THE FEAR OF being seriously hurt by some preventable medical mistake--the wrong dose of medicine, a glitch in a machine--has until now been fueled mainly by gory anecdotes about mistaken amputations and other such horrors. Now an alarming report from the prestigious Institute of Medicine suggests that medical errors kill between 44,000 and 98,000 people a year, more than breast cancer, AIDS or traffic accidents. The report is meant as a call to action, a first step to changing a "culture of medicine" that allows most errors to pass unrecorded and fails to look to the underlying systems that make errors likelier. But arriving at this frightening statistic is only the first and easiest part of a huge task.
No one in health care, obviously, wishes patients to die because of mistakes. Factors keeping institutions from focusing on the high rate of preventable error include fear of malpractice liability (which prevents accurate reporting and data-sharing) and a medical culture that, the report says, "creates an expectation of perfection and attributes errors to carelessness or incompetence."
Previous surges of concern over medical error have failed to make much headway against these systemic barriers. A drive to shorten the brutally long hours of doctors in training, spearheaded by columnist Sidney Zion after his daughter died in circumstances of apparent error, had some limited effect in New York City. In 1986 Congress created a national database of successful malpractice actions against doctors, intended to keep practitioners from skipping from state to state without accountability. But concerns about misuse have sharply limited access to those data.
This report takes the different tack of urging that accidents be seen as "a form of information about a system." It recommends "non-punitive" systems for reporting and analyzing such errors--mandatory for errors resulting in death or serious injury, confidential and voluntary for less serious incidents that could pinpoint lurking risks.
The hurry and stress of the emergency room or intensive care generate avoidable errors. So do long hours or unsafe working conditions for staff. So do the swiftly growing number of new medications, which doctors have been shown to prescribe without, in many cases, checking available information about a patient's other prescriptions. Some suggestions in the report are already being tried, such as a greater Food and Drug Administration role in packaging and drug names. Other steps would require congressional action, such as the creation of a central office to track patient safety.
Errors are, by definition, hard to anticipate. But the sheer scale of this unnecessary loss of life should act as a spur across the system.