Machine-Learning Assessed Abdominal Aortic Calcification is Associated with Long-Term Fall and Fracture Risk in Community-Dwelling Older Australian Women
Main finding · Clinical fracturesResult — go to the resultHazard ratio 1.42 95% CI 1.10–1.85Authors' conclusion
In conclusion, this novel automated method for assessing AAC, that can be easily and seamlessly captured at the time of bone density testing, has robust associations with long-term incident clinical fractures and injurious falls.
Study context
- Population
- Community-dwelling older women (mean age, 75 ± 3 years), Perth Longitudinal Study of Ageing Women; Abdominal aortic calcification (AAC) scored on DXA lateral spine images
- Participants
- 1,023 women
- Exposure
- Moderate to extensive AAC vs low AAC
- Outcome
- Incident falls and fractures
- Follow-up
- Long-term
Methodological detailsmachine-learning algorithm (ML-AAC24), adjusted for baseline fracture risk
- Machine-learning algorithm (ML-AAC24)
- Adjusted for baseline fracture risk
Main finding
Clinical fractures · 10 years
Hazard ratio
1.42
95% CI 1.10–1.85
Moderate to extensive AACvs Low AAC
HR 1.42 (95% CI 1.10–1.85)
As reported in the abstract
- Women with clinical fracture (overall)
- 253 (24.7%)
Observational data: associations, not effects.
Further findings
| Outcome | HR (95% CI) | |
|---|---|---|
| Fall-related hospitalization · 14.5 yearsAdjusted for baseline fracture risk | 1.35 (1.09–1.66) | |
| 0.512Hazard ratio (log scale) |
- Fall-related hospitalization — Women with injurious fall requiring hospitalization (overall)
- 393 (38.4%)
Fracture hospitalization · 10 years
- Women with fracture hospitalization (overall)
- 169 (16.5%)
- ML-AAC24 not associated with fracture hospitalizations, similar to manually assessed AAC-24
ML vs manual AAC-24
- Relative hazard estimates with ML similar to manually assessed AAC-24
Original publication
Machine-Learning Assessed Abdominal Aortic Calcification is Associated with Long-Term Fall and Fracture Risk in Community-Dwelling Older Australian Women. J Bone Miner Res. Published 30 October 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →