Machine Learning Prediction for Functional Impairment, Falls, and Fractures in Postmenopausal Women
Bibliographic record
Abstract
Abstract Machine learning-based prediction models have gained support for their robustness and flexibility, particularly with complex data common in geriatrics research. They may be well-suited to studying complex aging questions, such as predicting functional status among postmenopausal women. The objective of this study is to develop an interactive prediction model for prevalent, clinically relevant, geriatric health outcomes: functional impairment, falls, and fractures. We simulated a dataset (n = 20,000) to emulate baseline characteristics of the Women’s Health Initiative (WHI) study population, a large study of postmenopausal women. The simulated dataset had a mean age of 63.2 years (SD: 7.25), 82.4% non-Hispanic White. Maximum follow-up was 25 years. We examined random survival forest, XGBoost, and Lasso-Cox models. Models were trained and compared across multiple domain-specific feature sets comprising demographic, socioeconomic, behavioral, functional, and health-related features, selected from existing evidence, with time-dependent ROC-AUC, calibration, and Brier score. SHapley Additive exPlanations (SHAP) guided feature selection. The random survival forest outperformed other models in predicting functional impairment, falls, and fractures. Reduced models incorporating 30 features demonstrated comparable performance relative to the comprehensive models with 57 features, while improving computational efficiency. These reduced models demonstrated strong discrimination (ROC-AUCs >0.80) and predictive accuracy (Brier scores < 0.10) across most time points and were subsequently deployed in the interactive application. This study demonstrates the utility of machine learning-based prediction to guide targeted interventions by capturing the complex, heterogeneous nature of geriatric health outcomes. Deploying such advanced models into a user-friendly application may support data-driven decision-making in geriatric clinical practice.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 teacher head, 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".