Improving the autopsy rate at a university hospital

• Published in: The American journal of medicine Vol. 92; no. 4; p. 423
• Main Authors: Clayton, S A, Sivak, S L
• Format: Journal Article
• Language: English
• Published: United States 01-04-1992
• Subjects: Autopsy - statistics & numerical data; Cause of Death; Death, Sudden; Hospitals, University; Humans; Informed Consent; Internship and Residency; Medical Staff, Hospital; New York - epidemiology; Professional-Family Relations; New York
• ISSN: 0002-9343
• DOI: 10.1016/0002-9343(92)90273-e

Although the autopsy is widely recognized as an invaluable tool for medical education, as well as the gold standard by which the accuracy of diagnoses are measured, there has been a steady decline in the autopsy rate over the last 20 years. At Westchester County Medical Center, a university hospital of New York Medical College, we observed an alarming drop in our autopsy rate. We realized that the methods used to obtain consent from the family of the deceased were haphazard and often left to the junior houseofficer available. We hypothesized that we could increase the autopsy rate by explicitly involving senior housestaff in the task of obtaining autopsy consent after giving them formal instruction in the technique of asking for consent and by having them record information regarding their encounters with families.
Data concerning the frequency of autopsies at Westchester County Medical Center were collected for a 3-month period in 1990. A corresponding period in 1991 was designated the study period during which our intervention was initiated. All medical examiner's cases were excluded for both periods since autopsy consent for these patients is not at the discretion of the family. At the start of the study period and each month thereafter, formal instruction regarding obtaining permission for autopsy was given to all senior residents assigned to direct patient care duty. The senior resident was required to complete a data form regarding autopsy request on each patient who died.
Nine autopsies among 89 deaths (10%) were obtained during the study period in 1990, compared with 31 autopsies among 116 deaths (27%) in 1991 (p less than 0.01). In 1991, autopsies were more likely to be obtained when death was unexpected (p less than 0.05).
The institution of a formal program to educate and involve the senior resident staff in obtaining autopsy consent can significantly improve the autopsy rate at a university hospital.