What does this research mean for the field?

A machine learning-based clinical decision support tool utilizing registry data can accurately predict short- and long-term hip fracture risk without in-person assessment, identifying significantly more at-risk individuals than traditional secondary prevention approaches. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

To develop and validate an automated, high-performing machine learning clinical decision support tool (FRACTURE-ML) for nationwide hip fracture prediction without requiring in-person patient assessments.

August 29, 2026PLoS MedicineOpen Access

A clinical decision support tool for accurate hip fracture prediction: A nationwide cohort study

View Full Paper

Ask AI

Bookmark

To develop and validate an automated, high-performing machine learning clinical decision support tool (FRACTURE-ML) for nationwide hip fracture prediction without requiring in-person patient assessments.
Nationwide cohort included 3,542,647 individuals aged ≥50 years residing in Sweden without recent osteoporosis treatment, followed from baseline (2011–2013) through 2021 (142,327 incident hip fractures).
Evaluated 139,980 demographic, diagnostic, procedural, and medication variables across discovery (25%), development (65%), and holdout (10%) datasets using Cox models, XGBoost, and DeepSurv algorithms.
The DeepSurv-based FRACTURE-ML model using 2,500 predictors achieved an AUC of 0.89 (95% CI 0.88, 0.89) at 1 year and 0.88 (95% CI 0.87, 0.88) at 2 years, while a reduced 35-predictor model yielded an AUC of 0.87 at 2 years.
For 2-year prediction, FRACTURE-ML demonstrated higher sensitivity (0.84, 95% CI 0.82, 0.85) than standard fracture liaison services screening (0.12, 95% CI 0.11, 0.13; AUC 0.55, 95% CI 0.54, 0.55), with a specificity of 0.79 (95% CI 0.79, 0.79) versus 0.98 (95% CI 0.98, 0.98).

Discussion

No takes yet. Share an insight, caveat, or question.

Implication

Nationwide cohort study demonstrates accurate automated hip fracture prediction in older adults, indicating potential for population-wide screening without in-person assessment.

Cite This Study

Axelsson et al. (2026) studied this question.

synapsesocial.com/papers/6a9299958e5d7d1fc0c11a22 https://doi.org/10.1371/journal.pmed.1005190

View Full Paper

Ask AI

Bookmark

Implication

Nationwide cohort study demonstrates accurate automated hip fracture prediction in older adults, indicating potential for population-wide screening without in-person assessment.

A clinical decision support tool for accurate hip fracture prediction: A nationwide cohort study

Key Points

Discussion

Implication

Cite This Study

A clinical decision support tool for accurate hip fracture prediction: A nationwide cohort study

Key Points

Discussion

Implication

Cite This Study

Also Consider

Also Consider