Hospitals Could Dramatically Cut Mistakes

One out of every 25 hospital patients suffers complications related not to illness, but to treatment. And more than any other single cause, that treatment involves drugs.

A study by a Harvard School of Public Health adjunct professor highlights a low-cost way to catch two of every three drug errors before they occur.

The study, by Lucian Leape, adjunct professor of health policy in the Faculty of Public Health, examined the effect of including pharmacists on physician rounds where treatment is discussed and drugs prescribed.

The study showed the impact can be dramatic, with a 66 percent reduction in adverse drug reactions caused by prescribing errors during the nine-month study period. In addition, doctors were receptive to the pharmacists’ recommendations, accepting 99 percent of the 366 suggested revisions to drug orders during the study, which took place between October 1994 and July 1995 in Massachusetts General Hospital’s Intensive Care Unit.

“Pharmacists are the most underutilized resource at the hospital,” Leape said. “Doctors spend 5 percent of their time thinking about drugs; pharmacists spend 100 percent of their time thinking about drugs.”

The study, published in July 1999 in the Journal of the American Medical Association, follows several previous studies by Leape that examined ways to reduce medical mistakes.

Leape’s work drew nationwide media attention last fall, when a study he co-authored for the Institute of Medicine showed that between 44,000 and 98,000 hospital patients die each year because of medical errors.

An earlier study by Leape at Brigham and Women’s Hospital showed how a computerized drug ordering system – which can compare patient records to check for allergies, drug interactions, and other potential problems – can reduce errors.

While Leape said the computerized system works quite well, many hospitals are years away from having one. Adding a pharmacist to rounds is a low-cost alternative, since hospital pharmacists already perform a review function, but generally after the drug has been ordered, Leape said.

A human pharmacist has advantages over a computer system, Leape said, such as the ability to judge whether to administer a drug by mouth or injection, whether there’s another drug that may be better, and whether there’s a lower-cost drug that does the same thing.

“Any hospital can put a pharmacist on the floor tomorrow if it wants,” Leape said. “The driving concept here is that everybody makes errors – frequently – and if you want to reduce errors, you need to redesign systems and not try to redesign people.”

Adding a pharmacist to the team may not work at all hospitals, however. Doctors in some hospitals, Leape said, especially smaller institutions, come in to see patients individually, when their schedule permits. In those cases, Leape said, there are no physician teams that the pharmacist could join during rounds. Still, he said, careful scheduling could give pharmacists input on prescriptions even at those facilities.

Leape’s interest in medical mistakes was sparked in the late 1980s, as he switched to health policy work from a career as a Tufts University professor of surgery. He was working on a Harvard study of thousands of hospitalized patients and was struck by how many developed additional problems stemming from their treatments.

Leape believes medical errors could be cut by 90 percent over the next few decades if the medical establishment focuses on the problem in a cooperative, nonpunitive way. A 25 percent to 30 percent reduction is possible within five years, he said.

“I’m convinced we need a fundamental change in the way hospitals do business,” Leape said. “We need a shift to a culture that emphasizes safety.”