Rapid Liberation From Mechanical Ventilation, the ICU and Hospital by Using an ICU Dashboard and Alert Program

Background: At the Hospital of the University of Pennsylvania, respiratory therapists (RTs) are required to assess for spontaneous breathing trial (SBT) readiness on all ventilated patients every morning. The RTs manually assess for SBT readiness by gathering different parameters from the electronic health record (EHR). This can become somewhat burdensome in a busy ICU. In addition patients may be ready for SBT at any time throughout the day and more frequent readiness assessments could expedite the liberation process. Our committee looked to leverage the EHR to create a novel program that will continuously screen EHR elements to assist the care team with ongoing SBT readiness assessments. Methods: A novel computer program (called the ABC App) was created, in collaboration with the Data Science team and Penn Center for Innovation to continuously screen all vented patients in the MICU by extracting data from the EHR. The information is then displayed in real-time on an ICU Dashboard, with green indicating SBT ready and red for Not SBT ready. When patients meet criteria for SBT readiness, the system alerts the RT via text message to do an SBT. If the patient is over sedated (based on RASS) the RN and providers are also prompted via text alert to address sedation. For patients not meeting readiness criteria, the parameters limiting weaning (ie, FIO2 or PEEP) are displayed on the dashboard with icons to nudge clinicians to consider weaning that parameter. The ICU dashboard and alerts were implemented in the HUP MICU for a 6-month pilot assessment. Results: Duration of ventilation decreased from 4.6 to 4.0 d. ICU LOS decreased by 1.2 d and hospital LOS dropped by 1.4 d. Conclusions: A real-time electronic dashboard and alert system affected a reduction in duration of mechanical ventilation and ICU length of stay. The RTs have found the text alerts and dashboard to be highly beneficial in the ICU with many competing priorities. With these positive results, we plan to expand the ICU dashboard and text alerting system to all ICUs in the Penn Medicine Health System.