Study: Hospitals misidentify cause of death up to 23 percent of the time

Autopsies continue to detect clinically important diagnostic discrepancies, according to a new evidence report released today by the Agency for Healthcare Research and Quality. Based on an analysis of more than 50 studies spanning 40 years, researchers estimate that, in U.S. hospitals in the year 2000, the correct cause of death escaped clinical detection in between 8 percent and 23 percent of cases, with as many as 4 percent to 8 percent of all deaths having a diagnostic discrepancy that may have harmed the patient. In addition to clinically missed diagnoses, up to 5 percent of autopsies disclosed clinically unsuspected complications of care.

These diagnostic discrepancy rates do not simply reflect selection by clinicians of diagnostically challenging cases, according to the study authors. In fact, considerable evidence suggests that clinicians have trouble predicting which autopsies are likely to yield important new information. The researchers note that, although often referred to as “diagnostic errors,” these findings refer to discrepancies between clinical diagnoses and autopsy diagnoses and not necessarily to medical mistakes. While diagnostic discrepancies can result from a clinician’s failure to consider an appropriately broad listing of alternative diagnoses or misinterpretation of test results, there are also situations with atypical symptoms or limited diagnostic test information, they explained. These discrepancies, regardless of source, create inaccuracies in death certificates and hospital discharge data, both of which play important roles in epidemiologic research and health care policy decisions, the study authors said.

The evidence report examined the benefits of the autopsy as a tool in health care performance measurement and improvement. However, the researchers did not attempt to address other roles of the autopsy in medical education, furthering medical research, quality control within the medical specialty of pathology, verification of second opinion consultations and legal documentation of findings, or the bereavement process for surviving family members. The focus of the report on the autopsy’s role in detecting quality problems reflected an objectively quantifiable area to evaluate the potential negative effects of the trend toward fewer autopsies during the past 40 years.

In 1994, the last year for which national data exist, the autopsy rate for all non-forensic deaths fell below 6 percent, from a high of 50 percent in the 1960s. This decline is probably due to lack of reimbursement for autopsies, the attitudes of clinicians regarding the utility of autopsies in light of other diagnostic advances, and general unfamiliarity with the autopsy and techniques for requesting one, especially among physicians in medical training, according to the study authors. The evidence report was requested by the College of American Pathologists (www.cap.org).

A summary of the report, prepared for AHRQ by the University of California at San Francisco-Stanford University Evidence-based Practice Center, can be found online at http://www.ahrq.gov/clinic/epcsums/autopsum.htm and also from the National Guideline Clearinghouse? (NGC) at www.guideline.gov (Select NGC Resources). Printed copies of the summary and the full report, “The Autopsy as an Outcome and Performance Measure,” are available by calling AHRQ’s Publications Clearinghouse at 1-800-358-9295 or by E-mailing [email protected].

For more information, please contact AHRQ Public Affairs, (301) 594-1364: Farah Englert (301) 594-6372 ([email protected]) or Karen Migdail (301) 594-6120 ([email protected]).