Better Than Predicted Outcomes in an Open Intensive Care Unit Model: Low Mortality in Patients Cared for by Nonintensivist Physicians.
Chest. 2012 Oct 1;142(4_MeetingAbstracts):419A
Authors: Daniel S, Zhong F, Nguyen C, Martynowicz M, Dutta A, Lin T, Escalante J, Sabo J, Esquivel A
SESSION TYPE: Improving Processes and Outcomes in Adult Critical CarePRESENTED ON: Wednesday, October 24, 2012 at 02:45 PM - 04:15 PMPURPOSE: The LeapFrog ICU guidelines require that all ICU (Intensive Care Unit) patients are managed by intensivists. Recent data suggests that ICU patients managed by non-intensivists may have favorable outcomes. We studied ICU performance in a low intensity staffing ICU model at a tertiary care hospital, where patients admitted for monitoring were managed by non-intensivists. We evaluated the outcomes of patients cared for by different specialties.METHODS: A retrospective cohort study was conducted on all patients admitted by a single large practice group to the ICUs at the St. Luke's Episcopal Hospital, Houston for a year. Patients with an ICU length of stay of less than 24 hours were excluded. Hospital mortality and ICU length of stay for ICU patients were compared with those predicted using APACHE IV, and MPMο III prognostic models. These two outcomes were used to assess outcomes by specialty.RESULTS: Hospital mortality rate (6.37%, 95% CI 4.96-8.00) for 1054 ICU patients between July 2007 and June 2008 was 48% to 51% lower than predicted by the 2 models; APACHE IV model (12.25%, p<0.001), and MPMο III (12.92%, p<0.001). The ICU average length of stay was significantly shorter than predicted by APACHE IV (3.89±4.35 vs. 4.42±1.96, P<0.001). Hospital mortality for patients cared for by intensivists (16.72%) was 7% to 19% lower than predicted, for hospitalists (2.28%) was 77 % to 81% lower than predicted, for cardiologists (1.46%) was 76 % to 82% lower than predicted. Mortality for patients cared for by surgeons was zero versus predicted mortality of 3.81 to 9.64%.CONCLUSIONS: Despite a low intensity staffing, open ICU care model at a tertiary care teaching hospital, the clinical outcomes of hospital mortality and ICU length of stay for ICU patients managed by one practice group are better than predicted. Both non-intensivists and intensivists achieved better than expected outcomes.CLINICAL IMPLICATIONS: This study suggests that a subset of ICU patients admitted for monitoring can be safely managed by non-intensivists, which could help alleviate a worsening shortage of intensivists.DISCLOSURE: Suresh Daniel: Employee: Employed by the Kelsey-Seybold clinic. Fengwei Zhong: Employee: Employed by the St. Luke's Episcopal Health System. Christina Nguyen: Employee: Was employed by the Kelsey Research Foundation for the duration of the study. Half her salary was funded by the Kelsey Research Foundation and half by the St.Luke's Episcopal Hospital. Tony Lin: Employee: Employed by the Kelsey Seybold clinic. Jonathan Escalante: Employee: Employed by the St.Luke's Episcopal Health system. John Sabo: Employee: Employed by the St. Luke's Episcopal Health System.Adol Esquivel: Employee: Employed by the St.Luke's Episcopal Health System. The following authors have nothing to disclose: Marek Martynowicz, Amit DuttaNo Product/Research Disclosure InformationBaylor College of Medicine, Houston, TX.
PMID: 23698612 [PubMed - as supplied by publisher]