Objectives: A change in our children's hospital coverage model to providing full-time in-house supervision by intensivists allowed us to evaluate the impact of this change on patient safety outcomes. Our aim was to determine whether in-house attending coverage influenced the prevalence and outcomes of pediatric code events.
Design: We conducted a retrospective review of all code events between October 2005 and October 2007 (before in-house intensivist supervision) and compared the prevalence, interventions, and outcomes of these codes with those occurring between April 2008 and April 2010 (after in-house intensivist supervision). A code event was defined as any activation of the code system.
Setting: One hundred eighty-seven bed children's hospital.
Subjects: All children with code events.
Interventions: None.
Measurements and main results: There were 99 codes during these two periods: 39 codes occurring prior to in-house intensivist coverage (of which eight on the ward and 31 in the ICU) and 60 occurring following in-house attending coverage (30 on the ward and 30 in the ICU). Survival was significantly improved following the implementation of in-house coverage (odds ratio, 4.3; 95% CI, 1.7-10.8; p = 0.003). There was no significant change in the overall rate of codes during these two periods (0.82 codes/1,000 patient-days before implementation vs 1.17 codes/1,000 patient-days after implementation). However, there were significantly more codes on the ward following in-house intensivist coverage (0.2 codes/1,000 patient-days before implementation vs 0.71 codes/1,000 patient-days after implementation; p = 0.013). An intensivist was significantly more likely to be present during these events (odds ratio, 28; 95% CI, 3-273; p = 0.001); however, the acuity of the children with codes on the ward was significantly lower during the in-house coverage period (p = 0.001). There were no changes in the rate or outcomes of codes occurring in the ICU with this change in coverage.
Conclusions: In the period following implementation of in-house intensivist supervision, children with code events were more likely to survive to hospital discharge. Having an intensivist in-house 24 hr/d, 7 d/wk may be associated with improved outcomes in hospitalized children.