Imaging utilization rates vary by 47% across country

November 30, 2004

Geographic variations as large as 47% for medical imaging have been identified in the first systematic study of its kind, performed by researchers at Thomas Jefferson University in Philadelphia. A comparison of changes in geographic utilization rates from 1997 to 2002 reveals that cardiologists have higher growth rates for Medicare-reimbursed imaging than radiologists or other physicians who perform medical imaging, said Dr. Vijay Rao, radiology director at Jefferson.

Geographic variations as large as 47% for medical imaging have been identified in the first systematic study of its kind, performed by researchers at Thomas Jefferson University in Philadelphia. A comparison of changes in geographic utilization rates from 1997 to 2002 reveals that cardiologists have higher growth rates for Medicare-reimbursed imaging than radiologists or other physicians who perform medical imaging, said Dr. Vijay Rao, radiology director at Jefferson.

"In all of the 10 regions of the country, the utilization growth rate is higher - in fact, it is much higher - among cardiologists compared to radiologists," she said.

Data were drawn from the Medicare Part B fee-for-service statistics for 1997 and 2002 for regions surrounding Atlanta, Boston, Chicago, Dallas, Denver, Kansas City, New York, Philadelphia, San Francisco, and Seattle.

Outpatient imaging utilization rates for Medicare beneficiaries in 2002 ranged from 3959 per thousand in Atlanta to a low of 2694 per thousand in Seattle.

The rates attributed to radiologists varied 39% across the country. They were highest in Philadelphia, where radiologists performed 2511 exams per 1000 population in 2002. They were lowest in San Francisco, where the utilization rate was 1803 per thousand.

In contrast, the utilization rates for imaging procedures performed by cardiologists varied by 91% from 323 to 617 per thousand from region to region, Rao said. For other physicians, the highs and lows were 941 and 568, respectively.

More analysis is needed to understand the reasons for regional variations in the use of diagnostic imaging, she said.

"We don't know why there is such a big variation. It could be access. It could relate to the amount of disease, or it could relate to the population," Rao said.