Duty hour restrictions and surgical complications for head and neck key indicator procedures.

Objectives/hypothesis: Graduate medical education has traditionally required long work hours, allowing trainees little time for adequate rest. Based on concerns over performance deterioration with sleep deprivation and its effect on patient outcomes, duty hour restrictions have been mandated. We sought to characterize complications from otolaryngology key indicator procedures performed before and after duty hour reform. Study

Design: Retrospective cross-sectional analysis of National Inpatient Sample (NIS).

Methods: The NIS was queried for procedure codes associated with head and neck key indicator groupings for the years 2000-2002 (45,363 procedures) and 2006-2008 (51,144 procedures). Hospitals were divided into three groups: nonteaching hospitals (NTH), teaching hospitals without otolaryngology programs (TH), and teaching hospitals with otolaryngology programs (TH-OTO). Surgical complication rates, length of stay, and mortality rates were analyzed using logistic and linear regression.

Results: The number of procedures increased (12.7%), with TH-OTO contributing more in postrestriction years (21% to 30%). Overall complication rates between the two periods revealed no difference, regardless of hospital setting. Subset analysis showed some variation within each complication within each grouping. Length of stay increased at TH-OTO (2.75 to 2.78 days) and decreased at NTH (2.28 to 2.24 days) and TH (2.39 to 2.36 days). Mortality did not increase among the three hospital types (NTH, P < .58; TH, P < .96; TH-OTO, P < .06). During the latter period, TH-OTO procedures showed lower mortality (P < .0038, odds ratio [OR] = 0.45, 95% confidence interval [CI] = 0.27-0.77). Increasing Charlson comorbidity index increased overall mortality rate (P < .0001, OR = 2.63, 95% CI = 2.4-2.89).

Conclusions: Overall complication rates did not change for head and neck key indicator procedures. Moreover, concerns about reduced surgical case numbers appear unfounded, especially for otolaryngology programs. Level of evidence: 2c Laryngoscope, 127:1797-1803, 2017.