When I started general surgery residency in 2011, my training program was on probation for violating the 80 hour work week as mandated by the Accreditation Council for the Graduate Medical Education (ACGME). In addition, new regulations were being introduced that year that limited the maximum number of hours an intern and resident could work (16 and 24 hours respectively) and increased the time off in between 24 hour shifts for residents. Our program worked hard to get off probation with adoption of a night float system and strict recording of duty hours. Even our attending surgeons, who trained when 120 hours a week was the norm, prodded us to comply with these new restrictions. This was truly a new era for surgical residency.
My third month of residency I had the opportunity to rotate on one of the most coveted services: a busy private general surgery service run by one surgeon who performed operations that ran the gamut from run-of-the-mill hernias to Whipples (pancreaticoduodenectomies). Towards the end of my rotation, I saw that he had scheduled an Ivor-Lewis esphagectomy, a complex and challenging case, as the last case of the day. I had never seen one so I desperately wanted to scrub in for the case. It had been a long day in the OR and the case didn’t start until close to 6PM- the usual time when interns were supposed to start wrapping things up so they could sign out to the night float team. Regardless, I scrubbed in to the case with my senior resident and right when we were about to start the case, the attending surgeon turned to me and said, “It’s 6PM, shouldn’t you be signing out?” For general surgeons who trained in an earlier era prior to mine, this would have represented heresy! While I won’t reveal whether or not I stayed and violated the duty hours, I will admit that I have not seen another Ivor-Lewis esophagectomy in the 4 years since.
Our probation status was eventually repealed but the new ACGME rules remained. Many surgeons worried that these new rules, and the 80 hour work week, would negatively affect our general surgery training, limiting our exposure to operating and our acquisition of the much needed experience required to operate independently after training. There was also concern for patient safety with the new rules which is what had sparked the initiation of the first set of ACGME regulations in 2003. In 2003 the concern was that overworked and overtired residents would lead to patient errors. The stricter rules in 2011 tried to further mitigate resident fatigue; however there was a flip-side to reducing maximum shift lengths: loss of continuity of care. Because we had to work less hours, there were more frequent “hand-offs” or “sign outs” to other interns. Often you would sign out a patient to an intern or moonlighter who had never met any of the patients. And in an effort to try to get out of the hospital on time, the likelihood of forgetting something, something important, increased which could also lead to patient error. With so many questions and concerns regarding adequate residency training and patient safety it was only a matter of time before someone decided to study it.
Enter the Flexibility in Duty Hour Requirements for Surgical Trainees (FIRST) Trial published Online First this week in NEJM by Karl Billmoria et al. This study is the first of its kind to study the effect of randomizing residency training programs to different duty hour regulations, which many thought could never be done. Eligible general surgery residency programs and affiliated hospitals were enrolled in the FIRST trial and then stratified by tertiles based on a composite measure of death or serious morbidity. Within each strata, programs and their hospital affiliates were then cluster randomized to the “Standard Policy” group, required to follow current ACGME rules, or the “Flexible Policy” group which allowed maximum intern and resident shifts to be extended and time off in between shifts to be reduced. Both groups had to adhere to the ACGME mandated 80 hour work week, number of days off, and frequency of call regardless of group assignment. Residents were not blinded to their program’s assignment.
The trial was conducted as a non-inferiority trial to look at both patient and resident outcomes. Primary patient outcome was 30-day postoperative death or serious morbidity, obtained using the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP®). Secondary outcomes included other measures of death, morbidity and post-operative complications. Primary resident outcomes were resident-reported satisfaction with overall quality of resident education and overall well-being. Secondary resident outcomes included residents’ perceptions and satisfaction regarding the effect of their current duty hour regulations on various aspects of patient care and safety and resident quality of life and personal being. Separate adherence analyses were performed by surveying program directors at the end of the study. No data was collected on call schedules or resident duty hour logs.
Included in the final analysis were 115 programs (58 Standard Policy, 57 Flexible Policy) and 148 hospitals (70 Standard Policy, 78 Flexible Policy). There was no difference in the patient primary outcome of rate of death/serious morbidity between the two groups (9.00% Standard Policy vs 9.06% Flexible Policy, p=0.921). In adjusted analysis, the Flexible Policy group was noninferior to the Standard Policy for all secondary patient outcomes except for postoperative failure-to-rescue, renal failure, and 30-day postoperative pneumonia, all of which had non-significant differences but did not reach noninferiority.
With regards to resident primary outcomes, there was no difference in rates of dissatisfaction with overall education quality (10.7% Standard Policy vs 11% Flexible Policy) or overall well-being (12.1% Standard Policy vs 14.9% Flexible Policy) between the 2 groups. There were some notable differences in resident secondary outcomes. Residents in the Flexible Policy group were significantly less likely to be dissatisfied with continuity of care (OR=0.44 p<0.001) and quality of handoffs/transitions in care (OR=0.69 p=0.011) but were more dissatisfied with time for rest (OR= 1.41 p=0.020). Flexible Policy residents were also less likely to perceive a negative effect of duty hour policies on patient safety, clinical and operative skills acquisition, OR time and learning/teaching activities (OR for all <1.00 and all p<0.001). However, Flexible Policy residents did feel more effects of the duty hours on personal time away from the hospital with greater perception of negative effects on several measures (ORs all<1, all p<0.001). There was no significant difference, though, in perceived effects of duty hours on job satisfaction or morale.
So what do the results of the FIRST trial mean for duty hours for surgical (and non-surgical) residents going forward? In this study, flexible duty hour policies were noninferior to standard duty hour policies with regards to patient safety and overall resident well-being. Should this outcome change current ACGME regulations? The authors of the FIRST trial believe so, stating, “These results merit consideration in future debate and revision of duty hour policies.” Others interpret the results in a different light. In the accompanying editorial, Dr. John Birkmeyer respectfully applauds the authors of the FIRST trial for their “very ambitious, scientifically robust study.” However, he contends that the results prove that “surgeons should stop fighting the ACGME duty hour rules and move on. The FIRST trial effectively debunks concerns that patients will suffer as a result of increased hand-offs and breaks in continuity of care.” He instead argues for improving our health systems to reduce dependence on “overworked resident physicians.” He concludes by saying, “Although few surgical residents would ever acknowledge this publicly, I’m sure many love to hear, “We can take care of this case without you, go home, see your family, and come in fresh tomorrow.””
While this is likely not the end of the debate on resident duty hours, it certainly adds more data to the discussion. The FIRST trial will spark interesting debate going forward, especially as internal medicine residencies undertake their own RCT, the iCOMPARE trial, to similarly study effects of a more flexible duty hour policy.
Watch the NEJM Quick Take Video: A Trial of Flexible Duty Hours in Surgical Training
(Call to action!)