TUESDAY, July 24, 2001 (HealthDayNews) -- A new study says the alarmingly high numbers in a government report on deadly medical errors are overstated.
A high-profile paper from the Institute of Medicine (IOM) in 1999 estimated the number of fatal medical mistakes in this country at between 44,000 and 100,000, more than the number of people who perish in car wrecks, breast cancer or AIDS.
The numbers greatly overstate the true risk to patients, say researchers from the University of Michigan. They say the actual figure probably is between 5,000 and 15,000 fatalities a year.
The researchers say the reason for the overstatement is that the IOM used studies that relied on highly subjective patient charts, a method fraught with potential errors.
"The vast majority of these 'errors' are situations not where the wrong thing was done, but they tend to be instances where people thought that more care should have been given. Very few of these are egregious," says lead author Dr. Rodney A. Hayward. The new study appears in tomorrow's Journal of the American Medical Association (JAMA).
Hayward says the nation's health-care system is far from perfect, and his study is hardly cause for celebration, since the findings suggest the situation may be more difficult to fix than previously thought. But he says the results show the system is not suffering from an epidemic of incompetence.
The IOM report, titled "To Err is Human," sparked hand-wringing among health-care professionals and executive action from former President Bill Clinton, who called for closer scrutiny of deadly medical errors, including mandatory reporting of blunders at the state level.
Hayward and colleague Dr. Timothy Hofer enlisted the help of more than a dozen doctors to review 111 deaths at seven Department of Veterans Affairs (VA) clinics nationwide. The work was funded by a grant from the VA Health Services Research and Development Service.
Hayward and Hofer chose cases likely to involve preventable deaths, and excluded patients considered terminally ill.
As in earlier studies, the reviewers determined that about 22 percent of the deaths were possibly preventable with optimal treatment, and 6 percent were probably or definitely avoidable.
Yet, they estimated that only 6 percent of patients who died would have left the hospital alive had they received optimal care. And only half a percent of those who died, or about one in 10,000 patients admitted to the VA hospitals, would have been in sound mental health more than three months after discharge, a strong sign of how sick they were.
"A lot of these people are at the end of what medicine can do to prolong their lives," says Hayward.
The researchers say using charts to examine deadly medical errors may have three major flaws: patients often are extremely ill to begin with; they have poor short-term prognoses, and differences of medical opinion in the face of few treatment options can make choosing the best treatment tricky.
"Rather than these being certain decisions, often they represent just the opinion of one doctor, and sometimes other doctors didn't agree," Hayward says. "It demonstrates how difficult it is to determine when something is an error. A decision might be questionable, but the optimal treatment may not be known."
Carmela Coyle, senior vice president of the American Hospital Association, says, "The JAMA study provides new information and takes a closer look at estimates of medical error, and certainly more data is welcome."
But Joanne Turnbull, executive director of the National Patient Safety Foundation, says focusing on the numbers "is the wrong way to go."
"Rather, we need to focus our energies on research for solutions and a strong dedication to fixing the system. In addition, prevention of patient injury, through early and appropriate response to evident and potential problems, is the key to patient safety," she says.
What To Do
For a look at the Institute of Medicine report on medical errors, check the National Academy of Sciences.
The National Patient Safety Foundation has information on the state of health care in America.
