Although Medicare pays more than twice as much per enrollee for healthcare in some parts of the country than others, a new study questions whether patients benefit from the higher expenditures.
The study, called the "Dartmouth Atlas of Health Care," was a project of Dartmouth Medical School, the Robert Wood Johnson Foundation and the American Hospital Association. The atlas was released last week at the AHA's annual winter meeting in Washington.
The report shows that price-adjusted Medicare fee-for-service reimbursement to hospitals, physicians and other healthcare providers in 1993 was nearly $6,000 per enrollee in Miami, compared with just $2,700 in Lincoln, Neb.
But the study's authors say there is no indication that beneficiaries in Miami necessarily get better care than those in Lincoln. Instead, Medicare expenditures often are driven by such factors as the number of hospital beds in a region.
One of the study's authors, John Wennberg, M.D., director of Dartmouth Medical School's Center for the Evaluative Clinical Sciences, said the study's findings raise fundamental questions for federal policymakers as they seek to reduce inappropriate healthcare services billed to Medicare and rein in the program's costs.
"Are people who live in high-cost regions any better off than people who live in low-cost regions?" Wennberg asked.
The link between available hospital beds and increased admission of Medicare beneficiaries is especially apparent in what the study calls "high-variation" conditions, or such illnesses as gastroenteritis, pneumonia and congestive heart failure. Physicians do not agree on the need for hospitalization in such cases.
As the number of hospital beds per thousand residents increases, so do discharge rates for such conditions, the study said. In 1992-1993, in regions with fewer than 2.5 beds per thousand residents, the discharge rate for such conditions was 145 per thousand Medicare beneficiaries. In regions with more than 4.5 beds per thousand, the discharge rate was 219.8 per thousand.
Other findings include:
*The rate in 1992-1993 of coronary artery bypass graft surgery per 1,000 Medicare beneficiaries in the fee-for-service sector was four times as high in Joliet, Ill., the highest-rate region, as it was in Grand Junction, Colo., the lowest area.
*Angioplasty rates were eight times as great in Stockton, Calif., the highest-rate area, as they were in the lowest, Buffalo, N.Y. In some areas, higher rates of angioplasty were correlated with lower rates of coronary artery bypass graft surgery, and vice versa, although in most areas, the rates were similarly related.
*The percentage of mastectomies that were "breast-sparing" procedures varied from 1.4% to 48%. Breast-sparing mastectomies were more common in the Northeast and less common in the South, Midwest and Northwest.