Higher Complications and No Improvement in Mortality in the ACGME Resident Duty-Hour Restriction Era

To the Editor: Recently several reports, including resident surveys and complication and outcome assessments, have raised concerns about the effectiveness of the Accreditation Council for Graduate Medical Education (ACGME) duty-hour regulations instituted in 2003. Hoh et al1 further add to these concerns by demonstrating an increased complication rate, without improvement in mortality in more than 107 000 neurosurgical trauma patients treated in teaching hospitals, after the implementation of these rules.1 Among 115 604 neurosurgical trauma patients treated in nonteaching hospitals, no increased complication rate was found. Therefore, the authors attributed the increased complication rate in teaching hospitals to the implementation of ACGME resident duty-hour restrictions. In our opinion, this conclusion appears to be based on the wrong assumptions. The authors’ conclusion is based on logistic regression analysis in which a number of variables, such as sex and age, are included as covariates, but the injury severity index (ISS) is excluded. However, because the severity of a trauma is the key factor in determining both mortality and the occurrence of complications, the ISS should be included as a primary covariate. No differences in the complication and mortality rates were found after the ISS was included as a covariate in the logistic regression analysis. Therefore, the increased ISS of 1.1 in the teaching hospital group versus the much smaller increase of 0.2 in the nonteaching hospitals is a much more likely culprit than the implementation of ACGME resident duty-hour restrictions. We strongly agree with evaluating new rules. However, in such delicate matters as resident duty-hours, that evaluation should be as careful as possible. Hoh et al performed an admirable analysis of 4 large cohorts of patients, but the conclusion from these data that the ACGME duty-hour regulations have led to increased complication rates in neurosurgical trauma patients appears to be preemptive at best. Disclosure The authors have no personal financial or institutional interest in any of the drugs, materials, or devices described in this article.