Impact of introducing multiple evidence-based clinical practice protocols in a medical intensive care unit: a retrospective cohort study

Background: Recently completed clinical trials have shown that certain interventions improve the outcome of the critically ill. To facilitate the implementation of these interventions, professional organizations have developed guidelines. Although the impacts of the individual evidence-based interventions have been well described, the overall impact on outcome of introducing multiple evidence-based protocols has not been well studied. The objective of this study was to determine the impact of introducing multiple evidence-based protocols on patient outcome. Methods: A retrospective, cohort study of 8,386 patients admitted to the medical intensive care unit (MICU) of an academic, tertiary medical center, from January 2000 through June 2005 was performed. Four evidence-based protocols (lung protective strategy for acute lung injury, activated protein C for severe sepsis/septic shock, intravenous insulin for hyperglycemia control and a protocol for sedation/analgesia) were introduced in the MICU between February 2002 and April 2004. We considered the time from January 2000 through January 2002 as the pre-protocol period, from February 2002 through March 2004 as the transition period and from April 2004 through June 2005 as the protocol period. We retrieved data including demographics, severity of illness as measured by the Acute Physiology and Chronic Health Evaluation (APACHE) Ill, MICU length of stay and hospital mortality. Student's t, Kruskal-Wallis, Mann-Whitney U, chi square and multiple logistic regression analyses were used to compare differences between groups. P-values < 0.05 were considered significant. Results: The predicted mean mortality rates were 20.7%, 2 1. 1% and 21.8%, with the observed mortality rates of 19.3%, 18.0% and 16.9% during the pre-protocol, transition and protocol periods, respectively. Using the pre-protocol period as a reference, the severity-adjusted risk (95% confidence interval) of dying was 0.777 (0.655 - 0.922) during the protocol period (P = 0.0038). The average 28-day MICU free days improved during the protocol period compared to the preprotocol period. The benefit was limited to sicker patients and those who stayed in the MICU longer. Conclusion: The introduction of multiple evidence-based protocols is associated with improved outcome in critically ill medical patients.