed the initial analysis of primary care impacts. An article based on this analysis has been submitted to Health Services Research. The work on PCSAs has been delayed due to the complexity of working with the county based data.

Publications

• Urban-Rural Flows of Physicians
  Author(s): Thomas C. Ricketts, Randy Randolph
  Citation: Journal of Rural Health, 23(4), 277–285
  Date: 2007
  Reports findings from a study to determine whether there was a significant flow of physicians from urban to rural areas in recent years when the overall supply of physicians has been considered in balance with needs.