Autopsy director discusses falling rates

John Sinard, director of Yale-New Haven Hospital’s autopsy service.
John Sinard, director of Yale-New Haven Hospital’s autopsy service. Photo by Yale School of Medicine.

A recent report from ProPublica revealed that the frequency of autopsies in the United States has declined over the past century. What was once a typical hospital procedure is conducted on only 5 percent of patients today. The News sat down with John Sinard, director of Yale-New Haven Hospital’s autopsy service, to discuss the implications of this trend.

Q. Dr. Sinard, could you give us some background on the issue — how long have autopsy rates been decreasing, and why is this happening?

A. Autopsy rates in the United States in general have been decreasing since around the 1950s. The most commonly cited reason is that autopsies aren’t as necessary as they used to be. With all the new imaging technology, the perception is that we already know what is wrong by the time the patient dies. A number of studies however have shown this is not the case. Typically in 40 to 50 percent of autopsies, something is clarified or found that was not clinically known prior to the patient’s death.

Q. What can an autopsy reveal?

A. Autopsies are not like they are often portrayed on TV. The autopsy is an anatomic examination. It can address anatomic causes of death, such as infectious diseases, and it can determine the extent of disease, such as in the case of cancer. What it can’t do is identify physiological causes of death such as cardiac arrhythmia or a seizure. Also, in hospital autopsies, we don’t do the types of expensive pharmacological work that would be done for forensic autopsies. Occasionally we have families who request an autopsy because they’re concerned a patient has received too much of a drug, either in the hospital or in the nursing home, and an autopsy isn’t going to be able to answer those questions. It’s important not to confuse hospital autopsies with forensic autopsies; forensic autopsy rates in comparison have remained steady.

Q. In your opinion, under what circumstances would an autopsy be beneficial?

A. There are some cases where it’s particularly recommended: if there’s uncertainty as to why the patient is dead, if it’s not clear where and how extensive cancer was at the time of death or if there’s any suspicion that the patient may be suffering from a hereditary disease. Even an autopsy that finds nothing unexpected could be comforting for the family so they know patient was treated correctly.

Q. Do autopsies benefit medical research?

A. In general, the autopsies can be a source of tissue for medical research. Of course, when tissue is released for medical research from an autopsy, it’s always deidentified so researchers don’t know where the tissue came from. In general, over the years autopsies have helped in understanding the course of disease and why some tumors respond to certain therapies.

Q. How often does Yale-New Haven Hospital conduct autopsies?

A. We do approximately 220 a year. For this last year, 13.5 percent of hospital deaths were autopsied, but that’s not just Yale-New Haven Hospital deaths – it also includes patients from other hospitals in the areas or patients who died outside of the hospital.

Q. In addition to changes in medical practice, are there also financial reasons for this decrease?

A. From a financial perspective, families are not typically charged for autopsies. Insurance companies don’t pay for autopsies because they are included in quality assurance payments to hospitals. The hospitals get these payments regardless and so have no financial incentives to do autopsies.

Q. Do you believe autopsies are worth the cost?

A. When one considers the cost of medical care, I think that the autopsy is a great way to assess the care that a patient received and to a certain extent the appropriateness of that care. On a cost-by-cost basis, it’s one the least expensive thing that can be done. It’s a high yield for a low financial cost.

Comments

  • The Anti-Yale

    We were offered an autopsy when my mother died in 1985 and my father in 1992.

    They were dead.

    That’s all I needed to know.

    Anything else was self-indulgent technological hocus-pocus —- and extraneous.

    Paul Keane

    M.A., M.Div., M. Ed.