Realtime Adverse Event Predictive Model Reduces Readmissions
4:00pm - 5:00pmWednesday, February 13
Orlando - Orange County Convention Center
In the U.S. 3.3 million patients are readmitted to hospitals annually, costing over $41 billion out of which adverse drug events (ADE) contribute $3.5 billion. ADEs are defined as any harm related to medication intervention and about half of them are preventable through timely intervention. We developed a realtime predictive algorithm that identifies hospitalized patients at high risk for an ADE and facilitates timely pharmacy-led interventions. Since go-live a year ago at a large safety-net hospital, 45,000 inpatients have been screened, 500 ADEs have been prevented and 30-day readmission rates in high risk patients have been reduced by 4%. This user-centric EHR integrated model provides customized actionable insights to the end user. It was developed on design-thinking framework with scientific rigor and clinical engineering. It is robust, field-tested, and clinically impactful with actual $1 million savings and potential for almost $6 million per year. The innovation is easily replicable and scalable across settings.
Define adverse drug event and its impact
Summarize the concept of pharmacy-led interventions
Illustrate design thinking framework and end user collaboration