Appendix Table 1.

Multivariable Models in Prediction of Incident Symptomatic Knee Osteoarthritis (OA) and Internal Validation (Calibration and Discrimination) of the Risk Prediction Models

Selected PredictorsBasic ModelBackward Model
OR (95% CI)P ValueOR (95% CI)P Value
Age (years)1.07 (0.98–1.17).151.14 (1.02–1.28).02
BMI (kg/m2)1.12 (1.04–1.20)<.011.16 (1.06–1.26).001
Mild knee symptoms*1.73 (0.85–3.49).13
Knee problems while climbing stairs2.08 (1.03–4.23).04
Morning stiffness2.46 (0.97–6.25).06
Postmenopausal status0.54 (0.26–1.12).10
Physically demanding work§2.38 (0.83–6.83).11
AUC of the model0.63 (0.54–0.71)0.72 (0.64–0.80)
Calibration: Hosmer-Lemeshow P value.96.20
Variance explained (Nagelkerke), %6.521.9
  • AUC, area under the curve of the receiver operating characteristic; BMI, body mass index; CI, confidential interval; OR, odds ratio.

  • Bold indicates P < .05.

  • * Pain in or around the knee within the last 12 months.

  • Defined with the Knee Injury and Osteoarthritis Outcome Score (KOOS) on physical functioning.

  • Defined with the KOOS on stiffness.

  • § Defined as doing intense work (regularly lifting heavy objects at work).