In the group with no radiographic changes at baseline, less endurance in one-leg rise from sitting was a significant predictor of developing incident knee osteoarthritis (OA) 5 years later, controlled for one confounder at a time using logistic regression analyses. Variables dichotomised by the median value
| Predictor | n* | OR (95% CI)§ |
|---|---|---|
| *Number of patients included in each analysis; †bivariate analysis. Only one predictor at a time was included in the logistic regression together with maximum number of one-leg rises, because of the rather small samples; ‡maximum number of one-leg rises, controlled for age, sex, body mass index, and pain in multivariate logistic regression; §odds ratio (OR) with 95% confidence interval (CI) for the maximum number of one-leg rises (less than v greater than the median). | ||
| Age† | 90 | 2.48 (1.04 to 5.89) |
| Sex† | 90 | 2.50 (1.05 to 5.95) |
| Body mass index† | 88 | 2.78 (1.16 to 6.67) |
| Pain (VAS) baseline† | 72 | 2.79 (1.02 to 7.68) |
| All above‡ | 70 | 2.67 (0.92 to 7.76) |