Deaths from hospital errors double 1999 estimates.

Preventable hospital errors cause 195,000 U.S. deaths a year, twice as many as previously believed, according to a recent nationwide study of Medicare data. The new findings put hospital errors in third place among the leading causes of death in the United States, following heart disease and cancer. (Health Grades, Inc., Patient Safety in American Hospitals (2004) .)

"The equivalent of 390 jumbo jets full of people are dying each year due to preventable, in-hospital medical errors, making this one of the leading killers in the United States," said Samantha Collier; vice president of medical affairs at Health Grades, which conducted the study. The results also show that "there is little evidence that patient safety has improved in the last five years," she said.

A 1999 study by the Institute of Medicine (IOM) estimated that medical errors cause up to 98,000 deaths a year. That figure was extrapolated from data collected in three states during one year. (Linda T. Kohn et al., Institute of Medicine, To Err Is Human: Building a Safer Health System (1999).)

Health Grades--a Denver-based company that rates health-care quality--claims its study is more up-to-date and comprehensive because its researchers looked at Medicare patient records from all 50 states and the District of Columbia over three years.

The American Medical Association (AMA) discounted the Health Grades study, calling it "flawed" and unreliable because it was not subjected to strenuous peer review.

"The AMA strongly supports research that improves patient safety, but a recent report from the for-profit company Health Grades was not peer-reviewed, and the authors themselves recognize the flaws in their methodology, which relies on claims data that have inherent limitations, including not being able to explain complex situations or make cause-and-effect connections," said Donald Palmisano, AMA immediate past president.

The methodology the Health Grades researchers used was applied in a study published in the Journal of the American Medical Association (JAMA) last year that confirmed the IOM study's conclusions. (Chunliu Zhan & Marlene R. Miller, Excess Length of Stay, Charges, and Mortality Attributable to Medical Injuries During Hospitalization, 290 JAMA 1868 (2003).)

Like the JAMA researchers, the Health Grades group examined signs of medical mishaps, called "patient safety indicators" (PSIs), in the hospital records of the study population. PSIs--which include postoperative infection, bedsores, and transfusion reaction--were developed in 2002 by the federal Agency for Healthcare Research and Quality as a way of assessing and improving patient safety. The researchers ignored four obstetrics-related PSIs, which would not be relevant to a study of Medicare patients.

While IOM and JAMA researchers focused only on unambiguous errors like overdoses and in-hospital infections, the Health Grades study included factors subject to some clinical interpretation--for example, whether hospital staff responded quickly enough to signs of infection. The two PSIs that accounted for most of the reported deaths, researchers said, were failure to rescue dying patients and death in low-risk hospital admissions. Neither factor was included in the IOM and JAMA studies.

Plaintiff lawyers can use the new study to battle tort "reform" efforts on both the state and federal level, said Dennis O'Brien, a personal injury lawyer in Towson, Maryland. Lawyers can cite the figures as proof that "the reason [medical malpractice premiums] are up is not that people are getting more money but because of the number of errors that injure or kill patients," he said.

"It's shocking," O'Brien said. "Doctors and hospitals need to clean up their act and look at themselves as the source [of premium increases] instead of trial lawyers."

Collier said more must be done to reduce hospital errors. "If we could focus our efforts on just four key areas--failure to rescue, bedsores, postoperative sepsis, and postoperative pulmonary embolism--and reduce these incidents by just 20 percent, we could save 39,000 people from dying every year."

The Health Grades study is available online at www.healthgrades.com.