My earliest memory is of watching a cowboy movie on an old Zenith television and being ushered upstairs by my Granny with the promise of a new brother by morning. I know that I was exactly two years, two hundred and ninety-five days old because Granny was right.

I remember the old barn decked with dented pots and pans we’d pass on the way to my first swimming lessons. I remember the way my pencils would rest in the bowl of my grade school desk. I remember the names of every homeroom teacher, every pool manager, and teammates I haven’t seen in decades. When my dad asked recently, “What was the restaurant where Cousin’ Dan dropped the whole pizza on the floor?” I answered, “The Tomato Patch,” without hesitation.

One period I do not remember is my medical residency. That three-year span at the end of my twenties is a haze—I recall only the outlines of rotations, snapshots of “on call” rooms and nursing stations. Memories that should be vivid in my mind—my favorite clinic patients; getting engaged; getting married; Granny’s final days—have faded prematurely.

With a decade into life as an academic hospitalist and parent of two young daughters, my memory has made an improbable recovery. So, what happened during residency? In three words: I didn’t sleep.

Having trained during what many of my predecessors view as the relaxed, post-“duty hour” restrictions era, I was supposed to have benefitted from reforms that currently include: scheduled clinical and educational work limited to 80 hours per week; a cap of 28 hours on each extended, “in-house call” shift; and a minimum of 8 hours off between shifts, collectively intended to improve residents’ quality of life and to prevent drowsy doctors from making mistakes.

In its 2020 version of Common Program Requirements,1 the Accreditation Council for Graduate Medical Education (ACGME) notes that, “Residents have a responsibility to return to work rested…[and] are encouraged to prioritize sleep over other discretionary activities.” To put this in perspective, a resident given 8 hours off between shifts who prioritizes the 7 hours of sleep per night recommended by the CDC is left with 1 hour to accomplish the following discretionary activities: commute home; prepare dinner; eat; exercise; maintain relationships; care for children or other family members; shower; eat breakfast; commute to work; breathe.

Medical training is physically demanding by design—specifically, the designs of one cocaine-fueled surgeon practicing at the turn of the 20th century, when doctors believed in dousing head lice with gasoline and pacifying teething babies with opiates and, notwithstanding the introduction of measurable competencies and heightened expectations for scholarly productivity, success in residency is still measured foremost in terms of survival. This exceptionally low bar makes it easy to dismiss calls for humane working conditions as mere bellyaching from an entitled generation.

The harmful effects2 of acute sleep deprivation are serious and wide-ranging: worsened glycemic control, impaired immunity, weight gain, depressed mood, and increased alcohol use. Chronic sleep deprivation significantly increases one’s risk of developing both coronary artery disease and dementia.

Perhaps most unsettling is the link between sleep disturbance and an increased risk3 for suicidal behaviors, given that women residents are twice as likely as age-matched peers to die by suicide.

I tend not to reflect on difficult experiences, and in my earlier years as a clinician educator I viewed the deprivations of residency as integral to the training process. But the gaps in my memory have exposed the absurdity of that rationalization. A mountain of evidence confirms that memory fails without sleep, with one study4 going so far as to demonstrate atrophy of the brain’s memory center among people with chronic sleep-deprivation. One can only hope the hippocampus grows back.

Redesigning a system and changing a culture that have for more than a century taken for granted the superhuman contributions of young doctors is daunting. In terms of financing, residency programs don’t control their own budgets or the allocation of residency positions. Most programs can’t afford to shorten resident hours without offsets through hospital or government spending.

Convincing older doctors to embrace an assumption denied to them—that residents deserve to sleep as well as the rest of us—may be even more challenging, in part, because outcomes of the ACGME reforms have been disappointing.

The rigorously designed iCOMPARE trial,5 comparing residents’ sleep patterns between programs with and without extended overnight shifts, failed to show a meaningful difference in average daily sleep over a 14-day period. And despite clear evidence that sleep deprivation causes significant impairment in both diagnostic and procedural skills—insights that will come as no surprise to college students, multi-shift workers, or new parents—studies have also failed to show reductions in medical errors as a result of duty hour reforms.

Since publication in the New England Journal of Medicine, the iCOMPARE data have stood as an argument against future limits, an argument that ignores the study’s most overlooked finding; namely, that none of the residents in the trial slept adequately. Whether working extended overnight shifts or shifts capped at 16 hours, residents reported at least 1 period of excessive sleepiness on more than half of their days; sleep duration of less than 7 hours on half of their days; and sleep duration of less than 6 hours on a quarter of their days.

In other words, simply eliminating our most physically punishing shifts was not enough to allow residents the minimum sleep their bodies need. Reforms to date remain inadequate. After more than a century of sleepwalking, it’s long past time for us to wake.

I wish to acknowledge Jennifer Corbelli, MD, for her thoughtful review of this manuscript.

(The views and opinions expressed in this article do not necessarily reflect the official policy or position of the Department of Veterans Affairs or any agency of the U.S. government.)


  1. ACGME Common Program Requirements (Residency). ACGME-approved focused revision. ACGME-approved focused revision, February 3, 2020; effective July 1, 2020. Accessed August 15, 2021.
  2. Institute of Medicine (US) Committee on Sleep Medicine and Research. Colten HR, Altevogt BM, eds. Sleep Disorders and Sleep Deprivation: An Unmet Public Health Problem. Washington (DC): National Academies Press (US); 2006. 3, Extent and Health Consequences of Chronic Sleep Loss and Sleep Disorders. Accessed August 15, 2021.
  3. Bernert RA, Kim JS, Iwata NG, et al. Sleep disturbances as an evidence-based suicide risk factor. Curr Psychiatry Rep. 2015 Mar;17(3):554.
  4. Riemann D, Voderholzer U, Spiegelhalder K, et al. Chronic insomnia and MRI-measured hippocampal volumes: A pilot study. Sleep. 2007;30(8):955-958.
  5. Basner M, Asch DA, Dinges DF, et al. iCOMPARE Research Group. Sleep and alertness in a duty-hour flexibility trial in internal medicine. N Engl J Med. 2019 Mar 7;380(10):915-923.



Clinical Practice, Health Policy & Advocacy, Medical Education, Research, SGIM, Wellness

Author Descriptions

Dr. Sgro ( is an academic hospitalist, associate program director, and clinical assistant professor of medicine, VA Pittsburgh Healthcare System and University of Pittsburgh School of Medicine.