[Skip to Content]
[Skip to Content Landing]
Views 382
Citations 0
Comment & Response
July 16, 2019

Outcomes Associated With Overlapping Surgery—Reply

Author Affiliations
  • 1Department of Anesthesiology, Perioperative and Pain Medicine, Stanford University School of Medicine, Stanford, California
  • 2Department of Anesthesiology, University of Michigan School of Medicine, Ann Arbor
  • 3Department of Health Care Policy, Harvard Medical School, Boston, Massachusetts
JAMA. 2019;322(3):275-276. doi:10.1001/jama.2019.6475

In Reply Drs Urwin and Emanuel raise 2 concerns with our study on overlapping surgery.1 First, they note that the reported surgical times may not be representative, as the mean surgical times in the study are longer than times from other sources.2 These differences may stem from the fact that the institutions in our study were largely tertiary academic referral centers, where cases may take longer because of increased complexity and the presence of trainees. In this sense, our study may not be representative, as it did not include many relevant settings (ie, practice in community hospitals), but the magnitude of difference between the surgical times in our study and the times from these other sources (20 minutes) is consistent with the distribution of and variation in case times nationwide. For example, for THA, a large national study of 89 802 cases found a median time of 91 minutes with an interquartile range of 68-109 minutes and with a substantial proportion of cases taking 100 minutes or longer, and similar results were found in a large national study of patients undergoing TKA.3 In addition, the difference in times is not as large as the authors suggest when a true apples-to-apples comparison is made, because we reported mean times, whereas the studies cited by Urwin and Emanuel report median times. Because the distribution of surgical times tends to have a positive skew,4 a comparison of median times with mean times will tend to exaggerate differences across studies.