Autopsy's Fall Imperils Research

When pathologists talk about a lack of bodies, they don’t always mean staff. They’re referring to, and lamenting, the declining autopsy rate. If you died in 1950—during the golden age of autopsy—the chances were 50-50 that your body would be opened to determine the cause of death. The rate dropped to 41% in 1964, 35% in 1972, and 22% in 1975. Today, it is closer toone in 10 and still falling. Researchers are now voicing alarm about the implications this decline poses for the accuracy of estimating disease rates, the allocation of research funds, and the efficacy of scientific investigation in fields like the neurosciences, in which human tissue is needed. The average university hospital has an autopsy rate of 31%, while community hospitals are down to around 5%, according to Rolla Hill, who is a research pathologist at the Health Sciences Center at the State University of New York, Syracuse. Hill and Robert Anderson, chairman of the pathology department at the University of New Mexico School of Medicine, are authors of The Autopsy: Medical Practice and Public Policy (Austin, Texas, Butterworth, 1988). “The autopsy rate is so low now,” says Hill, “that you can hardly measure the decline.” Ironically, the rate is often quite -high among infants, in certain hospitals hitting 50%. but it dwindles to 1% or 2% in people over 70, those whose bodies could provide the clues that researchers need to better understand aging and chronic disease. “We know a lot more about the causes of death in old mice than we do about the causes of death in old people,” says Anderson. Despite this gap in information, it is unclear which questions would be answered by more autopsies, according to Pasquale Cancilla, a pathologist at the UCLA School of Medicine. He gets requests for tissue all the time, and “we try to fill them when the tissue becomes available,” but no one is actually studying whether availability ‘of tissue is slowing research. The great majority. of the information is anecdotal. “It is sort of jungle rules” for researchers who want tissue but are not associated with a neurologic instition says neuropathologist John Hardman, chairman of the pathology department at the John A. Burns School of Medicine at the University of Hawaii. So, informal methods have evolved in which investigators chat up the clinicians to get them to ask the families of the deceased for autopsy permissions. “It is a matter of establishing a network with key people who might be interfacing with the cases,” says Hardman. This approach has been unusu ally successful for Robert Terry, an Alzheimer’s researcher in the Department of Neurosciences and Pathology at the University of California, San Diego, School of Medicine. Indeed, Terry says that there are not enough pathologists at his institution to do the autopsies. He attributes this success to getting patients’ families involved with the Alzheimer’s. Disease and Related Disorders Association. With such participation, “the autopsies flood the labs,” says Terry, who adds that tissue is usually available for research on Huntington’s and Tay-Sachs diseases. He has less confidence, however, about obtaining tissue for research into other central nervous system diseases, noting, for instance, that “there has been a terrible shortage of autopsies” among dystonia victims. Malpractice Fears Other researchers are even more pessimistic about tissue availability. Mike Hart, director of the Division of Neuropathology at the University of Iowa College of Medicine, says that getting tissue for Alzheimer’s research is still a problem for him. Tissue banks are valuable, but freezing destroys tissue structure and thwarts certain immunological, biochemical, and histological studies. Hart says the decline in autopsies has set back research not only on Alzheimer’s, but also on amyotrophic lateral sclerosis, Parkinson’s disease, and multiple sclerosis. “We have seen about four MS autopsies in the 14 years that I have been at the University of Iowa,” he says. Although Terry is proud of his networking success, Hill sees it as a sign of how bad the situation has become. He says most physicians no longer ask for autopsies, much less push for them as they once did, so “if you just wait for an autopsy, you can wait for a long time.” He also says that the rising number of malpractice suits has had a chilling effect on autopsy, because of the widespread fear among physicians that the practice would uncover their mistakes. founded. One study (Martin J. Valaske, The Archives of Pathology and Laboratory Medicine, 108:462-68, June 1984) of insurance companies and hospital administrators reports that autopsies have aided their defenses against malpractice. Another reason for autopsy decline, according to Stephen Geller, director of Pathology and Laboratory Medicine at Cedars-Sinai Medical Center in Los Angeles, is the decline of the family physician. Geller says that the rise of specialists and the growing distance between doctor and patient have made it harder for physicians to ask their patients’ survivors for permission. But there is more to the decline in autopsy rates than physicians who aren’t asking for permissions. For one thing, it’s expensive, each one costing a hospital between $1,000 and $1,500. Furthermore, the value of autopsy as a research tool has, to an extent, been diminished by the proliferation of new high-tech diagnostic imaging techniques. Many clinicians see no need to second-guess these methods, says Robert Anderson. Even so, people who have studied the reliability of the new techniques don’t feel that they justify the demise of autopsy. Anderson looked at diagnoses of 11 diseases over a 50-year period, comparing accuracy of diagnosis before and after new techniques were introduced (The Journal of the American Medical Association, 261:1610-17, March 17, 1989). He says, “it is hard to see an effect.” The techniques have improved patient care, according to Anderson, “but doctors are not robots,” and they still make errors of interpretation. Hart agrees, saying, “we are still missing the same number of major disease processes as we used to.” Areas of research outside of the neurosciences are also affected by the autopsy decline. Jay Ornstein, director of the Autopsy Service at George Washington University, says autopsy “will affect the study of every type of disease.” In gastroenterology, for example, both hepatitis research and studies of vascular insufficiency of the intestinal track are suffering, according to Geller. With cancer research, however, the situation isnot as clear. Werner Kirsten, associate director of the National Cancer Institute’s Frederick Cancer Research. Facility in Frederick, Md., does'nt find himself short of autopsy tissue, since he can obtain excised tumors from the surgical pathology lab, he says, Autopsy material “is less significant for cancer than for cardiovascular or genetic diseases,” says Kirsten. Geller agrees with Kirsten that the cancer researcher may be less affected than the chronic disease investigator, but says this is changing. As diagnosis wiht fine needle apiration, which samples cells and doesn’t remove tumor chunks, becomes more common and patients go directly to radiotherapy or chemotherapy without tumor removal, “cancer researchers are also going to feel the crunch,” says Geller. Hill worries for a different reason: He says that in cooperative chemotherapeutic trials conducted by the National Cancer Institute, the cause-of-death data used to evaluate new drugs are taken from death certificates without the benefit of autopsy, when it is well known that there is a 10% to 25% discrepancy between such data and what autopsy would show. “So if you don’t get autopsies, you don’t know what is going on in those trials,” says Hill. He is further concerned about any epidemiological research that relies on death certificate data when “that data is just terribly wrong.” Use of this flawed data also has national funding implications; because allocation for individual diseases are, in part’ a function of prevalence, and prevalence is determined by death certificate diagnoses. Some people are working on the issue. Hart is on the Autopsy Task Force of the College of American Pathologists, which is encouraging legislation for third-party payment of autopsy and the setting of mandatory autopsy rates. The task force has also has been successful in getting a commitment from the National Academy of Sciences’ Institute of Medicine to appropriate $400,000 to study the autopsy problem. For his part’ Anderson is surveying clinicians to determine what they want from autopsies, in the hope that if pathologists knew what to provide, the clinicians would seek more permissions. In a 1987 report (JAMA, 258:364-9, July 17), the AMA’s Council on Scientific Affairs called for more autopsies on a number of grounds. Yet, what might bring about the greatest improvement in the autopsy rate might be an emerging interest in quality control, according to Cancilla. The Joint Commission on the Accreditation of Health Organizations, for example, has already de- veloped new guidelines for which hospitals will be required to document that they have asked for an autopsy, and “they will have to try to improve the autopsy percentage,” says UCLA’s Cancilla. Still, Hill is not that optimistic. Earlier this year he reviewed articles over the last decade and a half in nine major health policy journals. He says that there was only one article on autopsy and a half-dozen in which the topic should have come up but wasn’t mentioned. He concludes, “It isn’t in the consciousness of policy-makers yet.” Paul McCarthy is a freelance science writer based in Honolulu.

Autopsy's Fall Imperils Research

Interested in reading more?

Become a Member of