Specialists, take heart. The situation may not be as dire as you think.
According to a study just published in the Journal of the American College of Cardiology, patients treated by specialists may have lower mortality rates and recover more quickly than those treated by generalists.
The report comes at a time when many patients are having trouble getting past their HMO gatekeeper physician to the specialist they want to see. It could provide ammunition to those who claim the gatekeeper idea doesn't work as well as HMOs say it does.
As managed-care plans ration access to specialists, it's likely generalists are shouldering more of the workload. But how efficient are they? Do they do as good a job as specialists? And does limiting access to specialists really save money? The study looked at a specific adverse health event and how different kinds of doctors treat it.
The authors-Ira S. Nash, M.D.; David B. Nash, M.D.; and Valentin Fuster, M.D.-examined how heart attack patients fared under the care of internists, family physicians and cardiologists.
Previous studies have found patients being treated by cardiologists are more likely to undergo a specific clinically necessary procedure. Cardiologists are also more likely to use certain therapies of established effectiveness.
But what the researchers attempted to do is link the outcomes of treatment to the kind of doctor giving it.
They used the database collected by the Pennsylvania Health Care Cost Containment Council, or PHC4, which gathers information on cost and quality from all hospitals in the state (Feb. 17, p. 140).
They looked at all the cases of acute myocardial infarction, or AMI, in Pennsylvania in 1993. Of those, 16,996 were treated by a cardiologist, 12,960 by an internist and 6,971 by a family physician.
The researchers tried to "risk adjust" for the fact that patients have different risks of dying that have nothing to do with who their doctor is. They may be older or sicker or have other complications.
The risk-adjusted mortality rate for patients treated by a cardiologist was 8.6%. The rate for those treated by an internist was 10.8%, and by a family physician, 11.1%.
With the risk ratio for treatment by a cardiologist set at 1.00, a patient is 26% more likely to die in the hands of an internist and 29% more likely to die in the hands of a family physician, the researchers determined.
"That is a statistically significant difference," said study author David Nash, who is director of health policy and clinical outcomes at Thomas Jefferson University Hospital in Philadelphia. He calculated that if all 12,960 patients treated by internists had been treated by cardiologists, 285 fewer deaths would have occurred. Likewise, if all 6,971 of the family physicians' patients had been treated by cardiologists, 174 fewer people would have died.
"I believe this is the beginning of our rethinking the role of the gatekeeper," Nash said. "I predict many medical and surgical subspecialties will be publishing similar kinds of reports."
The study also found that those patients treated by a cardiologist had shorter hospital stays than what was predicted by an adjusted model. Patients treated by internists stayed longer than predicted. Those treated by family physicians were close to what the model predicted.
Length of stay is used here as a proxy for more economical treatment, but it should be noted that it's possible the specialists used more intensive treatments or spent more resources than the generalists during that shorter stay.
Study author Ira Nash, who is associate director of the Cardiovascular Institute at Mount Sinai Medical Center in New York, draws two conclusions from the report. First, it appears that for AMI, patients have better outcomes if their doctor is a cardiologist.
"No. 2, it points out the fact that many organizations like (PHC4) are interested in collecting outcomes data. The medical community needs to heighten its awareness of these efforts and participate in them to make them as valuable as they can be and as accurate," he said.
However, he added the caveat that if you want to draw conclusions about disparity in outcomes by different kinds of physicians, you have to adjust for variations in the patient populations they treat.
"If I'm Dr. A and you're Dr. B, and all the sick people come to me, I'm going to look like I have poorer outcomes unless some adjustment is made for that fact," he said. "That's the Achilles' heel of this whole body of research. Making those adjustments is an imperfect art."
It is, among other things, the reason HCFA stopped releasing hospital mortality data in 1993.
"This is a general question of great interest in the medical literature right now, the whole notion of risk adjustment," he said.