SESSION TITLE: Quality Improvement in the ICU I
SESSION TYPE: Original Investigation Slide
PRESENTED ON: Monday, October 28, 2013 at 01:45 PM - 03:15 PM
PURPOSE: To evaluate the outcomes of intensive care unit (ICU) patients with high predicted hospital mortality who survive hospitalization
METHODS: Among 83,898 ICU episodes between 2008 and 2011 at 21 community hospitals in Kaiser Permanente Northern California, we identified patients with a predicted hospital mortality of >=50% at the time of ICU admission (based on a validated electronic ICU severity score, the eSAPS3). We grouped patients by predicted mortality (50-59%, 60-69%, and >=70%) and hospital survival and compared their baseline and hospital characteristics. We further conducted a focused chart review of all survivors at a single hospital to provide a detailed description of pre- and post-hospital functional status and long-term mortality. We compared groups with analysis of variance and chi-squared tests; we evaluated survival with Kaplan-Meier analysis and log-rank testing.
RESULTS: We included a total of 1,700 ICU episodes, among 1,663 patients, in our analysis. Half of patients (50.6%) had a predicted hospital survival of 50-59% with a mean eSAPS3 score of 81 +- 2; roughly one quarter were in the 60-69% and >=70% groups, respectively. Observed mortality closely matched predicted mortality in all groups. Sepsis was the admission diagnosis in more than one-third of all patients. Among patients who survived their hospitalization, estimated survival was 66.5% (95% confidence interval, 62.6%-70.1%) at 6 months, 57.7% (53.5%-61.6%) at 1 year, and 48.7% (43.9%-53.4%) at 2 years. Long-term survival estimates were not different in the 42 patients who underwent focused chart review from a single center (p=0.80). Among these survivors, 84.6% were ambulatory (independently or with a cane/walker) prior to hospitalization. Among those surviving 1 month after discharge, 65.5% were ambulatory; at 6 months, 78.3% of survivors were ambulatory.
CONCLUSIONS: Many patients with high predicted mortality on ICU admission die during hospitalization. However, among those that survive, roughly half survive to 2 years and many are ambulatory after hospitalization.
CLINICAL IMPLICATIONS: Our findings may help clinicians and families make decisions about aggressive ICU care among severely ill ICU patients.
DISCLOSURE: The following authors have nothing to disclose: James Louisell, Gabriel Escobar, Benjamin Turk, Daniel Skully, Danny Sam, Vincent Liu
No Product/Research Disclosure Information