Contribute Try STAT+ Today

Somewhere around the eighth hour of my 28-hour shift in the hospital, a nurse told me that a patient had just been transferred from the intensive care unit to my floor. I stopped by the patient’s room to introduce myself. It was just after lunchtime, and Ms. S’s room was overflowing with family members who had flown in to be nearby because she was in critical condition; they were now celebrating her improvement.

I chatted with Ms. S, examined her, reviewed her medications, and answered questions that she and her family had. At some point, the conversation turned nonmedical. Standing by the window, I remarked on the view of Boston’s Charles River below. I mentioned to the family that whoever was still in her room the next morning would be treated to a beautiful view of the boat-laden Charles at sunrise.

Ms. S’s mother, “Nana,” was pleased by that idea. She joked that I would likely be at home in bed, hours asleep, by the time the sun rose the next morning.


Not giving it too much thought, I lightheartedly corrected her. I told Ms. S and her family that I’d be working in the hospital all night. I’d seen the sunrise over the Charles on my way into the hospital that morning and would still have several hours left in my shift after tomorrow’s sunrise.

Although I made that statement to assure them that there would be continuity to Ms. S’s care, it didn’t have the desired effect. Nana liked boats and rivers. She did not like the idea of tired doctors taking care of her daughter.


Before I realized that Nana was worried, she politely asked me to explain how long I would be in the hospital and how much of that time I would be awake.

“Twenty-eight hours, ma’am, and likely all of it,” I answered.

Catching on to Nana’s concern, I tried to convince her that it wasn’t that bad. I’d slept well the night before. I would eat meals provided by the hospital. There was even a stash of orange juice, ice cream, and peanut butter I could dip into during the wee hours of the morning if I needed a bit more energy. But that explanation didn’t work. Nana, Ms. S, and her family remained concerned.

Rightfully so.

It isn’t terribly reassuring to know that doctors who might need to make life or death decisions about your health could be doing so after having been awake for so long. Would they be on top of their game at hour 16? What about hour 22? In medicine, the devil can be in the details — what if the doctor was too tired to notice something small that might not actually be that small?

But according to the organization that sets the rules on how long resident physicians like me are allowed to work, the Accreditation Council for Graduate Medical Education (ACGME), this is OK. And just so we are clear, as you read this, thousands of young resident physicians are working in hospitals for shifts lasting up to 28 hours every few days and providing care to thousands of Americans. From emergency departments to intensive care units to operating rooms, they often do this without sleep but with the blessing of the ACGME.

When I tell this to my patients, it comes as a shock to many of them. Some think I am joking. Most are concerned. The data seem to agree with them. If given a choice, many patients would prefer a well-rested physician over one who might be tired from a long shift. In a survey of 1,200 Americans, less than 1 percent of respondents said that doctors should work more than 24 hours, while 90 percent said that shifts should be limited to 16 hours or less.

I would love to think that my medical degree gives me some superhuman sense of stamina, but reality and physiology tell me otherwise. I vividly remember a morning when, after signing out to the oncoming team, I had forgotten where the exit to the hospital was, an exit I’d taken hundreds of times before. Luckily, a colleague saw me and walked me downstairs, buying me a cup of coffee in the process to make sure I didn’t fall asleep on my bicycle ride home — something I had once done when I closed my eyes while waiting for a stoplight to turn green.

That’s far less terrifying than situations in which residents get behind the wheel after a long shift only to end up in car accidents. Some walk away, others die.

But what can you expect? Sleep-deprivation of 24 hours or more is equivalent to a blood-alcohol content of 0.1 percent, the threshold for being legally drunk.

The topic of how long resident physicians can work on a given day remains one of the most controversial topics in modern medicine; the way America trains its doctors is heavily contested. In 2003, the ACGME first implemented policies that limited how long resident physicians could work: no more than 80 hours a week averaged over the course of a month. They did so in an attempt to limit medical errors that stemmed from overworked residents. These policies were revised in 2011, further restricting consecutive hours worked: Interns, the youngest doctors in training, couldn’t work shifts longer than 16 hours.

Despite the large body of scientific evidence demonstrating that tired resident physicians make more medical errors than well-rested ones, some groups have successfully argued that these policies should be relaxed. As of July 1, 2017, the ACGME removed the previous safeguard that limited interns to 16 hour shifts; they can now work shifts of up to 28 hours.

Nana’s scowl aside, 28 hour feels unsafe.

Most people would pause before boarding a plane with a pilot who hadn’t slept for the last 26 hours, or wouldn’t feel comfortable getting into a cab with a driver who had been awake for the same amount of time. Why should they feel any different when it comes to doctors?

Christopher Bennett is a resident physician in emergency medicine at Harvard Medical School and Massachusetts General Hospital.

  • I recall all those long call nights from my training over 30 years ago. I recall one morning, after a shift covering intensive care and cardiac care units, that I sat dropping my car keys and picking them up for some time before I drove home. That was the day of THE perfect storm and I crawled up the stairs and slept through the entire storm. How I drove home, I cannot imagine.

    I am always amazed that in these discussions of the call schedule issues, nobody ever considers the effect of being on call every third night for a year. This chronic sleep deprivation has consequences as well. Anybody who wants to claim your time becomes your enemy. This includes family, friend, patients and their families. I suppose it should be studied more the way we study sleep apnea for chronic consequences.

    In my program, we instituted a night float system that help alleviate the problem significantly. Each resident was scheduled for one month of night float and worked 11 pm to 8 am every week night with weekends off. The on-call residents still had to be available in the hospital, but could reasonably expect to get to sleep on those nights unless there was an unusual glut of new admissions.

  • I think it is worth noting that the author of this article is a future emergency medicine doctor, currently in his training. As such, he has made an active decision regarding his future medical practice, that continuity of care does not appeal to him.

    A “28 hour shift” which the article is describing sounds to me like a typical call obligation on an internal medicine rotation, the type of thing a resident would be asked to perform about five or six times a month.

    I finished my medical training in 2001. As such, I spent my intern year working in those prehistoric days before the 80 hour work week rule came about ( most hospitals, mine included, began implementing the 80 hour work week in July 2002, a year before it was mandated).

    With that in mind, I will offer just a few Perspectives.

    1) The 2003 ACGME rule made noticeable progress toward a more humane resident training environment, and when the 28 hour continuous call maximum was mandated, IN ADDITION to an 80 hour weekly maximum hour cap, It was celebrated as a tremendous victory for the residents. Prior to this, it was not unheard of for residents to remain in the hospital for 36 or more hours continuous with no relief in clinical duties, with weekly hours routinely exceeding 100.

    2) at the time when the “work rule” was implemented, many of my attending doctors predicted that over time, it would produce a new generation of entitled Doctors, who would eventually complain that even this was too much. I don’t know that this has happened, but I worry it may have.

    3) it has been my own experience that working a “regular” schedule – I.e. arriving at the hospital for 7 or 8 in the morning and working until 4 in the afternoon,with the occasional call obligation, where one is asked to sleep in the hospital every fifth or sixth night- is far less disruptive to physiology, as well as an overall sense of well-being, versus working only 45 to 50 hours a week, but performing the work as a series of “swing shifts” where on a Monday you may arrive to the hospital at 7 AM and leave at 4, while on Thursday you are asked to come in at 4 PM and work until 1 AM, or worse, come at 7 PM and work until 4AM !

    The latter scenario, utilized extensively by the specialty of emergency medicine, seems far less healthy and safe to me, and I wonder if it may play a large role in the high burnout rate which the specialty of Emergency Medicine deals with.

  • Isn’t it a matter of supervision. Knowing that there could be a fatigue factor, wouldn’t it prudent to round on the patients every 8, or 4 hours depending on the case. Being an instructor, professor or whatever makes your CV sound better should take this responsibility. I trained in an era where if you were on call everyother day, would mean missing half the pathology. Who said, “life is short, art is long and experience difficult”?

  • This author negleted to mention the actual studies arguing for return back to less work hour restrictions. In fact HIS institution is actually participating in one (iCOMPARE). The one that is published (Bilimoria et Al, NEJM 2016) found no difference in surgical patient outcomes when using flexible, ie longer, hours than the more restrictive ACGME guidelines. iCOMPARE is the internal medicine equivalent and should be reported soon.

    These studies were RANDOMIZED TRIALS that specifically looked at patient outcomes. The data that many quote about an all nighter being equiviant to a BAC of 0.1 is pulled from reaction time testing of residents pre and post shift, NOT outcomes data. Additionally no resident is alone up until hour 28, usually new teams arrive by hour 24 and spend the next 3-4 hrs transitioning care via rounding.

    The system knows residents are not 100% by hour 28 and there are multiple levels of redundancy built in such as supervising (back up) residents, in house intensive care fellows, etc. The reason no benefit has likely been shown by decreasing work hours is that an entirely new and even more dangerous variable is introduced with more frequent shift changes: HANDOFFS.

    To this day I’d still rather have my mom cared for by a resident on hour 24 who admitted her, knows her, and has seen her disease course. The other alternative is a resident who just came on, listened to a 5 min one liner and has to look up a first name and med list that they scribbled down on a piece of paper…

    • From the FIRST trial in NEJM:

      In an analysis of data from 138,691 patients, flexible, less-restrictive duty-hour policies were not associated with an increased rate of death or serious complications (9.1% in the flexible-policy group and 9.0% in the standard-policy group, P=0.92; unadjusted odds ratio for the flexible-policy group, 0.96; 92% confidence interval, 0.87 to 1.06; P=0.44; noninferiority criteria satisfied) or of any secondary postoperative outcomes studied. Among 4330 residents, those in programs assigned to flexible policies did not report significantly greater dissatisfaction with overall education quality (11.0% in the flexible-policy group and 10.7% in the standard-policy group, P=0.86) or well-being (14.9% and 12.0%, respectively; P=0.10). Residents under flexible policies were less likely than those under standard policies to perceive negative effects of duty-hour policies on multiple aspects of patient safety, continuity of care, professionalism, and resident education but were more likely to perceive negative effects on personal activities. There were no significant differences between study groups in resident-reported perception of the effect of fatigue on personal or patient safety. Residents in the flexible-policy group were less likely than those in the standard-policy group to report leaving during an operation (7.0% vs. 13.2%, P<0.001) or handing off active patient issues (32.0% vs. 46.3%, P<0.001).

    • True, looking at patient outcomes can be a basis. However, most studies failt to take into consideration the effects of these hours to a resident. if we look only into outcomes, ask yourself, why would a physician put his patient at risk? Physicians help their patients at their own expense, clearly, you have not seen the other side.

    • Great comment!
      thankfully intellectual honesty and backing ones assertions with data are not completely dead practices

    • Hi Ralph thank you for your input. You are correct, these studies do not often look at “the other side,” ie. Work-life balance. That being said I have seen both sides, 1) as a family member of those with terminal illness (lung and brain cancer, as well as ALS) and 2) as a medical professional who spent the last 6 years of his life in residency and fellowship training programs. From that perspective I also incorporate the available data regarding patient outcomes. The author of this article approached the issue from patient safety, I therefore addressed his claims with patient outcomes data.

      As for the physician work-life balance side of things… Residency was one of the most grueling experiences of my life. I remember days without sleep, emotional exhaustion, missed doctors appointments (of my own) and the simple pleasure of managing to get my haircut post call. I also take pride in the fact that my program is known for one of the most intense schedules, and played a major role in designing the above studies. Having now trained at 4 major teaching hospitals, all of which utilizes different training models, I feel confident saying that programs that emphasize continuity of care and patient ownership tend to produce the types of clinicians I’d want to to be caring for my loved ones.

      I chose my training based on my desire to become the best physician possible fully realizing that anything worthwhile takes sacrifice. If anything, I’d argue that larger debt burden and low salaries caused more distress than any amount of sleep deprivation.

  • People who make these rules do not have work longer hours, nights or weekends. They may have worked in the past these inhuman hours but they forgot.
    As for the trained physicians, they work by choice to MAKE MORE MONEY, no one can force them.

  • As big a concern that extended hours are for Residents, there are no rules (that I know of) for Attending Physicians. For example, a surgeon on call for the ER can get called in for multiple patients and end up operating all night after having been up all day. This actually happened to me where a surgeon fixed my broken fibula at 6am the following morning after I hit the ED because other more serious injuries kept coming into the ER. (Summer Saturday – so not entirely unexpected.) Something else to consider…..

  • Keep in mind the other side of the patient safety issue, however, and that is that every time care of patient in hospital is transferred from doctor to another some information is lost with negative implications for quality of care. There needs to be a “sweet spot,” where doctors are with patients for long enough periods to minimise care-damaging hand-offs but not so long as to generate care-damaging hazards from sleep deprivation.

    • Yes, the Ethical Balance begins once we have a choice of medical schools and hospitals (always admit patients to 2 unrelated hospitals, each with due process in all matters administrative).

      See whether the hospitals you are considering at any level in your career meet the “Grassley Ethics Standard”:

      Senator Charles Grassley (R., Iowa) cited a “nonprofit hospital” in Missouri he caught suing indigent patients whose hospital costs are supposed to be covered by the tax-exemption. Contact Dr. Karen Summar in the senator’s Washington office: 202 224 8990.

      See also “Donations of Professional Services Virginia” for a solution to the shortfall-problem of ACA-coverage: Tax-credits at the national and state levels: Voters want as good a plan as Congress itself!


  • This is why I specify in my medical chart, when I go into a hospital that no doctor responsible for any aspect of my treatment have worked more than 12 hours prior to making any decision related to my care and that no surgeon operating on me will have worked more than 8 hours by the time my surgery is scheduled to be complete.

  • Crazy. The doctors have no choice. Are these decisions made by greedy hospital administrators? An important issue for STAT to investigate.

Comments are closed.