Improving the Quality of CKD Care With Risk Prediction and Personalized Recommendations: The GEMINI Project
Bibliographic record
Abstract
Background: CKD affects 1 in 7 Americans and can lead to progression to dialysis, cardiovascular disease, and early mortality. Effective interventions exist to slow the progression of CKD and prevent heart failure, but implementation remains a challenge, and use of guideline recommended testing and therapies remain low. Routine, complete collection of guideline recommended blood and urine tests allowing accurate risk prediction with personalized treatment recommendations can improve CKD care, when integrated into clinical workflow. Our objective is to implement risk prediction algorithms clinical decision support for identifying patients at risk of CKD progression in 5 large nephrology practices representing more than 100 nephrologists and 100,000 patients with CKD. Methods: Data for estimated glomerular filtration rate, albuminuria, demographics, other laboratory tests and comorbid conditions will be extracted from the electronic health record (EHR). Patients already on dialysis will be excluded. The remaining individuals will be risk stratified using Klinrisk's proprietary risk prediction equations. A dashboard with disease specific educational information, personalized treatment recommendations and links to the EHR's of the identified patients will be created. Results: We will aim to enroll 5 leading large US nephrology practices in the next 12 months. Eligible patients will be identified and quality of care as defined by appropriate testing (proportion of patients with albuminuria testing within 12 months), and appropriate therapy as recommended by the relevant guidelines (RAASi, SGLT2i, and non-steroidal MRA use) will be measured in the pre and post implementation period. Conclusions: A highly accurate machine model for CKD progression when paired with EHR linked clinical decision support will improve testing and management of intermediate and GEMINI high-risk patients with CKD. Larger randomized trials of clinical decision support and practice audit applications will be needed to impact CKD management in primary care. Funding: Commercial Support - Bayer U.S.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.049 | 0.048 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.004 | 0.003 |
| Open science | 0.002 | 0.008 |
| Research integrity | 0.001 | 0.003 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".