Why Geographic Variation in Medicare Spending Matters

They argue that variations in the utilization and spending throughout the country can be influenced by factors such as burden of disease, race, ethnicity, income, insurance status, market dynamics, and local regulation. This report stresses that more research on the driving factors of geographic variation must be done before policy changes and reimbursement cuts are made that could potentially hurt physicians and their patients.

One issue the AHA raises with the Dartmouth Atlas data is the one dimensional approach taken to account for health status by considering only the healthcare expenditures in the last two years (or six months) of life for Medicare beneficiaries. The AHA points out that this analysis does not take into account the different illness intensities and subsequently varying measures that were taken to treat patients before death. Barnato et al conducted a cohort study, "Development and Validation of Hospital 'End-of-Life' Treatment Intensity Measures" Medical Care 2009, which supports the AHA's opinion.

It analyzed the different measures of hospital end-of-life treatment measures in hospitals in Pennsylvania and concluded that there were marked variations in the end-of-life treatment intensities across the state. Authors fear that the Dartmouth Atlas Project's end-of-life analysis of Medicare spending, may not consider end-of-life treatment intensity properly.

Patients that survive due to intensive measures are not considered, and hospitals that use higher intensity end-of-life treatment may unjustly appear wasteful. Along these lines, they argue that a patient's intensity of treatment is something that can, and should be allowed to vary based on the decisions of the patient and their physician.