The abstract is read at the publisher; the summary is JClub's.
Algorithm-based CKD auto-loading in Geisinger’s EHR problem lists increased CKD diagnosis within 1 year from 28.4% to 65.7% (vs 22.6% to 20.7% in OLDW).
- Why it matters: CKD is frequently underrecognized and undiagnosed, limiting timely diagnosis and management. This work tests whether automated EHR problem-list updates can improve real-world CKD care.
- What they did: Using summary-level data from 2008–2023, the study compared Geisinger (2013–2023 auto-load) with 4 Optum Labs Data Warehouse health systems (2008–2012 vs 2013–2023). It evaluated incident, undiagnosed CKD cohorts (n=21,995 vs n=208,402) and applied difference-in-differences.
- The result: The difference-in-differences estimate for CKD diagnosis within 1 year was 39.1% (95% CI: 38.1%, 40.1%; p<0.001), with favorable effects on follow-up eGFR, parathyroid hormone, and phosphorus testing (all p<0.001) and increased RAASi and statin use (p<0.05).
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.