Tuesday, June 1, 2010

Answer to Question # 4: Why do teaching hospital costs vary to the degree that they do, even though their physicians are salaried and do not charge f

Using the Dartmouth Atlas of Health Care data an article in Time magazine (June 29, 2009) by Michael Grunwald compares the costs as revealed by Medicare spending per patient in the last two years of life in five major large hospital teaching centers, all with salaried physicians. The costs itself are of some importance, but more importantly they reflect the style of medicine practiced at each medical center. From most expensive to least they are: UCLA Medical Center $93,842, Johns Hopkins Hospital $85,729, Massachusetts General Hospital $78,666, Cleveland Clinic Foundation $55,333 and Mayo Clinic $53,432. The reasons for these differences are variable but do not include physician entrepreneurship. Although the medical center was paid for the physician services on a fee-for-service basis, the physicians were paid by salary or in some cases on an hourly basis.

There are many known factors causing these differences and many that are harder to define. Certainly the idea that the physicians at the Mayo and Cleveland Clinics have access to information that is unavailable to UCLA, Mass. General and Johns Hopkins in this computer age is absurd. Thus comparative effectiveness research may be somewhat helpful, but it will not solve the problem of making the more expensive centers more like the less expensive ones. There is little to no difference in the availability of advanced technology, but outcomes are possibly worse in the more expensive centers. The more that is done having no benefit the greater the chance of mishaps. The more expensive hospitals have more beds; the patients are in the hospital more often and have more consultant and sub-specialist visits. Alas more (The American Way) is not better. There are of course other complicating factors; hospitals serving less privately insured patients need to maximize billing to compensate for the fact that government programs do not cover the costs of their activities. Even for salaried physicians there are subtle but real pressures to enhance income.

Each teaching center has its own medical culture which is the result of many forces, both historical and economic. The physicians of the Mayo Clinic have a long tradition of quick informal consultation, creating an environment of collegiality and helpfulness not requiring costly formal consultation. The patients at the Mayo and Cleveland Clinics frequently travel long distances for their care and are thus probably more amenable to a conservative approach and more likely to have private insurance; as time passed these institutions developed a more conservative practice of medicine utilizing less consultation, hospital days and ancillary testing.

The wide variations between the costs of care in these fine large teaching hospitals give pause to the concept of the widely touted proposal of bundled payments. This is because many of the patients in these prestigious hospitals are members of the managed care organizations that have evolved over time. Again, it is the medical culture, the skills the physicians have in history taking, physical diagnosis, interpretation of simple tests, ability to conceptualize cases, understanding probabilities and risk-to-benefit ratios along with the ability to communicate effectively.

This difference in medical culture as a primary cause of differences in cost is well documented by Dr. Atul Gawande in his article, The Cost Conundrum, in the June 1, 2009 New Yorker magazine. He compared two centers in Texas, McAllen and El Paso. McAllen spent twice as much ($15,000 vs. $7504) as El Paso per Medicare enrollee/year. Dr. Gawande found that McAllen’s much higher costs were clearly due to an over-delivery of medical care by doctors, without better results.

Dr. Gawande reported on another community, Grand Junction, Colorado which practices in a fee-for-service setting. It had achieved Medicare’s highest quality of care scores. They provide this excellence as one of the lowest health care cost areas in the country. The secret: the physicians have the courage and spirit of collegiality to meet regularly in small groups to review each other’s charts and discuss how to improve care. This is in marked contrast to that found in most centers of aggregated impersonal computerized review. In addition they implemented a regional electronic medical record system reviewing each other’s data, somewhat akin to my suggestion of a national medical record.

We need a medical culture on the national level that is willing to support and effectively teach each other while regularly reviewing cases. We must provide evidence-based care tailored to each patient’s needs. This idea is similar to my suggestion of an active real time peer review system.