With the recent birth of the Pennsylvania TQIP Collaborative, statewide data identified unplanned admissions
to the Intensive Care Unit
(ICU) as an overarching issue plaguing the state trauma community. To better understand the impact of this unique population, we sought to determine the effect of unplanned ICU admission/readmission on mortality to identify potential predictors of this population. We hypothesized that ICU bounceback
) patients would experience increased mortality compared to non-ICUBB
controls and would likely be associated with specific patterns of complications.
The Pennsylvania Trauma Outcome Study database was retrospectively queried from 2012-2015 for all ICU admissions. Unadjusted mortality rates were compared between ICUBB
counterparts. Multi-level mixed-effects logistic regression models assessed the adjusted impact of ICUBB
on mortality and the adjusted predictive impact of 8 complications on ICUBB
58,013 ICU admissions were identified from 2012-2015. From these, 53,715 survived their ICU index admission. The ICUBB
rate was determined to be 3.82% (2,054/53,715). Compared to the non-ICUBB
patients had a significantly higher mortality rate (12% vs. 8%; p <0.001). In adjusted analysis, ICUBB
was associated with a 70% increased odds ratio for mortality (AOR: 1.70, 95% CI: 1.44-2.00; p <0.001). Adjusted analysis of predictive variables revealed unplanned intubation, sepsis, and pulmonary embolism as the strongest predictors of ICUBB
ICUBBs are associated with worse outcomes and are disproportionately burdened by respiratory complications. These findings emphasize the importance of the TQIP Collaborative in identifying statewide issues in need of performance improvement within mature trauma systems.
LEVEL OF EVIDENCE
Level III epidemiological study