Table 2. Multivariable Association between Predictors and Type of Walking Aids Prescribed.
| Variables | Low | High | Odds Ratio (95% CI) a | P-value |
|---|---|---|---|---|
| Age | 61 | 74 | 0.95 (0.44 to 2.06) | .89 |
| Sex | Women | Men | 0.32 (0.11 to 0.93) | .04 |
| Body mass index, kg/m2 | 23.4 | 31.0 | 3.11 (1.50 to 6.44) | <.01 |
| Knee pain intensity* | 1.0 | 3.5 | 0.95 (0.44 to 2.08) | .90 |
| Knee flexion, ° | 85 | 98 | 0.60 (0.33 to 1.09) | .09 |
| Knee extension, ° | 5 | 15 | 2.05 (0.95 to 4.27) | .06 |
| Active Knee lag, ° | 2 | 9 | 1.71 (0.89 to 3.28) | .11 |
| CoP ML-SD, cm | 0.22 | 0.41 | 2.55 (1.23 to 5.30) | <.001 |
aOdds Ratios (ORs) with 95% CIs were derived from proportional odds regression on type of walking aids – an ordinal outcome variable of 4 categories (see further explanation in the text). ORs for requiring walking aids with a larger base-of-support were estimated comparing men with women or the 75th (High) with the 25th (Low) percentile for continuous predictors. For example, other variables being equal, increasing the CoP ML-SD variable from its lower quartile (0.22cm) to its higher quartile (0.41cm) was associated with a 2.55-fold (95%CI, 1.23- to 5.30-fold) increase in the odds of requiring walking aides with a larger base-of-support.
*Assessed using a visual numeric pain scale (0–10), with higher scores indicating worse knee pain.
CoP = center-of-pressure
ML = mediolateral
SD = standard deviation
° degrees