Thomas L. Higgins, MD, MBA; Brian Nathanson, PhD; Daniel Teres, MD
Potential Financial Conflicts of Interest:Consultancies: T.L. Higgins (Cerner), B. Nathanson (Cerner).
Higgins TL, Nathanson B, Teres D. What Conclusions Should Be Drawn between Critical Care Physician Management and Patient Mortality in the Intensive Care Unit?. Ann Intern Med. 2008;149:767. doi: 10.7326/0003-4819-149-10-200811180-00014
Download citation file:
Published: Ann Intern Med. 2008;149(10):767.
TO THE EDITOR:
We were intrigued by the results of Levy and colleagues' study (1), which demonstrated higher odds for hospital mortality in patients managed by critical care physicians. We recently used the Project IMPACT database (Cerner, Kansas City, Missouri) to update the Mortality Probability Model (MPM). Although our sample of 124 885 patients at 98 hospitals between October 2001 and March 2004 is not identical, the overlap is substantial. We analyzed the variable, “Percentage of patient stay managed by critical care physician/team.” In our sample, 68 179 patients (54.6%) were coded as 0% critical care management (CCM), 50 694 (40.6%) were coded as 100% CCM, and 6012 (4.6%) had a value between 0% and 100%. The mortality rate was higher in the CCM group at every decile of risk. However, most patients (64% of non-CCM patients and 54% of CCM patients) were in the lowest decile of MPM-III mortality risk, in which risk-adjusted outcome was exactly as expected (standardized mortality rate for both non–CCM- and CCM-managed patients, 1.0 [95% CI, 0.95 to 1.05]). There were also no differences in standardized mortality rates in patients in whom the admission model (MPM0-III) risk prediction was greater than 50%. Standardized mortality rate differences occurred at the second and third deciles of MPM-predicted risk, exactly where the calibration curve for this model deviates from “perfect” prediction (2). Although our analysis uses MPM, we suspect similar calibration considerations may apply to the Simplified Acute Physiology Score (SAPS), which was used by Levy and colleagues. A calibration curve of their observed versus predicted mortality, along with measures of model fit, would be informative.
to gain full access to the content and tools.
Learn more about subscription options.
Register Now for a free account.
Copyright © 2016 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use
This PDF is available to Subscribers Only