Waking Up to the Problem of Fatigue Among Medical Interns

At 3 a.m., intern Michael Greger, awakened for the fifth time that night, listened as a nurse ticked off a long list of blood test results for one of his patients, then fell back into an exhausted stupor. Later in the morning, when he checked the patient’s chart, Greger was horrified: He had failed to realize that one of the blood tests clearly showed the man was in imminent danger of having a fatal arrhythmia, a heart rhythm disturbance. The patient was rushed to intensive care.

* It was 2 a.m., anesthesiology resident Steven K. Howard had been working for more than 16 hours and was facing another hectic, sleepless night in the operating room. After preparing two syringes containing drugs for his next patient, Howard had a vague feeling something was amiss. He checked the syringes and discovered that one contained the wrong drug, a medication that would have triggered a fatal stroke.

* Plastic-surgery resident Risa S. Moriarity had been working for more than 50 hours without sleep when she started to perform complicated colon surgery. Minutes after the nine-hour procedure began, Moriarity briefly nodded off, instruments in her hand. After her repeated attempts to stay awake failed, a sympathetic senior surgeon sent her home.

Although none of these episodes resulted in harm to a patient, they are emblematic of an issue that long has escaped scrutiny: the risk of errors made by medical residents, collectively the most inexperienced, overworked and sleep-deprived of American doctors. A landmark report issued 18 months ago by the Institute of Medicine estimated that as many as 98,000 hospitalized patients die each year as a result of medical errors, many of them preventable and most of them unreported.

But the report, which urged organized medicine to examine the error reduction programs undertaken by other industries, barely mentioned fatigue as a factor in mistakes made by the nation’s 100,000 interns and residents.

The reason, according to a member of the panel, is simple: The institute is obligated to consider the most rigorous scientific evidence. And although many experts believe fatigue is a significant factor in medical errors--just as it is in mistakes in aviation, aerospace and other industries--there are only a few studies that indicate a causal relationship.

One often-cited report is an anonymous survey published in 1991 in the Journal of the American Medical Assn. Researchers found that 41% of 145 residents cited fatigue as a cause of their most serious mistake; in nearly one-third of these cases, the patient died as a result of the error. Another report, which appeared in the British journal Nature in 1997, found that staying awake for 24 hours impairs cognitive and motor skills to the same degree as having a blood alcohol level of 0.1%--above the legal limit for driving drunk in most states.

The dearth of rigorous research into the relationship between residents’ fatigue and the errors they make may be a reflection of the subject’s enormous sensitivity. For years the medical establishment has defended the residency training system, largely unchanged since it was pioneered a century ago at Johns Hopkins Hospital. Many doctors have insisted that the grueling three to seven years of hands-on specialty training after medical school--which, after all, they survived--teaches neophyte physicians to subordinate their needs for sleep and food to the unpredictable and often consuming demands of patient care. These defenders insist that residents, who provide much of the round-the-clock care in the nation’s teaching hospitals, learn to transcend fatigue and function effectively.

But safety researchers, sleep experts and a growing number of influential physicians contend that this view represents the triumph of self-delusion over science. They point to numerous studies conducted since 1980 in transportation, at NASA and elsewhere that clearly demonstrate that sleep deprivation causes errors. These studies consistently show that fatigue erodes every aspect of performance: judgment, vigilance, mood, motor coordination, cognitive skills, reaction time and even the ability to recognize error.

In medicine, fatigue is inextricably linked to another well-documented and often overlooked hazard: the night shift. Human performance declines measurably between the hours of midnight and 6 a.m. and most sharply between 3 a.m. and 5 a.m., in accordance with innate circadian rhythms. A disproportionate number of accidents occurs at night, such as car crashes. There is no reason to suspect that the same is not true in medicine, scientists say, and a few studies have suggested this to be the case.

“There is no question that fatigue is foundational to the question of medical errors,” said Mark R. Rosekind, former director of the Fatigue Countermeasures Program at NASA’s Ames Research Center in California and a consultant to the National Transportation Safety Board. “There’s a huge amount of science that has established that this is an issue you have to pay attention to, but we’re just now getting the medical profession to begin to appreciate this,” added Rosekind, one of the few scientists who has extensively studied fatigue in aerospace, aviation and medicine.

“Medicine is the only high-hazard industry that has successfully ignored this issue,” agreed Howard, the former anesthesiology resident who is now associate director of the Patient Safety Center of Inquiry at the VA Palo Alto Health Care System in California. Howard, an associate professor of anesthesiology at Stanford University, noted that airline pilots are barred from flying more than eight hours a day, truck drivers are limited to 10 consecutive hours behind the wheel, and rest breaks are mandatory for air traffic controllers. “You can be sure that if medicine were regulated, or if we had a smoking gun, we wouldn’t have these work hours.”

To Bertrand M. Bell, the maverick medical educator who chaired a New York state commission that drafted the nation’s first and only regulations limiting residents’ hours, long shifts are only part of the problem. The bigger issue in Bell’s view is lack of supervision by senior doctors known as attending physicians. In most residency programs, Bell noted, interns are most closely supervised by second-year residents who are as sleep-deprived and overburdened as interns and have only one additional year of experience.

“Mistakes by interns and residents kill more people than medication errors,” declared Bell, a professor of medicine at Albert Einstein College of Medicine in New York. “There are a lot of reasons. As a resident, you’re taught to hide what you don’t know. Then there’s this idea that the only way you’re going to remember how to do it right is to make a mistake and learn from it. And there’s the fact that the lowest person--the intern--is given the most responsibility for the care of patients.”

Increasingly, however, there are signs that public pressure to reduce preventable mistakes is forcing medicine to begin to confront the question of whether resident work hours are an institutionalized source of errors. The American Medical Assn. and the American Academy of Sleep Medicine are sponsoring a conference that will explore the impact of sleep deprivation on residents. Discussions are underway on Capitol Hill to introduce federal legislation modeled after New York’s regulation limiting resident work hours. And the group that accredits the nation’s 7,700 residency training programs, the Accreditation Council for Graduate Medical Education, has begun to crack down on training programs that overwork residents.

“This problem is getting worse, and it’s crossed a threshold that the ACGME is not going to tolerate,” said David C. Leach, the council’s executive director. Increasingly, he said, “residents are working too hard, not to meet the patient’s needs but to meet the system’s needs.

“There’s more work to do, less time to do it and less support staff,” added Leach, who noted that hospital stays have become shorter and patients are much sicker than they were as recently as five years ago. At the same time residents are spending more time on scut work--drawing blood, transporting patients, arranging patients’ housing--to compensate for layoffs by revenue-pinched hospitals.

Some officials of the AMA are advocating reform as well. “It’s totally ridiculous to think that the longer you work, the better doctor you are,” observed AMA trustee J. Edward Hill, a family practice residency director, disputing one of the profession’s time-honored tenets. “We can really see a difference in our ability to care for patients between the early morning, when we’re rested, and at 4:30 in the afternoon after taking care of patients all day.”

*

Collective denial, a reverence for tradition and the absence of regulation are not the only reasons medicine has lagged so far behind industries such as aviation in addressing the link between fatigue and errors. Medical mistakes are frequently complex and the consequences less obvious than, say, failure to de-ice an airplane.

Nor has the legal system found a way to hold medicine accountable: Successful malpractice cases alleging that sleep deprivation caused or contributed to serious injury are exceedingly rare; patient safety experts estimate that only 5% of medical errors of any kind are ever reported.

“It’s much harder to pin a problem on the fact that a resident didn’t get enough sleep,” said Boston internist turned medical writer Timothy B. McCall, who wrote a passionate indictment of residents’ work hours and their effect on patient care that appeared in the New England Journal of Medicine in 1988 as he was finishing his residency. “It’s not like when a 747 crashes and the evidence is right there on the runway.”

Unlike pilots, who have an obvious vested interest in error prevention and usually belong to powerful labor unions that aggressively negotiate on their behalf, residents are a captive population afraid to complain--or to admit they are exhausted--because their careers depend on the goodwill of their supervisors, particularly their residency directors. These senior physicians have the power to derail, or even to end, a resident’s career with a bad recommendation.

It’s left to each residency program, most of which are sponsored by hospitals, to set its own work hours and regulations under the aegis of committees of the accreditation council. Some specialties, such as internal medicine, limit work weeks to 80 hours averaged over a period of four weeks. Emergency medicine limits shifts to 12 hours and work weeks to 72 hours. Surgery has no limits on hours or the length of shifts.

The accreditation council’s ultimate power is withdrawal of a program’s accreditation, an action that would cost a hospital millions of dollars. The federal government pays training programs about $100,000 per resident annually, while most trainees earn between $26,000 and $50,000, depending on specialty and experience. So far, the council has never withdrawn accreditation from a residency solely for work hours violations, although the accrediting group said they are widespread. A council survey found that in 1999 nearly one-third of all training programs were guilty of violating work rules. These programs were cited by the council, which gives programs time to correct the violations and then checks to determine whether they are in compliance.

New York, which trains more doctors than any other state, adopted rules in 1989 that limit work weeks for all residents to 80 hours and shifts to a maximum of 24 hours. These regulations followed an avalanche of publicity and a grand jury investigation into the death of Libby Zion, an 18-year-old college freshman who died several hours after being admitted to New York Hospital-Cornell Medical Center, where she was treated by a sleep-deprived intern under the supervision of a second-year resident.

Yet by all indications, New York’s regulations are widely flouted and poorly enforced. A series of raids conducted by state health officials at prestigious teaching hospitals several years ago found numerous violations, including the case of one resident who worked a staggering 136-hour week.

*

No group of doctors is more vehemently opposed to limiting residents’ hours than surgeons, who led the unsuccessful opposition to the New York regulation and were instrumental in killing similar measures in Massachusetts and California. Surgeons argue that long hours actually benefit patients by fostering “continuity of care” that forges a bond between doctors and patients.

Richard Reiling is a Harvard-trained Ohio surgeon and former residency director who takes a dim view of sleep research and its implications for medical errors. “There have never been good studies that show damage or injury to patients results from sleep loss by doctors,” said Reiling, who represents the American College of Surgeons at the AMA. “You can have just as much disaster occur after having had a full night’s sleep.” Surgeons, he said, “are built differently” and learn to become impervious to exhaustion. “That’s part of the selection process in surgery,” added Reiling, who dismisses complaints about fatigue as “whining.”

John L. Cameron, residency director and surgeon-in-chief at Baltimore’s Johns Hopkins Hospital, widely regarded as having one of the nation’s toughest training programs, said he considers an 80-hour work week preferable to the 93-hour average reported last year by his trainees. But Cameron notes that, in surgery, long hours don’t end with residency; he arrives at the hospital every day by 6 a.m. and works at least 12 hours. On Saturdays and Sundays he usually leaves around 1 p.m.

“Our residents spend a long time, longer than they should, but it’s not going to be a 40-hour week,” said Cameron, who adds that he holds 8:30 a.m. meetings with his residents every Sunday morning--dubbed “Sunday school”--where they can air grievances.

Like Reiling, Cameron doubts that fatigue caused by 60-hour shifts or spending every second or third night in the hospital on call, both of which occur at Hopkins, result in errors.

Efforts to convince Cameron and other senior physicians that fatigue causes errors are complicated by the fact that sleep deprivation doesn’t inevitably result in mistakes. “It’s like drunk driving,” said Mark Rosekind. “Not everyone who drives drunk is going to have an accident. And in a way that’s unfortunate because it reinforces the attitude that ‘Hey, I can handle it.’ ”

And the ability to “handle it” is a core value in medicine, which is, in Rosekind’s words, “a ‘Right Stuff’ kind of environment”--a culture shared by astronauts, pilots, Navy Seals and other highly trained elite groups. “Right Stuff” cultures prize exceptionally hard work, toughness, intelligence, self-sufficiency and a refusal to complain. Residents at Johns Hopkins are even called “Osler marines” after William Osler, the legendary physician whose 19th century vision that young doctors would best learn to take care of patients by living on the wards and apprenticing themselves to experienced doctors became the modern residency. “These are smart people, very competitive and driven in a high-stress environment, which requires a certain attitude to get through,” said Rosekind, whose wife is a Hopkins-trained pediatrician.

“The problem in medicine is that when we talk about the issue of fatigue, it’s immediately perceived as a weakness,” he added. “But sleep is a vital physiological function, like eating and breathing.”

Virgil D. Wooten, a physician, pilot and a special medical consultant to the Federal Aviation Administration, agreed. “A high level of training and experience does prevent mistakes--up to a point,” said Wooten, medical director of a sleep disorders clinic in Cincinnati. But he noted that nothing can overcome the inevitable deterioration caused by sleep loss. “And there is no way to make the human being perform normally at night when people are programmed to be asleep.”

Leach, the accrediting council’s director, insists that the metamorphosis of the health care system makes it imperative that organized medicine confront the issue of fatigued residents and medical errors.

“Residents live in the cracks of a broken health care system,” he observed. “They get things done. But what we badly need is a system that has many more elements of safety built into it.”