Increased rate of complications on a neurological surgery service after implementation of the Accreditation Council for Graduate Medical Education work-hour restriction.

OBJECT The Accreditation Council for Graduate Medical Education instituted mandatory 80-hour work-week limitations in July 2003. The work-hour restriction was met with skepticism among the academic neurosurgery community and is thought to represent a barrier to teaching, ultimately compromising patient care. The authors hypothesize that the introduction of the mandatory resident work-hour restriction corresponds with an overall increase in morbidity rate. METHODS This study compares the morbidity and mortality rates on an academic neurological surgery service before and after institution of the work-hour restriction. Complications are individually assessed at a monthly divisional conference by neurosurgical faculty and residents. A prospective database was commenced in July 2000 recording all complications, complications that were deemed to be potentially avoidable ("possibly preventable"), and complications that were deemed unavoidable. The incidence of morbidity and mortality from July 2000 to June 2003 is compared with the incidence from July 2003 to June 2006. RESULTS The overall rate of morbidity and mortality increased from 103 to 114 per 1000 patients treated after institution of the work-hour restriction, although this increase was not statistically significant (χ(2)(1, N = 8546) = 2.6, p = 0.106). The morbidity rate increased from 70 to 89 per 1000 patients treated after institution of the work-hour restriction (χ(2)(1, N = 8546) = 10, p = 0.001). The overall mortality rate was diminished from 32 to 27 per 1000 patients treated after institution of the work-hour restriction (χ(2)(1, N = 8546) = 3.2, p = 0.075). Morbidities considered avoidable or possibly preventable were seen to increase from 56 to 66 per 1000 patients treated (χ(2)(1, N = 8546) = 5.7, p = 0.017). Avoidable or possibly preventable mortalities numbered 3 per 1000 patients treated, and this rate did not change after introduction of the work-hour restriction (χ(2)(1, N = 8546) = 0.08, p = 0.777). CONCLUSIONS The morbidity rate on a neurological surgery service is increased after implementation of the work-hour restriction. Mortality rates remain unchanged.