TOPICS > Health

Critical Choice

May 27, 1999 at 12:00 AM EDT


ELIZABETH FARNSWORTH: The new research, published today in the New England Journal of Medicine, found that heart attack patients stand a better chance of surviving if they’re taken, not just to the closest hospital, but rather to one with lots of experience with heart attacks. The lead author of the report joins us now. He is Dr. David Thiemann, a cardiologist at Johns Hopkins hospital.

ELIZABETH FARNSWORTH: Dr. Thiemann, briefly tell us who you were studying and what you found.

DR. DAVID THIEMANN, Johns Hopkins Hospital: We were studying elderly patients who had the confirmed diagnosis of heart attack. Medicare, the Health Care Financing Administration, collected data in 1994 and 1995 on essentially every elderly patient who had a heart attack during that period and put together a large database that we were able to analyze. We found that patients who had heart attacks and were taken to low-volume centers did much worse than those who were taken directly to high-volume centers, even after adjusting for technology and the doctor’s specialty and every clinical factor involving the patient.

ELIZABETH FARNSWORTH: And what do you mean by low and high-volume?

DR. DAVID THIEMANN: Well, high-volume centers had basically taken care of more than about four Medicare heart attack patients per week or translated into the entire population of patients with heart attack, about one patient per day. The low volume centers had actually less than one patient per week.

ELIZABETH FARNSWORTH: And Dr. Thiemann, what do you mean by elderly?

DR. DAVID THIEMANN: The Medicare database’s patients over 65 only but our findings probably apply to younger patients, as well. If anything, there was greater benefit among the younger patients than among the very elderly. So that although overall there was about a 17 percent difference between patients at the lowest and the highest volume hospitals, in the youngest age group, the 65 to 69-year-old patients, the difference went to as high as 38 percent.

ELIZABETH FARNSWORTH: Before we get into why this might be happening, what’s the significance for our viewers who have either had heart attacks or have friends or relatives who have had them? Should they be sure that if an ambulance comes, for example in San Francisco the ambulance is supposed to take you to the closest hospitals, we checked on it, should they be sure that doesn’t happen, that they go to a hospital that has a higher volume of treatment of heart attack victims?

DR. DAVID THIEMANN: I don’t think this study has implications for individual patients for several reasons. The most important thing for a patient who is having what they think might be a heart attack is simply to call 911 and get into an ambulance, regardless of where the ambulance is going. The other reason is that this is primarily an issue for emergency medical system policy makers and for politicians. It suggests that emergency medical system policy perhaps could be revised. But unfortunately, or fortunately, patients don’t have control over where the ambulance is going to take them.

ELIZABETH FARNSWORTH: How would you revise the emergency policy system?

DR. DAVID THIEMANN: I think there are two answers. One is that for patients who are obviously having a heart attack, emergency medical system people and paramedics may consider taking them directly to a high-volume center, if the transit time isn’t significantly different. For patients who may be having a heart attack but don’t have the obvious diagnosis, the situation gets much murkier, because if the ambulance took everybody who might be having a heart attack to high-volume centers, they’d quickly overwhelm every big hospital in the country.

ELIZABETH FARNSWORTH: And we should be clear here for the record, there was an editorial in the “New England Journal of Medicine” that accompanied your article, that this study, your study’s looking at averages, right? It doesn’t mean that a given hospital with low volume heart attack treatment is not — does not do very well treating heart attacks? Is that right?

DR. DAVID THIEMANN: That’s absolutely correct.

ELIZABETH FARNSWORTH: Is this why individuals shouldn’t try to make their own decision?

DR. DAVID THIEMANN: Yes. There are many excellent low-volume hospitals. Part of the problem, however, is there’s sort of a statistical Catch-22 here in that in hospitals with very low volume, it’s impossible to have an adequate sample size to even tell whether they’re good or bad. And this is one of the fundamental debates in quality assurance, in the debate about whether patients should be taken directly to a bigger hospital, or whether we ought to try to improve the quality at smaller hospitals or try to identify which hospitals are best regardless of volume.

ELIZABETH FARNSWORTH: And is this sort of your research so far a beginning point for more research, I gather?

DR. DAVID THIEMANN: One of the issues raised by this study is how to identify in the field patients who have a high likelihood of having a heart attack. If paramedics using either automatically interpreted electrocardiograms or using a simple questionnaire could identify patients who were likely to be having a heart attack and take only those patients to the higher volume centers or only the very sick patients to the higher volume centers, we might be able to maximize the benefit without overwhelming the system.

ELIZABETH FARNSWORTH: Dr. Thiemann, why do you think the high-volume hospitals do so much better? Is it just that practice makes perfect?

DR. DAVID THIEMANN: It looks as if part of the difference is due to individual practice characteristics, particularly the use of aspirin and beta blockers, which explain about a third of the difference.

ELIZABETH FARNSWORTH: Excuse me. In other words, the way that use those makes some difference?



DR. DAVID THIEMANN: But most of the difference isn’t explained by any of the individual practice variations that we were able to measure. It suggests that the experience not just of the doctors, but of the entire health care team, from the emergency room through the coronary care unit nurses to the tension in additions in the operating room and the catheterization laboratory, is key, because in this analysis, we included the specialty of the physicians, and even a cardiologist at a low-volume hospital does not have as good an outcome for his patients as cardiologists at high-volume hospitals.

ELIZABETH FARNSWORTH: Do most cities have high-volume hospitals, and what about rural areas?

DR. DAVID THIEMANN: Part of the problem I think is that in the United States there is a lot of economic pressure for new drugs and new devices, but we have not invested large sums of money in the emergency medical system. So in Europe, for example, it’s fairly common to give clot-busting drugs to heart attack patients in rural areas in the field before they ever get to the hospital. That’s very rarely done in this country because we haven’t invested in the training and the infrastructure needed to do so. One of the surprising findings in our study is that simply living in a rural area is an independent risk factor for worse survival after heart attack. And that suggests that this is a particular area that might be targeted by health service interventions.

ELIZABETH FARNSWORTH: Do you recommended that hospitals be developed which are essentially cardiac attack units? They have like a trauma unit. If you now have a severe trauma, you may be taken to a hospital because it has this severe trauma unit. Do you think that should happen with heart attacks too?

DR. DAVID THIEMANN: The trauma analogy is interesting and seductive, but it doesn’t apply to cardiac disease for a couple reasons. One is that there are many more cardiac patients than trauma patients. Another is that the cardiac patients tend to have a lot of other problems. So I think the notion of stand-alone hospitals that do nothing but hearts is probably not destined to survive. However, I think the notion of large, high-volume centers that have a special expertise in cardiac disease in addition to taking care of a range of general patients probably is something that deserves serious consideration.

ELIZABETH FARNSWORTH: And you have touched on this throughout, but I gather you aren’t sure yet whether low-volume hospitals could be brought up to par with the high-volume hospitals if certain things are taught or certain treatments are begun.

DR. DAVID THIEMANN: This is a fundamental debate within quality assurance and improvement efforts in medicine. It’s sort of whether to bring the mountain to Mohammed or Mohammed to the mountain. If experience is what matters, as seems likely from this study, then merely being at a low-volume hospital is a severe disadvantage. And it’s probably easier for a nurse or a doctor to maintain their skills and to improve them in a place where they’re seeing heart attack patients constantly than at a place where they see them only infrequently.

ELIZABETH FARNSWORTH: All right. Well, Dr. Teaman, thank you very much for being with us.