|
Abstract:
|
It has been demonstrated that clinical prediction tools could help shape healthcare provider behavior towards more evidence-based clinical practice, particularly in emergency department (ED) practice where time pressures for diagnostic, treatment and disposition decisions persist. Although well-validated risk prediction models automated through electronic health records (EHR) system with sensible implementation strategy could greatly enhance its adoption, the need for complete risk factor information in a timely fashion is not circumventable at the time of clinical decision making. This is especially challenging in complex conditions such as acute heart failure (AHF), where more than a handful of risk factors contribute to the risk calculation. Motivated by implementation of STRATIFY, a risk prediction model developed to identify low-risk AHF patients who could be safely discharged home, I will demonstrate the use of the “precondition” method to develop strategies in calculating predicted risks in the presence of various incomplete risk factor scenarios.
|
