Overview

The FIRST Trial was a national, cluster-randomized, pragmatic noninferiority trial to test whether surgical-patient outcomes under flexible, less-restrictive duty-hour policies would be no worse than outcomes under standard ACGME policies. Secondary outcomes also studied included resident satisfaction, resident perceptions of patient care, resident education, and resident well-being. Of the 136 eligible general surgery residency programs, 117 (87%) programs and 151 affiliated hospitals participated. Programs and their affiliate hospitals were randomized to the control group (standard-policy group) or the intervention group (flexible-policy group). Outcomes were measured using ACS NSQIP data and the American Board of Surgery In-Training Examination (ABSITE) survey.

Summarized Results

Patient Outcomes

Patient outcome analyses were performed on 138,691 general surgery patients. The rate of death or serious complication did not significantly differ between study groups (9.1% in the flexible-policy group and 9.0% in the standard-policy group, P=0.92). In regards to other secondary patient outcomes, the flexible-policy group was found to be noninferior to the standard-policy group for serious complications, any complication, unplanned reoperation, sepsis, surgical-site infection, and urinary tract infection.

Resident Outcomes

Resident outcomes were analyzed for 4,330 general surgery residents. There was no significant difference in resident satisfaction with overall education quality (11.0% in the flexible-policy group and 10.7% in the standard-policy group, P=0.86) or resident satisfaction with overall resident well-being (14.9% vs 12.0%, respectively; P=0.10). Residents in the flexible-policy group compared to the standard-policy group were less likely to be dissatisfied with continuity of care (odds ratio 0.44; 95% CI, 0.32 to 0.60; P < 0.001) and quality and ease of handoffs and transitions in care (odds ratio 0.69; 95% CI, 0.52 to 0.92; P=0.01), and less likely to perceive a negative effect of institutional duty-hour policies on patient safety, continuity of care, clinical-skills acquisition, operative-skills acquisition, autonomy, operative volume, availability for elective and urgent cases, conference attendance, time for teaching medical students, the relationship between interns and residents, and professionalism (all odds ratios>1.00, P < 0.001 for all comparisons except P=0.003 for professionalism). There were no differences in ABSITE or board examination scores between study arms.

In contrast, the flexible-policy group was more likely to be dissatisfied with time for rest (odds ratio 1.41; 95% CI, 1.06 to 1.89; P=0.02) and perceive negative effects of duty hour policies on resident outcomes that depend on time away from the hospital (odds ratios >1.00, P < 0.001). However, most of these concerns decreased as residents progressed through their residency. In fact, compared to interns, overall well-being was significantly better in the flexible-policy group for PGY2-3s, and even more so for PGY4-5s.

When residents were asked what proportion would choose a standard-arm policy program vs a flexible-arm program if they could go back in time, only 14% preferred standard policy.

Finally, it could be that cumulative years under flexible duty hour policies would result in worse patient outcomes and worse resident wellness. In examining the long-term results of the FIRST trial, we found that there was no significant change over time in resident well-being or patient outcomes. Flexibility continued to have a positive impacts on continuity of care.

ACGME Policy Changes

Accordingly, after a period of detailed review, the ACGME revised duty hour requirements in accordance with the flexible duty hour requirements utilized in the FIRST Trial beginning July 1, 2017.