By Brad Segal
Amidst a roller-coaster presidential campaign, on November 4th the Accreditation Council for Graduate Medical Education (ACGME) presented a plan to change resident duty hour limits. That the specifics have largely flown under the radar is perhaps unsurprising given the current news cycle. But the understated revision to, “Resident Duty Hours in The Learning and Working Environment” is the latest twist in a relatively contentious issue within medical education (see 2016 NEJM op-ed vs. responses). The proposal is currently undergoing requisite comment period until December 19. This week I’ll briefly lay out the history of duty hours to help explain the significance of ACGME’s proposal, and I will then go through general empirical arguments for and against such a change. My next post will examine how well these argument hold in light of the most recent data available.
Today the physicians’ training experience immediately following medical school is no longer the whir of dangerous sleep deprivation lampooned in the House of God. Amid mounting evidence that resident sleep deprivation caused medical errors, and under threat of federal legislation, in 2003 the ACGME first introduced national guidelines restricting resident work hours to 80 hours per week (averaged over 4 weeks), and capped residents to 30 hours of continuous in-house call. Then in 2009 the Institute of Medicine (IOM) released a 427-page report reviewing scientific evidence on resident work hours, sleep deprivation, and fatigue-related errors. The evidence overwhelmingly suggests that sleep deprivation significantly impairs most aspects of cognition. Hence the IOM ultimately recommended that residents not exceed 16 hours of continuous work before dedicated rest.
The ACGME subsequently modified duty hour guidelines in 2011 and limited first-year residents (‘interns’) to working 16-hour stretches. The reason ACGME’s most recent proposal is curious, though, is that it back-tracks on the 2011 intern duty-hour limits, raising their in-house cap to 28 hours. In response to this proposal a national advocacy group, Public Citizen, claimed it, “would expose residents, their patients and the general public to the risk of serious injury and death.”
The tension between the IOM’s report and ACGME’s 2016 policy changes highlight a clear divide. Parallel to the debate leading up to the 2001 reforms, today the primary areas of disagreement center on the extent to which duty hour restrictions—or lack thereof—affect; (1) patient safety, and; (2) resident training.
Proponents of the new guidelines frequently contend that the current 16-hour cap has ill effects on medical training. Many note that the continuity of caring for acutely sick patients—for instance, from the time they present in the ED to implementing resuscitative measures in the ICU—is educationally invaluable to understand the natural course of severe illness. These sorts of life-threatening clinical scenarios, however, do not often play out over 16 hours, and even if they do, they are unlikely to neatly coincide with the start of an intern’s 6am shift.
The ACGME proposal itself highlights the argument that current restrictions hinder professional development. They claim the 16-hour cap prematurely separates interns from their teams, “delaying maturation of clinical skills, and threatening to create a ‘shift’ mentality in disciplines where overnight availability to patients is essential in delivery of care.” In a written statement, ACGME’s CEO reinforced this motivation for the change, contending that the expanded duty hour limit, “promotes professionalism, empathy, and commitment.”
Despite the cognitive toll from resident fatigue, many proponents of the new guidelines will additionally cite improvements in patient safety as a reason to raise intern duty hour limits. A study which followed over 2,000 interns before and after the 2011 ACGME changes found that interns self-reported more frequent medical errors after introduction of the 16-hour limit. These errors were self-reported and not independently confirmed. Still, another study of neurosurgical services in New York which compared hospitals with and without residency programs reported that the 2011 changes might have resulted in increased rates of complication among certain patient subpopulations. But to caveat this study as well; there was no effect on overall mortality; and in 2011 ACGME also modified other resident policies, limiting causal inference to the duty hour reforms.
Regardless, proponents will often cite ‘handoffs’ as a mechanism which explains the ostensible consequences of shortened duty hours. Interns with 16-hour shifts must transition clinical responsibility for their patients with increased frequency compared to before 2011. If handoffs between interns are not done properly, this discontinuity in patient care might cause an ominous new symptom to be overlooked, or follow-up on an important test result to be forgotten. Such omissions risk subsequent complication or medical error. Though over a decade old, some evidence suggests that the effects on patient safety from duty hour limits are overall mixed; negative consequences from more frequent handoffs might offset—or perhaps outweigh—the benefits of decreased resident fatigue.
These arguments constitute a formative response to the IOM’s 2009 report on resident fatigue. It is understandable how clinical investigators conducted the Flexibility in Duty Hour Requirements for Surgical Trainees (FIRST) Trial with equipoise. The study, just recently published in the New England Journal of Medicine, was a randomized controlled trial among residencies nationwide and which offers evidence on the empirical arguments raised for—and against—the ACGME proposal.
But like a Rorschach test, interpreting the FIRST trial data is not entirely clear-cut; sometimes it runs counter to the arguments raised in this post, and in unexpected place it supports other claims. Given all of this, I feel that how the results inform the ACGME proposal warrants closer attention, and so keep an eye out for my summary and conclusions in two weeks.