A hospital's ability to rescue patients from complications after high-risk surgery determines mortality rates
Research Activities, July 2012, No. 383
Hospitals that perform low numbers of a particular surgery typically have higher mortality rates than high-volume centers. These differences in mortality rates are not associated with large differences in complication rates. Instead, they seem to be associated with the ability of a hospital to effectively rescue patients once complications occur, concludes a new study.
Researchers used Medicare data from 2005 to 2007 to identify 37,865 patients who underwent one of three high-risk cancer operations: gastrectomy, pancreatectomy, and esophagectomy. Hospitals were ranked on their average annual volume and divided into quintiles with an equal distribution of patients through each quintile. The researchers also analyzed the incidence of major complications and failure to rescue from complications that do arise.
While patients were of similar age and sex at very high-volume and very low-volume hospitals, low-volume hospitals were found to treat more blacks and sicker patients. The risk-adjusted mortality rate for gastrectomy was 7.5 percent in very high-volume hospitals versus 17.7 percent in very low-volume institutions. It was even more dramatic for pancreatectomy, with a 3.1 percent mortality rate at very high-volume hospitals versus 13.3 percent for very low-volume ones.
Major complication rates remained similar for both types of hospitals, except for pancreatectomy, which had a 1.7-fold difference in major complication rates between very high- and very low-volume hospitals. As a result, patients undergoing this surgery at very low-volume hospitals were 3.2 times more likely to die once complications set in. The researchers also noted marked differences in failure-to-rescue rates for patients undergoing esophageal surgery. Such patients developing complications at very low-volume hospitals had a three-fold increased odds of death compared to their peers at very high-volume hospitals.
The authors call for more research to help all hospitals improve their ability to rescue patients from complications after surgery. The study was supported in part by the Agency for Healthcare Research and Quality (HS17765).
See "Hospital volume and failure to rescue with high-risk surgery," by Amir A. Ghaferi, M.D., M.S., John D., Birkmeyer, M.D., and Justin B. Dimick, M.D., M.P.H., in the December 2011 Medical Care 49(12), pp. 1076-1081.