Regional Variations in Diagnostic Practices

Year in Research Nominee for 2010

This article examines geographic variation in diagnostic practices by studying Medicare recipients who move to regions with higher or lower diagnostic intensity than the original region. While variations in Medicare spending by region have been thoroughly documented, the drivers behind these differences are less well understood. The authors collected information on diagnostic histories and risk factors of 225,726 Medicare recipients who changed their place of residence in 2001, 2002 or 2003. The study population was stratified into quintiles based on the intensity of Medicare spending by location. Medicare recipients who moved from higher-spending regions to lower-spending regions were compared to recipients who moved from lower-spending regions to higher-spending regions and to recipients who did not move during the study time.

Key Findings:

  • Medicare beneficiaries moving into higher-intensity areas received more diagnostic testing, a higher level of recorded chronic conditions, and higher risk scores than beneficiaries who did not move or who moved into lower-intensity areas.
  • The average number of diagnoses per beneficiary rose as the study population aged, but increased most sharply for individuals moving to higher-intensity regions. For example, beneficiaries in the second quintile who moved to a higher-intensity region had their number of diagnoses increase an average of 85 percent over the duration of the study, while beneficiaries who moved to a lower-intensity region had the number of diagnoses increase by only 53 percent. Beneficiaries who did not move saw their average number of diagnoses increase by 64 percent.
  • There was no difference in mortality rates between beneficiaries who moved into higher-intensity areas, lower-intensity areas, or did not move.

This article provides insight into geographic variation in health care costs and suggests that the act of moving into a higher-intensity region results in higher than average increases in diagnostic testing and treatment. This variation has serious repercussions for health care reforms relying on accurate risk adjustment.