Randomized Trial of AKI Alerts in 6,030 Hospitalized Patients

F. Perry Wilson from Yale University conducted a double-blinded, multicenter randomized controlled trial involving 6,030 adult inpatients with Acute Kidney Injury (AKI). The trial ran over 22 months across six diverse hospitals to evaluate the impact of EHR-based AKI alerts on a composite primary outcome of AKI progression, dialysis, or death within 14 days. The results, published as an open-access paper, showed no overall clinical benefit from the alerts.

Use Cases

• Analyzing the impact of electronic alerts on patient mortality based on the trial's primary outcome measure.
• Comparing clinical outcomes between teaching and non-teaching hospital settings using the per-hospital analysis results.
• Studying the adoption of specific AKI care practices following an alert intervention as mentioned in the secondary outcomes.
• Evaluating the design and results of a large-scale, pragmatic randomized controlled trial in a real-world hospital setting.

Strengths

• Data originates from a rigorous, double-blinded, multicenter randomized controlled trial design.
• Includes results from 6,030 patient participants, providing a substantial sample size.
• Trial was conducted across six diverse hospital settings, including teaching and non-teaching centers.
• Primary and secondary outcomes are clearly defined and reported with statistical measures (relative risk, confidence intervals, p-values).

Limitations

• The actual trial data (rows, columns, sample data) is not directly accessible; only the published results summary is provided.
• Column-level documentation is absent; field semantics and data structure must be inferred from the paper.
• Row count for the underlying dataset is unknown, which may limit suitability assessment for direct analysis.

Provenance

Source

Yale University (author F. Perry Wilson), published via paperswithcode.

Collection Method

Double-blinded, multicenter, parallel, randomized controlled trial; participants electronically identified and randomized.

Time Range

Trial conducted over 22 months (specific dates not provided).

Geography

Six hospitals in the United States (inferred from author institution and trial registration), ranging from small community to large tertiary care centers.