Five years later, medical errors still a leading killer

Patients were supposed to be so much safer by now.

In November 1999, one of the most influential reports in medical history, "To Err is Human," called for a drastic reduction in medical mistakes after finding that accidental overdoses, infections and other care-giver errors had become a leading cause of death. With up to 98,000 US patients dying annually, the authors declared, "it would be irresponsible to expect anything less that a 50 percent reduction in errors over five years."

But five years later, few hospitals can prove they have reduced patient injuries from medical errors much at all, and no one would argue that the 35 million patients hospitalized this year in the United States are dramatically safer. Thousands of hospitals have launched safety campaigns, but experts differ on whether they're having much effect, especially in the face of shocking mistakes, such as the teenager at Duke University who died after receiving a heart and lung transplant of the wrong blood type last year.

"The evidence of improvement is indeed unimpressive. It's hard to get away from that," said Dr. Lucian Leape of the Harvard School of Public Health and a coauthor of "To Err is Human." "The problem of medical errors is immense." Comparing it to airline safety records, he said hospital patients are at least 1,000 times more likely to die from mistakes than airline passengers.

But Leape and 80 patient-safety specialists who gathered last week at the National Academy of Sciences to assess changes over the last five years say they are encouraged that so many doctors, nurses and hospital administrators at least talk about preventing mistakes today. Until a few years ago, there was a "shame-and-blame" culture that made health-care professionals loathe to acknowledge errors at all. And some hospitals are making dramatic progress fighting errors -- at Brigham and Women's Hospital, a computerized drug-ordering system has slashed serious medication errors by 55 percent, in part by eliminating doctor's sloppy handwriting.

"People are out there struggling in 1,000 different ways how to get on top of this patient-safety thing," said James Conway, chief operating officer at the Dana-Farber Cancer Institute, which has made error prevention a top priority since chemotherapy overdoses in 1994 that killed one patient and irreversibly damaged the heart of another. "People are clamoring. . . . They want to be told what to do" to reduce the risk of accidentally hurting patients.

The question of whether patients are safer now than five years ago is impossible to answer objectively, in part because Congress, under pressure from the hospital industry, rejected the Institute of Medicine's recommendation to create a mandatory national reporting system to track errors. Today, there is no national yardstick for assessing safety, though the federal Agency for Healthcare Research and Quality has begun compiling an annual report based on fragments of information such as surgical instruments left inside patients. ("Retained instrument" reports dropped 40 percent from 1994 to 2002.)

As a result, analysts at last week's conference, sponsored by the Commonwealth Fund, a philanthropy that partially underwrote "To Err is Human," don't even agree on whether the error problem is getting worse or better. On the positive side, an extraordinary number of people and institutions are tackling the safety issue -- from the private agency that accredits US hospitals to health insurers who want to reduce patient claims. And heavily publicized tragedies, such as the dehydration death of a baby at Johns Hopkins Children's Center in 2001, have spurred dramatic safety measures at affected hospitals.

On the negative side, however, thousands of hospitals have not taken such widely recommended safety steps as computerizing drug orders to prevent misunderstandings and give staff the latest information on drug interactions and allergies. Meanwhile, the national nursing shortage and hospital financial problems leaves many wards short-staffed, increasing the risk of both mistakes and not promptly noticing a patient's worsening condition.

In addition, patient safety is such a new field that people are still discovering hidden pitfalls of seemingly good ideas. For instance, researchers thought that they could reduce dangerous falls among the hospitalized elderly by having them wear hip pads, but the patients hated them. "Not surprisingly, the elderly don't want to wear hockey pads while they're lying in bed," said Dr. Kaveh Shojania, assistant professor at the University of Ottawa.

Dr. Robert Wachter of the University of California at San Francisco believes that, overall, medicine has made "a little bit of progress" in making patients safer. But he argues that the improvement comes after decades in which mistakes soared as new technologies and treatments allowed doctors to do more for -- and to -- their patients. An intensive-care patient, he points out, is typically on numerous intravenous medications, any one of which is lethal if given incorrectly, while breathing is maintained on a respirator that could bring instant death if it malfunctions. Compare that, he said, to 1950s medicine that was "quite ineffectual but quite safe."

Some authors of "To Err is Human" have tempered their expectations since their report's release, accepting that long-term improvement requires a cultural change in medicine that is hard to regulate and slow to unfold. But they are still disappointed at the decentralized, uneven nature of the improvements.

"Many of us who worked very closely on this project had hoped that we would have made more progress by now," said Janet Corrigan of the Institute of Medicine, study director for the errors report. "We do realize that what we're trying to change is one-seventh of the economy. The health-care sector is tough to turn around."

Scott Allen can be reached by e-mail at allen@globe.com.