656 CHEST Original Research CRITICAL CARE Original Research D ecisions to withdraw or withhold life support are routinely made in the ICU when patients, surro- gate decision-makers, and the health-care team tran- sition from curative to comfort care. 1 The two most important factors influencing such decisions are patient preferences and patient prognosis. 2,3 Numerous addi- tional patient-, provider-, and surrogate-related fac- tors impact such decisions and create significant variability in decision-making. 4-6 In an era when up to 20% of all adults die in the ICU and one-third of all health-care dollars in the United States are used in the last year of life, understanding how decisions to limit life support are made and implementing strat- egies to improve decision-making have been the sub- jects of continued research. 1,7-10 Additionally, there has been a growing trend and recommendations toward the use of continuous, 24-h, intensivist staffing of ICUs. 11 This is typically accom- plished by alternating daytime and nighttime intensivist shifts. The impact of adding continuous, attending intensivist coverage in the ICU has been associated with improvement in a number of patient outcomes Background: A growing trend is the implementation of 24-h attending physician coverage in the ICU. Our aim was to measure the impact of 24-h, in-house, attending intensivist coverage on the quality of end-of-life care and the timing of end-of-life decision-making. Methods: A retrospective cohort study was conducted of all ICU deaths 6 months before and 6 months after the implementation of mandatory 24-h attending intensivist coverage in a medical ICU. Data relevant to end-of-life care per established consensus recommendations were abstracted from the medical record. Results: The following changes were observed after implementation of 24-h intensivist coverage: Time from ICU admission to decision to withdraw mechanical ventilation and time to decision to change to do-not-resuscitate code status both were shortened by 2 days (both P 5 .03). Quality measures, such as increased family presence around time of death ( P 5 .01) also improved. Other findings, which did not reach statistical significance, included the following: Time to family con- ference was shortened by 2 days ( P 5 .09), time to decision to limit any life support was shortened by 1 day ( P 5 .08), time to death was shortened by 2 days ( P 5 .08), and intubations against patient wishes decreased (from three to none; P 5 .12). Conclusions: The implementation of mandatory 24-h, in-house, attending intensivist coverage was associated with earlier decision-making across a number of domains related to end-of-life care. Positive trends were noted in quality of end-of-life care as reflected in the presence of family at the time of death. CHEST 2013; 143(3):656–663 Abbreviations: DNR 5 do not resuscitate Physician Staffing Models Impact the Timing of Decisions to Limit Life Support in the ICU Michael E. Wilson, MD; Ramez Samirat, MD; Murat Yilmaz, MD; Ognjen Gajic, MD, FCCP; and Vivek N. Iyer, MD, MPH Manuscript received May 9, 2012; revision accepted August 8, 2012. Affiliations: From the Department of Internal Medicine (Dr Wilson), Divisions of Pulmonary and Critical Care Medi- cine (Drs Gajic and Iyer), Department of Internal Medicine, Mayo Clinic, Rochester, MN; Department of Internal Medicine (Dr Samirat), University of Miami Jackson Memorial Hospital, Miami, FL; and the Department of Anesthesiology and Intensive Care (Dr Yilmaz), Akdeniz University, Antalya, Turkey. Funding/Support: Financial support for this study was provided by the Mayo Clinic and Mayo Foundation. Correspondence to: Vivek N. Iyer, MD, MPH, Pulmonary and Critical Care Medicine, Mayo Clinic, 200 First St SW, Rochester, MN 55905; e-mail: iyer.vivek@mayo.edu © 2013 American College of Chest Physicians. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details. DOI: 10.1378/chest.12-1173 Downloaded From: http://journal.publications.chestnet.org/ on 09/12/2014