About 195,000 U.S. Patients Die Each Year From Medical Errors, Study Finds

An average of 195,000 patients die annually from preventable hospital errors, according to a HealthGrades study released Tuesday, the Wall Street Journal reports. For the study, researchers examined data from 37 million Medicare beneficiaries in all 50 states over three years (Davies, Wall Street Journal, 7/27). Researchers looked for 16 types of errors identified as important by the Agency for Healthcare Research and Quality and estimated a national death rate based on Medicare beneficiaries whose deaths could be attributed to those types of mistakes (Allen, Boston Globe, 7/27). Data from those beneficiaries were adjusted for age. The majority of patient deaths were attributed to an error called "failure to rescue," or mistakes in diagnosing or treating an illness that occurs after an operation, such as pneumonia (Wall Street Journal, 7/27).

The study is the first to estimate deaths caused by failure to rescue (Boston Globe, 7/27). Some say that coding such errors is "both broad and subjective," which makes it difficult to estimate the number of deaths they cause, according to the Journal. Other leading causes of preventable deaths were bed sores that led to infection; postoperative infection; and postoperative embolism or respiratory failure, the study says (Wall Street Journal, 7/27). The study is not being published in a scientific journal but, rather is being released as a promotion for HealthGrades' rankings services of hospitals and physicians (Boston Globe, 7/27).

If CDC counted medical errors as a cause of death, they would be the sixth leading cause of death -- higher on the list than diabetes, influenza, pneumonia and Alzheimer's disease, the study says. According to the study, the additional cost to treat Medicare beneficiaries who experienced medical errors -- including those who lived and those who died -- is $2.9 billion per year. Many of the proposals to reduce medical errors, such as computerized physician order entry and electronic health records would not prevent the majority of mistakes, according to the study. HealthGrades' findings suggest the number of preventable deaths is double estimates from a 2000 Institute of Medicine study. That study was "criticized at the time for the shockingly large number of deaths it estimated" -- between 44,000 to 98,000 preventable deaths annually -- and "sparked a frenzy of initiatives to improve patient care, the Journal reports. The HealthGrades study says that medical errors probably have not increased since the IOM report but rather that the IOM's estimate is "too low," according to the Journal (Wall Street Journal, 7/27). The IOM study "stuck to more clear-cut mistakes, such as overdoses and postsurgical infections," which are easier to quantify, according to the Globe (Boston Globe, 7/27). The IOM study examined data from three states.

Some patient safety analysts said the HealthGrades study "confirms their suspicion" that the IOM estimate was too low, the Globe reports. Dr. Lucian Leape of the Harvard School of Public Health and a co-author of the IOM study, said that the IOM estimate was based on a conservative definition of medical errors and that it underestimated the number of preventable deaths. Dr. Kenneth Kizer of the National Quality Forum said HealthGrades' estimates are likely lower than the true total because they do not factor in errors at nursing homes, private doctors' offices and other outpatient settings (Boston Globe, 7/27). He added, "Any way you look at it, medical errors are a frightening problem."

Some health care analysts questioned the HealthGrades study's methodology and findings, partly because the difficulty of measuring medical errors has led to diverse estimates, the Journal reports. Critics questioned whether Medicare records are accurate or detailed enough to determine whether care providers failed to react to fatal illnesses expediently. Samantha Collier, author of the HealthGrades study and vice president of medical affairs at the company, said that while the "failure to rescue" category is broad, the number of fatal medical errors would still be high even if mistakes attributed to "failure to rescue" mistakes were cut in half. She added, "The lack of recognition [of medical errors] ... fosters a culture of denial and complacency" (Wall Street Journal, 7/27). The study is available online. Note: You must have Adobe Acrobat Reader to view the report.