David Meltzer, Physician and Economist, Discusses the New Hospitalist Movement The hospitalist movement continues to gain ground in the United States, and David Meltzer, MD, PhD, believes that's probably a good thing. He feels certain, however, that research directed toward evaluating the outcomes of hospitalists will hold the answer to their usefulness . David Meltzer, MD, PhD (Photo credit: D. K. Christopher, University of Chicago BSD AV Department) Meltzer, who is an associate professor in the Pritzker School of Medicine, the Department of Economics, and the Harris School of Public Policy Studies at the University of Chicago, has been studying for several years the effectiveness of hospitalists, general internists who spend a substantial part of their time taking care of hospitalized patients rather than ambulatory patients. Long interested in the idea of specialization and its benefits, Meltzer said during a recent interview that he thinks the University of Chicago hospitals offer a fruitful environment in which to study the impact of hospitalists on patient care, and that "the idea that the hospital could be used as a laboratory, that we could systematically collect data on the patients we're caring for, and use it to improve quality of care, is terribly exciting." Meltzer's research on the effectiveness of hospitalists allows him to use his dual training in medicine and economics. He has recently published a comparison of hospitalists with traditional internists (Ann Intern Med. 2002;137:866-874). His study demonstrated that as hospitalists became more experienced, they reduced the cost of caring for the average hospitalized patient by about $780 per stay, the length of the average stay decreased, and their patients had lower 30- and 60-day mortality rates than patients cared for by traditional internists. To learn more about his thoughts on the hospitalist movement, JAMA recently spoke with Meltzer, who is also codirector of the Robert Wood Johnson Clinical Scholars program and the MD/PhD Program in the Social Sciences at the University of Chicago. An edited transcript of the interview follows. JAMA:How did you become interested in the hospitalist movement? Dr Meltzer: From the time I was in medical school, around the era of the Clinton attempt at health care reform, everyone was talking about the idea that physicians are too specialized. As a medical student, I thought it seemed that there is so much to know, maybe it makes sense to be specialized. And as an economist, I was reading the studies that were done on the effect of specialization on cost and outcome, and was concerned that they were seriously flawed by the fact that patients typically choose doctors for some [specific] reason, so patients don't randomly assort to physicians. JAMA:How did you happen to start studying the hospitalist movement? Dr Meltzer: I recognized that there was an opportunity to study it on the general medicine services. Patients were assigned to whichever doctor happened to be on call on any given day, so people with renal failure might get assigned to a nephrologist or they might get assigned to a general internist—it was a random assortment, which gave me the idea that you could use this as a very nice natural experiment. After I finished my residency, and returned to the U of C, a few colleagues and I created a very simple data system infrastructure. We set it up so we would interview patients when they came into the hospital and then interview them again by phone a month after discharge. We measured patient satisfaction, readmission rates, use of the emergency room, functional status—all sorts of nice outcome measures. We replaced the attending physicians on one service who were traditional, general internists—who typically spent about a month a year on the wards and the rest of the time in ambulatory, in the clinic, or doing research—with two general internists who were each spending 6 months each year, each on for 1 month and then off the next. JAMA:It sounds like a great opportunity to create a living experiment. Were there any limitations to the study? Dr Meltzer: We looked at only two people and what happened to them over time compared with a large number of nonhospitalists, so that's one of the important limitations of the study, and why the multicenter study we are doing right now is so important. JAMA:How do you explain the results of the study? Dr Meltzer: All the differences we're finding over time seem to be explained by increases in disease-specific experience—the more you do it, the better you do it. The more pneumonia you see, the better you get at [caring for patients with] pneumonia. You don't necessarily get better at asthma if you take care of pneumonia, but you get better at pneumonia. JAMA:Did you run into any problems when you were doing the study? Dr Meltzer: We had some problems at the end of the second year. The two physicians who had been doing this for 6 months per year for 2 years just got really tired, started burning out, and they decided they didn't want to be hospitalists anymore or wanted to reduce their time on the wards. Six months is a reasonably long time to spend on the wards, and when they weren't on the wards, they were busy in clinic. Whenever they were on the wards and weren't in clinic, their clinic patients may have been frustrated that they couldn't reach them. The two physicians tried to get their patients comfortable with having one cover for the other, but people like having one doctor and not two doctors, and so it was really hard. The other thing that was frustrating for them was that they were spending so much time on the wards and they hadn't really figured out ways to change things. They were just regular attending physicians who were kind of involved in what was going on around the hospital, and they didn't have protected time to deal with individual patients' problems. They'd just see the same problems again and again. JAMA:Did you figure out a way to improve this situation? Dr Meltzer: We went to the hospital [administration] and said, "Look, these two doctors are saving about $600 000 per year in inpatient expenditures, and they're burning out, so why don't you share some of the savings with us and we'll use it to create protected time so that people can do quality improvement research and cost containment, and help deal with all the system's problems that so badly needed to be addressed?" JAMA:It seems that hospitalists may offer a lot of benefit to patients. Are there any negative aspects to having a hospitalist care for a patient? Dr Meltzer: One of the downsides to specialization is that it leads to fragmentation of care. The typical concern with hospitalists is that you don't have physicians caring for their own patients, so that relationship is broken. JAMA:Are you concerned about the lack of continuity? Dr Meltzer: There are different ways to think about this. It's something that hospitalists worry about a lot. And it's not necessarily driven only by the hospitalist movement. It's to some extent being allowed by primary care physicians. It's both a push by the hospitalists to some extent and an opportunity by the primary care physicians, many of whom may now feel it's not worth it to hop in the car and drive to the hospital and drive back to see only one patient. A long time ago, they might have had 10 patients in the hospital, but as more time is spent on disease prevention, and hospitalization rates have fallen, a physician may have many fewer patients in the hospital. JAMA:So it's not worth it financially? Dr Meltzer: It's not worth it financially. The other thing is that some primary care physicians may feel less comfortable in a hospital environment, particularly if the patients who come into the hospital are much sicker. So maybe there's a push from the hospitalist movement, but we have to recognize that there's an openness to it from nonhospitalists, or from other physicians. The other thing I'd say is that this isn't the first time in medicine this issue has happened. Every time you get neurosurgery or you have a cardiac catheterization, you use a specialist, and your primary care physician often is not the attending of record who takes care of you during your hospital stay. So we often have these sorts of exchanges. And when we think about those we say, "Why do I want my neurosurgeon doing my neurosurgery? It's because I think he or she is going to do a much better job doing neurosurgery than my primary care physician, and that's worth the fact that they may not know me as well." And I think that that's a very rational decision. But if you say, "What about treating my heart attack?" Well, then you could say there's some evidence that a cardiologist might do a better job, so maybe you'd say you'd want your cardiologist, but maybe you'd say, "I want a doctor who knows me." So it's kind of ambiguous. Then you could say, "What about a pneumonia?" Well, maybe every general internist should know how to treat a pneumonia equally well, so then you should just have your regular doctor do it, but then maybe not. Let's collect evidence, and find out. And the evidence that we have suggests, at least for this set of general internists, that people who do it more do it better, so there seems to be some advantage. JAMA:Are patients willing to be seen by hospitalists? Dr Meltzer: Yes, and the concerns that many people have [about not seeing their own physician] aren't that large. There are some people who have very large concerns, and really want their own doctor, and I would argue that those should be respected, and that's what's completely wrong about mandatory hospitalist models. JAMA:Is there concern about the future of general internists? Dr Meltzer: The question is, If general internal medicine bifurcates itself into doctors who do inpatient medicine and doctors who do outpatient medicine, will it weaken both? Will general internists be less good in the outpatient setting because they don't have experience in the inpatient setting? Will patients have less of a feeling of connection to them? Will there be less long-term career satisfaction? I don't think we know the answers to those things. I think those are things we need to think about and be wary of. I also think that it's important that hospitalists have some sense about what happens in an ambulatory setting as well. JAMA:What are some ways to improve connectedness between these two groups of physicians? Dr Meltzer: More attention needs to be paid to good communication between [physicians working in] inpatient and outpatient settings. JAMA:Do you think that in the future more patients will be cared for by hospitalists? Dr Meltzer: Yes, I think so. The growth has been incredible. And there are predictions that it [the hospitalist specialty] will grow to be the size of cardiology or larger within a decade. JAMA:Would you agree that it's important to apply the principles of evidence-based medicine to the question of what should a doctor do and how should a hospital function? Dr Meltzer: Right, and if you look at the hospitalist literature, it started out asking, Are hospitalists better than nonhospitalists? That question is going to gradually get less interesting. There's going to be a progression toward asking, "What are the factors of good hospitalists: What antibiotics do you get? How many days do you stay in the hospital? When are you safe to go home? How much support do you need once you get home?" And the hope is that hospitalists, because they're focused on these things, become leaders in doing the research that identifies those factors, and then leaders in making change. Hospitalists are natural opinion leaders. And the clinical focus that they have on the inpatient services gives you wonderful targets for making change, and then seeing that change translated into practice.
JAMA – American Medical Association
Published: Jan 22, 2003