Op-ed: Restrictions on physician hours may not improve patient safety

A version of this op-ed was published on June 2nd, 2010 in USA Today.

During medical training, I routinely stayed awake for more than 30 hours straight to care for my patients in the hospital. I strove to be the tireless physician who would be with his patients until they went home.  It turns out I may have jeopardized their safety instead.

For many years, the medical field has been rightfully concerned about doctor fatigue. Since 2003, medical interns and residents have been limited to 80-hour work weeks and 30-hour shifts. A year later, as if to put an exclamation point on the industry’s concerns, a New England Journal of Medicine study confirmed that doctors made substantially more medical errors when they worked frequent shifts of 24 hours or more.

The Institute of Medicine, an independent non-profit that provides authoritative advice on national health issues, subsequently recommended further limits in 2008, including capping continuous shifts at 16 hours and mandating naps during extended periods of work.  These controversial changes have yet to be implemented by the Accreditation Council for Graduate Medical Education, which sets the rules.

But if fatigued physicians are more prone to make errors, shouldn’t better-rested doctors make fewer mistakes?

Not necessarily.  A study released last month by the Cincinnati Children’s Hopsital Medical Center found that further restricting doctor work hours in accordance with the IOM recommendations did not enhance sleep or improve patient safety. In fact, researchers found not only was sleep time unaffected, but physicians’ work-life balance actually worsened.

Shortened work shifts can carry potential risks, some of which were also pointed out by the IOM, but without specific recommendations on how to resolve the problem. Shorter worker shifts:

1. Result in more “patient handoffs” between doctors, which introduces a potential source of errors. A study released in March and led by Vineet Arora, assistant professor of medicine at the University of Chicago Medical Center, found that important clinical data, like patient drug information, was not communicated between the outgoing and incoming doctor 60% of the time. That’s alarming when you consider that patients are passed between doctors an average of 15 times during a typical five-day hospitalization. And when resident physicians at Boston’s Massachusetts General Hospital were surveyed in 2008, more than half said flawed handoffs resulted in patient harm, including prolonged hospitalization, disability or death.

As Dr. Arora points out, there are “concerns about either a tired physician who knows the patient or a well-rested physician that may not know the patient.”

2. Can hamper physician education. The field of surgery, for instance, requires manual dexterity and physical endurance that comes from time spent in the operating room. A British Medical Journal analysis found that a maximum 80-hour workweek did not provide surgeon trainees with the necessary technical mastery.

3. Present ethical dilemmas for doctors. What if a hospitalized patient becomes seriously ill at a time when a doctor is forced to leave because his time limit had been reached? A New England Journal of Medicine editorial said that arbitrarily restricting hours “will signify to our trainees that the overriding consideration is the duration of the shift.”

4. Increase costs. Implementing the IOM recommendations is estimated to cost $1.7 billion, as cash-strapped hospitals have to hire more staff to account for the decrease in physician working hours.  That may not be money well-spent for reforms with such questionable benefit.

Nuance is required to prevent the unintended consequences of work restrictions. Rather than forcing doctors to nap or go home at a defined time, flexibility is needed so physicians can temporarily stay during medical emergencies. During a crisis, patients will appreciate being cared for by a doctor who already knows them, rather than someone new.

And lessons can be learned from the aviation industry, which require pilots to maintain a “sterile cockpit,” prohibiting non-essential communication during takeoffs and landings. Similar focus should be required of physicians when patients are discussed during shift changes.

Failing to do so not only adversely affects how physicians are trained but makes doctors slaves to the clock, stifling the professionalism needed to properly treat patients. And that, ironically, may hurt patients more in the long run.

Comments are moderated before they are published. Please read the comment policy.

  • http://www.linkedin.com/in/mehulsheth Mehul Sheth

    At some point decreased hours do not increase the value of time spent in the hospital. At the risk of sounding like an old curmudgeon that says things were better in my day, there are articles showing that further decreasing the hours below 80/week only increase leisure time and not rest. I agree that there should be a balance, but residency is meant to cram 10,000 hours (the time it seems to take to become proficient in an endeavor) into roughly 3 years. Taking into account vacation time this comes out to about 70 hours a week….why I think the 80 hour limit is where we should hold the line.

  • madoc

    I read some of the Cincinnati Childrens Hospital study. You need to be careful on this one. It could be of importance.
    But it was run for 1 month and there was a puzzling finding (a contradiction).

  • http://www.linkedin.com/in/mehulsheth Mehul Sheth

    Agreed that it was not the most stringent of studies, but just thinking back to my time in training (from ’03-’09, ’03-’06 my wife and I were both in residency) I think more time away from residency would have only yielded a little more time helping my knowledge fund. Of course it’s hard to say, maybe if the standard is 50 hrs/week then residents will spend 30 hours a week reading, but I felt pretty comfortable limiting hours to 70-80/week as long as it was firm, again to get to that target of 10,000.

    • twicker

      Dear Mehul,

      As a quick note, the “10,000 hour rule,” while greatly popularized by Gladwell’s Outliers, isn’t a hard-and-fast rule. As it happens, medical diagnosis represents one area where the rule doesn’t seem to apply. In fact, the person whom Gladwell cited as the expert in this field, K. Anders Ericsson, also published a study (with Lehmann, 1996) that suggested that there was little to no gain in expertise in diagnosis beyond the first year of residency (see http://www.eigenauer.com/criticalthinking/expertperformance.pdf ).

      So – it may be that it becomes harder to hit the 10,000 hour mark, but it may equally be that, in this domain, given the amount of experience a newly-minted MD already brings into the situation (through their 3rd and 4th year rounds, through their coursework, and through their years of general life experience of being around humans who end up with medical conditions), reducing the residency hours may not necessarily lead to a real reduction in physician effectiveness. It would, of course, lead to a reduction in the number of truly unusual cases seen as a resident, but most doctors will either go to practices where they don’t see extraordinary cases (unless they refer them), or into fellowships (where the extraordinary cases become very specialty-specific), or into hospitals/medical universities, where they still have colleagues around from which to learn.

      I realize that Gladwell is all about the 10,000 hours as The Answer; while I enjoy his writing, and do think he provides a lot of great food for thought, I’m not sure that his Answers are always the writ-large answers they may seem to be from his prose.

  • random med student

    What about the residents’ health? Who cares about them? I think an option that has 50 hour work weeks compensated at 40,000 a year and overtime for work done in excess of 50 hours would be a solid idea.Oh, well.

    • DO Student Doc

      I managed a construction company in Alaska before going to med school. I can tell you that when you’re trying to get a lot of work done inside a very definite time frame (10,000 patient-contact hours in two years or completing a building project in less than 20 weeks that would take 40 weeks down in the states) you gotta cram in as much as you can. It ceases to be about the health of the individual worker (to some extent) and becomes about accomplishing the objective, completing the project.

      We would occasionally would find construction workers that didn’t want to put in the hours required of them, that weren’t motivated by the goal, that wanted to “have a life” over their summer in Alaska. They were asked to find other employment. I would assume the same to be true of residency programs (“malignant” residencies aside). It’s a few years of very difficult work to accomplish a goal that will carry us through our careers. After that we can worry about “health” and “having a life”, such as it is.

      Just my $0.02 worth ($0.05 CAD) as another med student.

  • madoc

    Yes you are the random medical student.
    You have some good ideas. Be careful though. They may try to cut you down for offering an alternative to virtual savery.

  • madoc

    virtual slavery
    I do know how to spell.
    I just got excited by the student.

  • http://www.linkedin.com/in/mehulsheth Mehul Sheth

    I’m sure that asking residents to work 80 hours instead of 50 sounds like the health of the residents is not being taken into account, but the question is where do we draw the line, where do we find the balance. should we go down to 40, 30, etc. It would be wonderful if residents could get paid more, unfortunately the history is such where house staff will always get paid minimal as there still has to be a supervising doc to pay and that salaries are usually paid by the gov/t.

    The question is how many hours will allow proficiency (to maintain integrity and trust in the eyes of patients). If we cut back to 40/50 hours I’m sure the next thought would be to lengthen the number of years in residency, and I don’t think anyone involved wants that…

    • madoc

      My supervising doc was never there.
      You never called a staff doc after 500 PM. I don’t think minimal pay will always be. When the young docs take action you will see change.

  • Residentdoc

    I believe this shows residents are underpaid by at least 1.7 b dollars

  • aidel

    The recent data that I’ve seen indicates that cutting down on residents’ hours has not made any (positive) difference for patient safety. If you break it down, a lot of those ‘long hours’ spent at the hospital are not actually spent learning/working. The least the resident can do is to be present.

    • madoc


      What recent data are you referring to?
      What would be the key study?

  • http://silvercensus.com/ Jane Marian

    I think it makes more sense to just have smaller hospitals.

  • TNF

    lol….just realized that if we treated one of our lab rats like you guys treat residents, the feds would shut us down for animal cruelty. ;-)
    Sometimes I wonder if I did the right thing going into research instead of medicine. Articles like this help me realize that the grass isn’t always greener on the other side.

  • twicker

    Quick note about this portion:
    “2. Can hamper physician education. The field of surgery, for instance, requires manual dexterity and physical endurance that comes from time spent in the operating room. A British Medical Journal analysis found that a maximum 80-hour workweek did not provide surgeon trainees with the necessary technical mastery.”

    Seems you’re conflating some concepts.

    Physical endurance doesn’t actually come from time spent in surgery; in fact, from the research I’ve seen on physical endurance, it comes best from doing weight training, which then means that your muscles will be strong enough to basically ignore the far-smaller loads they would have to bear while in the OR. Further, the question really isn’t manual dexterity per se; the question is specific knowledge of how to perform certain manual actions. The surgeon in question could train as a painter for hours, developing amazing dexterity in that field, and have very little of it apply to her practice of surgery.

    The real missing element would be the transfer of the tacit knowledge required for surgery or other medical operations — the knowledge you get primarily from being there. You can read about how to do Surgery X on a normal patient, and maybe read a bit about how to perform it on a patient with Complication Y or Z or Y+Z; you only learn about all the small complications that come up in real life, and all the ways of dealing with those complications, by seeing a panoply of patients, each with her or his own set of particularities. And the way to solve that is not to have people regularly work until they’re about to fall over in a faint; it’s to require a long-enough residency/fellowship to put in the required time to see the conditions. That may require re-writing the laws some (I don’t know at what point the loans have to start being repaid), and it may require that we reduce the cost barriers to entry to the medical profession (which I think we, as a society, need to do anyway), but it’s not simple issues of general manual dexterity or of physical stamina.

  • http://www.linkedin.com/in/mehulsheth Mehul Sheth

    No question that residents are underpaid. But physicians are the only profession where you can have a garunteed 6 figure income WITH job security. The recent financial crisis shows that the two rarely go together.

    With this fact in mind, there will surely be folks willing to put up with long hours and little salary to get to that position. The only thing that might change if med school tuition continues to skyrocket, then the cost may outweigh the benefit.

    I would love if med students or residents would come together to get higher wages, but I fear the transient nature of residency, combined with the eventual promise of a solid income will prevent it from happening. None-the-less I challenge the current residents to try as you are underpaid.

Most Popular