In the March 2001 issue of Health Care News, we reported “as many as 98,000 Americans are dying each year as a result of medical errors. Former President Clinton and Congress jumped on the issue quickly, putting other issues like medical privacy and Medicare reform on hold. The result was a slew of executive orders, new legislation, increased spending, and growing medical bureaucracies.”
Not surprisingly, the source of that 98,000 figure—an Institute of Medicine study issued late last year—ignited a major debate. In our report, we noted the data used in the study were old and the sample too small to reach accurate conclusions.
Now a new study, published in the July 25, 2001 issue of the Journal of the American Medical Association, supports our conclusion, disputes the IOM study, and calculates the number of medical error-caused deaths at between 5,000 and 15,000 a year.
The new research, led by Rodney A. Hayward of the Veterans Affairs Center for Practice Management and Outcomes Research in Ann Arbor, Michigan—faults the IOM report for too little consensus among the doctors consulted on what constitutes a deadly error.
The JAMA-reported research examined the reliability of reviewer ratings of medical error. Hayward and his colleagues asked what it meant—in terms of the probability of a patient’s immediate and short-term survival if care had been optimal—for a death to be described as “preventable by better care.”
Fourteen board-certified, trained internists reviewed medical records from 1995 and 1996, conducting 383 reviews of 111 hospital deaths at seven Department of Veterans Affairs medical centers. Patients considered terminally ill who received comfort care only were excluded. Reviewers estimated whether the deaths could have been prevented by optimal care (rated on a 5-point scale), and also estimated the probability (from 0% to 100%) that patients would have lived to discharge or for three months or more if care had been optimal.
Like previous studies, the Hayward-led team found nearly one-quarter (22.7 percent) of active-care patient deaths were rated as at least possibly preventable by optimal care; 6.0 percent were rated as probably or definitely preventable. Reliability of the estimates among reviewers was similar to results achieved in previous studies (0.34 for two reviewers).
Reviewers estimated 6.0 percent of patients whose records were studied would have left the hospital alive had optimal care been provided. But after considering three-month prognosis and adjusting for the variability and skewness of reviewers’ ratings, clinicians estimated that only 0.5 percent of patients who died would have lived three months or more in good cognitive health if care had been optimal—representing roughly 1 patient per 10 000 admissions to the hospitals studied.
Hayward and his team conclude,
Medical errors are a major concern regardless of patients’ life expectancies, but our study suggests that previous interpretations of medical error statistics are probably misleading. Our data place the estimates of preventable deaths in context, pointing out the limitations of this means of identifying medical errors and assessing their potential implications for patient outcomes.
Lucian L. Leape of the Harvard School of Public Health, coauthor of the Institute of Medicine report, defended his findings. He says Hayward’s conclusions were based on too small a sample of patients and were derived by way of “statistical torturing.”
For more information . . .
the full text of “Estimating Hospital Deaths Due to Medical Errors: Preventability Is in the Eye of the Reviewer,” written by Rodney A. Hayward and Timothy P. Hofer and published in the July 25, 2001 issue of the Journal of the American Medical Association, is available on the JAMA Web site at http://jama.ama-assn.org/issues/v286n4/rfull/joc02235.html. (You must register at the site.)
For additional information on quality of health care and patient protection, visit the Web site of the National Center for Policy Analysis at http://www.ncpa.org/pi/health/hedex2b.html.