372-P: Predicting Real-World Severe Hypoglycemia in Americans with Diabetes (iNPHORM)
Bibliographic record
Abstract
Most prediction models for diabetes-related iatrogenic severe hypoglycemia (SH) have derived from trial/administrative records subject to poor generalizability, ascertainment bias, and incomplete data capture. Redressing this gap, iNPHORM leveraged the clinical and methodological advantages of prospective self-report to develop and internally validate a 1-year SH prediction model for use in real-world clinical contexts. Adults (18-90 years old) with insulin- and/or secretagogue-treated type 1 or 2 diabetes (T1D, T2D) were recruited from a US-wide probability-based internet panel and followed for one year. Monthly emailed questionnaires assessed SH incidence and related factors. To model recurrent 1-year SH (daytime + nocturnal) , Andersen-Gill Cox proportional hazards regression was performed on participants completing ≥1 follow-up. Missing data were multiply imputed with chained equations. Machine learning penalized regression with lasso was used to select clinically plausible predictors. A total of 986 (T1D: 17%) participants were analyzed (retention rate: 86.2%) . The mean age was 51 (SD: 14.3) years, 49.6% were male, and the median duration of T1D/T2D was 12 (IQR: 14) years. Among T2D participants, 38% were on insulin (without secretagogues) , 38% on secretagogues (without insulin) , and 24% on insulin plus secretagogues. Across follow-up, 35.1% (95% CI: 32.2-38.1%) reported ≥1 SH, and the annual rate was 4.97 (95% CI: 4.13-5.99) . Combination insulin-secretagogue therapy; use of an insulin pump and continuous glucose monitoring; decreased age; increased previous SH requiring healthcare utilization; chronic kidney disease; and food insecurity predicted 1-year SH risk. The optimism adjusted c-statistic was 0.75. iNPHORM is the first long-term, prospective study on SH prediction in the general US population with T1D and T2D. Our 7-variable model can be used to identify patients at high-risk of SH, leading to more valid, cost-effective prevention strategies in the real world. Disclosure A.Ratzki-leewing: Consultant; Eli Lilly and Company, Other Relationship; Sanofi. S.B.Harris: Consultant; Abbott, AstraZeneca, Eli Lilly and Company, Novo Nordisk, Sanofi, Other Relationship; Abbott, AstraZeneca, Bayer Inc., Dexcom, Eli Lilly and Company, HLS Therapeutics, Janssen Pharmaceuticals, Inc., Novo Nordisk, Sanofi, Research Support; Applied Therapeutics Inc., AstraZeneca, Canadian Institutes of Health Research, Juvenile Diabetes Research Foundation (JDRF) , Novo Nordisk, Sanofi, The Lawson Foundation. J.E.Black: None. G.Zou: None. S.Webster-bogaert: None. B.L.Ryan: None. Funding Sanofi Global
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 | 0.003 | 0.011 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.009 | 0.002 |
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".