Abstract: SA-PO547
Provider-Level Exposure to AKI Alerts in an Ongoing Randomized Trial
Session Information
- AKI: Clinical, Outcomes, Trials - II
October 27, 2018 | Location: Exhibit Hall, San Diego Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Acute Kidney Injury
- 102 AKI: Clinical, Outcomes, and Trials
Authors
- Wilson, Francis Perry, Yale School of Medicine, New Haven, Connecticut, United States
- Martin, Melissa, Yale School of Medicine, New Haven, Connecticut, United States
- Yamamoto, Yu, Yale University , New Haven, Connecticut, United States
- Moreira, Erica, Yale University , New Haven, Connecticut, United States
- Parikh, Chirag R., Yale University and VAMC, New Haven, Connecticut, United States
- Feldman, Harold I., University of Pennsylvania, Philadelphia, Pennsylvania, United States
- Garg, Amit X., London Health Sciences Centre, London, Ontario, Canada
- Palevsky, Paul M., University of Pittsburgh/VA Pittsburgh Healthcare System, Pittsburgh, Pennsylvania, United States
- Ugwuowo, Ugochukwu Caleb, Yale University , New Haven, Connecticut, United States
- Biswas, Aditya, Yale University , New Haven, Connecticut, United States
Background
Electronic alerts for acute kidney injury (AKI) are increasingly considered in clinical care because of the potential harms that may come from missed diagnoses of AKI. Alert fatigue, however, may reduce the effectiveness of AKI alerts.
Methods
The Electronic Alerts for AKI Amelioration (ELAIA-1) trial began in March 2018 and will randomize 6,030 patients to real-time, electronic health record-based AKI alerts or usual care across 6 hospitals. Alerts fire when the electronic health record is opened by an MD, NP, or PA while AKI diagnostic criteria are met. In this preliminary analysis, we describe the differential exposure to AKI alerts at the provider level.
Results
Two months into the study, 10,665 alerts had fired for 217 patients with AKI over a six-week period, at a median of 30 (IQR 13-72) alerts per patient. At the patient level, the median alert duration was 1.1 (0.5-2.9) days. A total of 996 unique providers were exposed to alerts with a median of 0.09 (.05-0.31) alerts per day per provider. Fellows received the most alerts per day while attending physicians and PAs received the fewest. Alert burden was highest in the ICU (Figure) and substantially lower in the hospital wards and emergency department.
Conclusion
Despite the large total number of alerts, provider alert burden was generally low. Fellows experienced the highest median number of alerts per day, a potential signal for higher risk of alert fatigue. Providers caring for patients in the ICU may also be at higher risk of alert fatigue.
Exposure to AKI Alerts by Provider Type
| Provider | Unique Providers | Alerts Per Day, Median (IQR) | Alerts Per Day (Max) |
| Attending | 450 | 0.08 (0.04 - 0.16) | 3.77 |
| Fellow | 119 | 0.21 (0.08 - 0.49) | 2.35 |
| Resident | 416 | 0.12 (0.04 - 0.40) | 4.59 |
| Physician Assistant | 93 | 0.10 (0.02 - 0.29) | 3.64 |
| Nurse Practitioner | 58 | 0.14 (0.06 - 0.33) | 3.29 |
Funding
- NIDDK Support