Overestimation of Cardiovascular Mortality Risk by Kaplan-Meier in Competing Risks Settings: A Web-Based Calculator and NHANES Analysis
Bibliographic record
Abstract
Background: Traditional Kaplan-Meier (KM) event rates are widely used for cardiovascular risk prediction and tend to overestimate absolute event risk for patients by censoring competing events, such as non-cardiovascular death. Competing risks analysis (CRA), which account for such terminal events, offers more accurate estimates. However, its application in a web-based health analytics remains limited. Methods: Using a simulated cohort (n = 2,500; 100 repetitions) and the 1999–2000 NHANES cohort (n = 2,480) with 2019 National Death Index mortality linkage, the researcher compared KM estimates to CRA’s Cumulative Incidence Function (CIF), implemented via Aalen-Johansen estimators and Fine-Gray subdistribution hazard models. We assessed relative differences (bias) in 5-, 10-, 15-, and 20-year cardiovascular mortality predictions across risk strata. Findings informed a web-based calculator prototype that dynamically estimates age-specific KM and CIF probabilities while highlighting potential misclassification risks. Results: KM consistently overestimated cardiovascular mortality risk compared to CIF. In the NHANES cohort, KM estimated the 5-year risk to be 5.85% higher than the actual rate (4.37% vs. 4.13%) and 20-year risk by 28.3% (20.02% vs. 15.60%). In the simulated data, KM overestimated the 5-year risk by 7.63% (5.84% vs. 5.42%) and the 20-year risk by 31.17% (21.37% vs. 16.25%). KM-based models tend to misclassify a substantial portion of patients into higher-risk groups compared to CIF-adjusted models. Conclusion: This study demonstrates that Kaplan-Meier consistently overestimates cardiovascular mortality in comparison to competing risk methods across five time points, through using both simulated and nationally representative data. We quantify this overestimation and provide an online calculator that shows differences by age. Our tool improves the usability and interpretability of competing risks analysis for older adults in digital health settings, in contrast to tools like SCORE2.
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 | 0.184 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.003 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| 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".