Learning Alerts for Acute Kidney Injury
Study Details
Study Description
Brief Summary
In this trial, individuals with Acute Kidney Injury will be randomized to usual care or to an electronic AKI alert system. In three phases, the alert system will be trained to identify those most likely to benefit from the alert, and inclusion criteria will be modified.
Condition or Disease | Intervention/Treatment | Phase |
---|---|---|
|
N/A |
Detailed Description
Acute kidney injury (AKI) carries a significant, independent risk of mortality among hospitalized patients. Recent studies have demonstrated increased mortality among patients with even small increases in serum creatinine concentration. International guidelines for the treatment of AKI focus on appropriate management of drug dosing, avoiding nephrotoxic exposures, and careful attention to fluid and electrolyte balance. Early nephrologist involvement may also improve outcomes in AKI. Without appropriate provider recognition of AKI, however, none of these measures can be taken, and patient outcomes may suffer. AKI is frequently overlooked by clinicians, but carries a substantial cost, morbidity and mortality burden.
The investigators conducted a pilot, randomized trial of electronic alerts for acute kidney injury in 2014. The trial, which randomized 2400 patients with AKI as defined by an increase in creatinine of 0.3mg/dl over 48 hours or 50% over 7 days, found that alerting physicians to the presence of AKI did not improve the course of acute kidney injury, reduce dialysis or death rates. However this study was conducted in a single hospital, and the alert itself did not describe specific actions that a provider could take. In the present proposal, the investigators seek to expand upon their prior study to determine both the modes of alerting that would be most effective and to determine if targeting alerts (such as to patients on medications that may worsen acute kidney injury) will improve effectiveness.
This study will be a randomized, controlled trial of an electronic AKI alert system. Using the Kidney Disease: Improve Global Outcomes creatinine criteria, inpatients at several hospitals will be randomized to usual care versus electronic alerting. Over three separate study intervals, uplift models will be created to identify clusters of patients expected to benefit most from alerting, and randomization will be tailored to that group.
Study Design
Arms and Interventions
Arm | Intervention/Treatment |
---|---|
No Intervention: Usual Care No alert will be fired |
|
Experimental: Alert An alert informing the provider of acute kidney injury will be fired. |
Other: Alert
An alert informing the provider of the presence of acute kidney injury will be fired.
|
Outcome Measures
Primary Outcome Measures
- Proportion of patients with progression to a higher stage of AKI OR Dialysis OR Death [14 days from randomization]
Eligibility Criteria
Criteria
Inclusion Criteria:
-
Phase 1 of the trial will enroll all inpatients at study hospitals who meet Kidney Disease: Improving Global Outcomes Stage 1 Creatinine criteria for AKI
-
Phase 2 and phase 3 of the trial will use a machine-learning based algorithm to target enrollment to those patients expected to derive benefit from alerting, based upon our experience in the prior trial period.
Exclusion Criteria:
-
Prior randomization
-
Admission to hospice service
-
End-stage renal disease
-
Initial hospital creatinine greater than 4.0mg/dl.
Contacts and Locations
Locations
No locations specified.Sponsors and Collaborators
- Yale University
Investigators
None specified.Study Documents (Full-Text)
None provided.More Information
Publications
None provided.- YALEAKIALERTLEARN