When new rules for medical residents were implemented four years ago, the Accreditation Council for Graduate Medical Education limited interns to 16-hour shifts. But at dozens of U.S. hospitals, first-year internal medicine residents are working 30-hour shifts in a study that's taking yet another look at resident work hours.
The 2011 rules were an update to guidelines set in 2003. Each revision has provided more data for accurate assessment of work hours on performance and patient safety. Today's interns -- excluding those in the study -- may work only 16 hours per shift and a total of no more than 80 hours in a week.
From the outside world, where a 40-hour week is considered the norm, this still seems like a lot, especially given that there are only 168 hours in a week. For those physicians who trained before the restrictions went into effect, however, there are significant concerns about the level of experience and education that newer trainees receive.
The original restrictions, put in place in 2003, were intended to combat mental fatigue and trainee errors. Initially, the collected data painted a picture of little or no change, but more recent trends have suggested a subtle decrease in mortality Granted, there are several possible confounders, including increasing use of electronic health records, better access to medical knowledge databases, and improved understanding of disease processes, but the overall trend in mortality for patients cared for by residents appears favorable.
That bodes well for the process, but it’s only one metric. Questions remain about the impact the duty-hour changes have on the knowledge and training of new physicians. United States Medical Licensing Examination scores have remained consistent but there are so many potentially confounding variables that this means little. Many other possible metrics, such as trainee confidence or bedside manner, are also exceedingly difficult to measure accurately.
One of the biggest concerns is the possibility that the restriction of work hours will generate more errors during patient handoffs. For most residents, the greatest chance for forgotten information comes during the transfer of care from one physician to another. We collect reams of data on each patient and order multiple tests that may not be performed until after checkout, making it imperative that we communicate as much pertinent information as possible. Even the most meticulous handoff procedure may miss some details. And more frequent handoffs, which are necessitated by work-hour restrictions, can compound these errors of omission and potentially negate the benefits gained from reducing fatigue. There are several studies looking at handoff procedures and the best way to minimize errors, but so far no one model has prevailed.
The other big concern among duty-hour restriction detractors centers on the decreased exposure to complicated pathology in the early years of training and subsequent deficiencies in education and confidence. As an intern during the initial 80-hour work week rules (2008), I personally saw more pathology on overnight call than during daytime call. I suspect this is because it is more likely that more complex patients will be evaluated and admitted by specialists during daytime hours, but a quick literature search found no specific supporting evidence. Regardless of the reason, there were far more "interesting" cases coming to our service from the ER at 2 a.m. than at 2 p.m.
The system in place at the residency where I trained facilitated the gradual introduction of personal responsibility for admissions. Each intern always had an upper-level resident and an attending faculty member overseeing the admission process. As in all things, some upper-level residents and attendings were better at it than others. The interns took responsibility as we were able, but we were never alone. I understand that this isn’t always the case, and I shudder to think of how many errors I would have made if I were solely responsible for even a simple admission as an intern.
But I also understand that had I not had the opportunity to fail in a controlled setting, I would not be the physician I am today. I would not know the limits of my endurance, or how to push past them to do what must be done. I wouldn’t even know I was capable of doing so. That’s not to say that I feel the restrictions don’t allow this for current trainees, but this experience needs to be incorporated into training, no matter what the duty-hours restrictions call for.
Regardless of your feelings about work hours, or even the profession of medicine as a whole, there is no denying that our job is full of stress. Good stresses and bad stresses, sure, but it's potentially one of the most stressful and demanding callings on both personal and professional levels.
Family physicians literally hold the power of life and death in our hands on a regular basis. We sacrifice time with our families with the hope that the care we provide will allow someone else to spend time with theirs. We spend hours in clinics and hospital wards helping patients learn to help themselves. We help escort new life into the world and, on the same day, hold the hand of the dying as they breathe their last.
In short, we have been given the awesome responsibility of caring for the lives of others, and with that responsibility comes a social contract that we will do so to the best of our ability. That means figuring out the best way to equip doctors in training for as many eventualities as possible, while at the same time preserving their sanity, their health and their compassion.
We must be willing to try new techniques and strategies in the pursuit of that education. Just like the researchers studying work hours at the University of Pennsylvania, we must iterate until we get it right, and be ready for further change when a better way presents itself.
Gerry Tolbert, M.D., is a board-certified family physician who practices in northern Kentucky. A lifelong technophile, his interests include the intersection of medicine and technology. You can follow him on Twitter @DrTolbert.