Osteoarthritis

Hip Morphology–Based Osteoarthritis Risk Prediction Models: Development and External Validation Using Individual Participant Data From the World COACH Consortium

Arthritis Care Res (Hoboken)Published Prediction model development and external validation study

Authors' conclusion

In hips free of definite RHOA, our AUCs for the incident RHOA models showed good predictive performance in similar populations. However, the added predictive value of the morphology measurements was small, and model performance was heterogeneous in leave-one-cohort-out cross-validation.

Study context

Population
Hips without definite radiographic hip OA (radiographic hip osteoarthritis (RHOA)) at baseline, nine prospective cohorts of the World COACH consortium
Participants
35,984 hips
Comparison
Generalized linear mixed model (GLMM) with vs without hip morphology measurements; hip morphology-based GLMMs vs random forest (RF) models
Outcome
Incident RHOA: hips without definite RHOA at baseline that developed definite RHOA
Follow-up
Four to eight years
Methodological detailsindividual participant data from nine prospective cohort studies, World COACH consortium, stratified cross-validation, leave-one-cohort-out cross-validation, GLMM and RF models
  • Individual participant data from nine prospective cohort studies
  • World COACH consortium
  • Stratified cross-validation
  • Leave-one-cohort-out cross-validation
  • GLMM and RF models

Performance

As reported in the abstract

Discrimination for incident RHOA (AUC), stratified cross-validation
Test or modelValueAUC
GLMM with hip morphologymean 0.80 ± 0.010.80

GLMM with hip morphology

Valuemean 0.80 ± 0.01
AUC0.80

Further findings

Incidence of RHOA · four to eight years

Hips developing incident RHOA
4.7%

Generalizability, leave-one-cohort-out cross-validation

Comparable discrimination (AUC range) and calibration for hip morphology-based GLMMs and RF models
0.56-0.88

Original publication

Hip Morphology–Based Osteoarthritis Risk Prediction Models: Development and External Validation Using Individual Participant Data From the World COACH Consortium. Arthritis Care Res (Hoboken). Published 14 January 2026. doi:10.1002/acr.25629

Composed by AI from the published abstract and checked against it. How Briefs are made →

Get them every week

New studies on this topic in My Week and in your Monday e-mail.

We e-mail you a sign-in link and a code. No password, no spam.