Home Past IssuesAbout UsContact Us Twitter Icon Facebook Logo LinkedIn Logo
Dartmouth Medical School Dartmouth-Hitchcock Medical Center


Geography affects diagnoses, too

By Amos Esty

Medicare enrollees in some regions of the country are more likely to be diagnosed with certain chronic conditions than are similar patients in other regions, according to researchers at the Dartmouth Institute for Health Policy and Clinical Practice (TDI).

Research from Dartmouth has long shown that there are widespread variations in the use of many medical and surgical procedures. The new study, published in the New England Journal of Medicine, reveals that it is not just in treating patients that there are extensive variations but also in diagnosing them.

There seemed to be no survival benefit from the added diagnoses.

Intensity: The researchers, led by Elliott Fisher, M.D., divided the nation's 306 hospital referral regions (each of which comprises an area within which patients are likely to receive subspecialty care) into five quintiles based on the intensity of the medicine practiced there. Then they identified Medicare enrollees who moved sometime between 2001 and 2003 to a referral region in a different quintile to see if moving to a locale providing more-intensive or less-intensive care would affect the likelihood that a patient would be diagnosed with one of nine chronic conditions.

The study showed that Medicare enrollees who moved from a lower- to a higher-intensity region received more medical attention, such as doctors' visits and diagnostic tests, than they would have if they had not moved. The result was an increase in the likelihood of a diagnosis. For every step up in intensity—for example, for a patient who moved from the lowest to the second-lowest quintile—there was a 5.9% increase in the average number of diagnoses. And enrollees who moved to regions in the highest-intensity quintile had, on average, 19% higher risk scores (using a Medicare formula that measures the health of a population) than those who moved to a region in the lowest-intensity quintile. In other words, similar patients appeared to be in worse health in higher-intensity than in lower-intensity regions.

But the relative risk of death one year and three years after the patients' moves was the same, regardless of where a patient moved. So there seemed to be no survival benefit from the added diagnoses.

Implications: The authors concluded that the findings could have implications for attempts to compare the effectiveness of care or to develop new payment models. Such efforts often build in adjustments that take into account the relative health of different populations. That is, if patients in one city are sicker than patients in another city, then it would be expected that patients in the first city would require more medical care. But relying on diagnoses alone may introduce bias if the rate of diagnoses depends on the local medical culture.

As a result of this new finding regarding the impact of geography on care, says Fisher, "it will be important to account for the severity of a condition, not just its presence or absence."

If you'd like to offer feedback about these articles, we'd welcome getting your comments at DartMed@Dartmouth.edu.

These articles may not be reproduced or reposted without permission. To inquire about permission, contact DartMed@Dartmouth.edu.

Back to Table of Contents

Dartmouth Medical SchoolDartmouth-Hitchcock Medical CenterWhite River Junction VAMCNorris Cotton Cancer CenterDartmouth College