Medical history is full of false starts and surprise setbacks. In 1628, the British physician William Harvey performed the first blood transfusion, and his patient died soon thereafter. A century and a half later, in 1796, Edward Jenner’s famous smallpox vaccine resulted in the deaths of several patients. In 1964, Dr. James Hardy performed the world’s first heart transplant — which led to his patient surviving for a grand total of only 90 minutes.
Of course, these disappointments didn’t convince medical researchers to give up on blood transfusions, vaccinations, or heart transplants. The persevering physicians of our past worked to solve new problems – and there were plenty of them – as they popped up. Techniques were perfected, safety was improved, and, ultimately, lives were saved.
It would be good to keep this in mind during the current debate over physician work-hour restrictions.
Ever since the 1984 death of Libby Zion as the result of a medical error by a physician at the tail end of a 36-hour shift, some in the medical profession had advocated for shortening the brutal work hours of young doctors in training. In 2003, out of the growing realization that extreme sleep deprivation leads to medical errors that harm patients, the Accreditation Council for Graduate Medical Education, or ACGME — which sets the rules for physicians in training throughout the country — instituted a 24-hour-per-shift and 80-hour-per-week limit on doctors in training. Gone were the 36-hour shifts and 100-hour workweeks of the past, and gone too, the hope went, were the dangerous blunders that came with the territory. This threat wasn’t imaginary — the danger of sleep deprivation is richly documented. One study in 2000 showed that 24 hours without sleep leads to the same hit to memory and reasoning tests as being legally drunk. More specific to medical practice, a 2006 study found that when a physician is sleep-deprived, he or she commits around 300 percent more of the sorts of preventable errors that can lead to a patient’s death.
In the 13 years since the ACGME implemented its rules, critics of the new system have popped up. They point to the fact that overall medical errors have not actually decreased since the more forgiving sleep rules took effect. This is because shorter shifts are a double-edged sword: Yes, they lead to less sleep deprivation, but on the other hand they lead to more medical handoffs, or transfers of patients from one overseeing physician to another. Anyone who has played the game Telephone as a child can understand why handoffs are inherently risky — whenever a patient’s care is transferred from one doctor to another, there’s potential for miscommunication (or plain noncommunication), which can have dire consequences for patients. Experts now believe that the errors resulting from these increased handoffs are canceling out the benefits that came with reduced sleep deprivation.
So the answer is simple, right? Acknowledge that addressing the sleep problem has caused another problem — more handoffs — to pop up, and figure out to solve that problem. Weirdly, some doctors are instead arguing for a return to the old way of doing things. Arguing that work-hour limits have not produced the promised improvements to patient safety, these physicians say we must remove them.
At the moment, this camp is trying to build up scientific evidence for their position. Dr. Karl Bilimoria, a surgeon at Northwestern’s Feinberg School of Medicine, for example, led an enormous national experiment of 4,400 surgical residents at 119 institutions. Known as the FIRST study, the experiment received a controversial exemption from the ACGME to expose half of the physicians in the study to more “flexible” work hours, eliminating some of the rules set in place to fight sleep deprivation. The other half of the physicians were placed in a control group and adhered to the current work-hour limits.
Naturally, the group working longer hours had fewer handoffs, and the group working shorter hours had more. The study aimed to demonstrate that work-hour limitations do not improve patient outcomes, with Dr. Bilimoria stating — while his study was still in progress — that he intended to use the results to push the ACGME to allow for longer work hours. Published last month, the study’s results showed no difference in patient outcomes between the two groups. In other words, they’re in line with what we already know: Giving physicians more sleep likely leads to fewer errors from sleep deprivation, but more from the increased number of handoffs.The study’s authors chose to frame their results not as an opportunity to improve handoffs so as to ultimately improve patient outcomes, but as proof that bad outcomes happen at about the same rate no matter what, so physicians might as well return to working longer hours.
The FIRST study isn’t the only experiment designed to build a case for rolling back work-hour limits. Investigators at the University of Pennsylvania, Johns Hopkins University, and Harvard Medical School are pursuing another hugely expensive study, known as iCompare, toward the same end. As stated on their trial’s website: “The goal of [this study] is to provide evidence to help policy makers evaluate whether the current duty hour standards should be changed.” In an editorial in JAMA, the journal of the American Medical Association, Dr. James A. Arrighi, the director of clinical residencies within the Lifespan hospital system, writes approvingly about a future where “a more liberal approach to specific duty hour requirements … allows trainees to attain a level of competence … in an environment that has no work hour limits.”
In Europe, there’s been no such pining for the bad old days of overworked young physicians. For whatever reason, the transition to more reasonable work hours for physicians there has gone much more smoothly. In 1998, the European Working Time Directive issued a requirement to limit doctors in training to a maximum workweek of 48 hours. Despite this relatively minuscule number of hours and the high number of handoffs that comes with the territory, European medicine has not suffered. There is no serious argument that German, Swedish, or Danish physicians are poorly trained or that their patients are receiving poor care. While acknowledging the huge differences between our medical systems, the proof of principle remains in plain sight: Doctors don’t need to work insane hours to provide excellent care.
So instead of investing considerable time, effort, and money into attempting to prove that work-hour limitations haven’t solved all of the medical system’s problems — no one’s saying they have — researchers should instead be seeking out ways to further improve that system and make handoffs safer. We must identify where and why mistakes happen and address these shortcomings wherever they are found. The idea of instead simply rolling back the clock to an era where physicians would often fall asleep in the middle of performing surgery is dangerously shortsighted.
Just as the first transfusions, vaccinations, and transplants began as disappointments, the first attempt at limiting the dangers of sleep deprivation in physicians did not produce the overwhelming reduction of medical errors some had hoped. But the lesson is the same: Just because new obstacles appear, don’t turn around. Keep pressing forward. There’s no reason to regress to a system that put patients — and young physicians — at risk.
Dr. Farzon A. Nahvi is an emergency medicine physician at NYU Hospital Center and Bellevue Hospital Center in New York City.