Only a few years ago, it looked as if the U.S. had finally managed to put a leash on surging health-care expenditures, which consume some 14% of gross domestic product. Now, with costs picking up steam again, the search is on for ways to control spending--not only to ease the pain for current employers, workers, and taxpayers, but also to lighten the massive burden that lies ahead when the baby boom generation enters its retirement years.
One potentially useful tack is suggested by a new study by economist Jonathan Skinner of Dartmouth College and physician John E. Wennberg of Dartmouth Medical School that focuses on geographic disparities in Medicare spending. Specifically, the study looks at differences in outlays and care for patients in the last six months of their lives, a period that accounts for some 30% of total Medicare expenditures.
Such differences can be stark. Comparing Miami and Minneapolis in the mid-1990s, for example, the researchers find that Medicare spending per patient in the final months of life, adjusted for illnesses, local prices, age, sex, and other variables, was more than twice as high in Miami. Average days per person spent in intensive care were nearly four times greater. And billed doctors' visits both in and outside of the hospital varied as much as fivefold (chart).
The higher expenditures and more intense treatment provided in a city like Miami might make economic sense, of course, if they paid off in terms of better health outcomes. But the study's analysis of such high-spending areas across the U.S. turns up no evidence that they enjoy reduced mortality rates. And past research indicates that many patients prefer less intense care after they are informed of the risks and benefits of various treatments.
What apparently does affect Medicare spending is the level of local resources. The higher the number of hospital beds or specialists relative to the population, report the researchers, the greater the outlays for people in their final months of life are likely to be. Such spending is also strongly correlated with the presence of for-profit hospitals.
These findings point to the tantalizing possibility that both Medicare and overall health spending could be reduced significantly without harming people's health. Indeed, Skinner estimates that bringing national care levels close to those prevailing in a city like Minneapolis would cut projected Medicare spending by 20%, producing big surpluses well into the next century.
The sticking point is getting health-care providers to change their practice styles without lowering the quality of care. The problems many health maintenance organizations have had trying to effect similar changes suggests that such a major reform would be far from easy.