Table 4. Results from a linear mixed quantile regression model with absolute percent bias in the odds ratio estimated via PQL or QUAD as the dependent variable and absolute percent difference in the odds ratios as estimated via PQL and QUAD as the independent variable, adjusted for data set characteristics (β1, σ2 u, proportion with the outcome (p), total number of observations in the data set and data set composition).
| Data | QUAD | PQL | |||
| Generation Parameter | Value | Slope (95% CI) | Vara | Slope (95% CI) | Vara |
| Absolute percent difference in ORQUAD and ORPQL | – | 6.50 (4.58, 8.43) | 8.17 | 1.17 (0.79, 1.56) | 1.28 |
| β1 | Log(1) | Ref | Ref | ||
| Log(1.5) | −0.01 (−0.08, 0.06) | 0.01 (−0.02, 0.04) | |||
| Log(2) | −0.01 (−0.10, 0.08) | 0.00 (−0.05, 0.04) | |||
| σ2u | 0 | Ref | Ref | ||
| 1 | −0.04 (−0.12, 0.04) | 0.00 (−0.04, 0.05) | |||
| 4 | −0.11 (−0.20, −0.02) | 0.00 (−0.03, 0.04) | |||
| 16 | −0.11 (−0.20, −0.02) | 0.02 (−0.01, 0.06) | |||
| p | 0.05 | Ref | Ref | ||
| 0.2 | −0.05 (−0.12, 0.02) | −0.17 (−0.22, −0.11) | |||
| 0.5 | −0.12 (−0.20, −0.04) | −0.20 (−0.27, −0.14) | |||
| Total n | 150 | Ref | Ref | ||
| 450 | 0.00 (−0.07, 0.06) | −0.14 (−0.20, −0.09) | |||
| 1500 | −0.01 (−0.08, 0.07) | −0.25 (−0.32, −0.17) | |||
| Dataset composition | Many large cluster | Ref | Ref | ||
| Many small clusters | 0.10 (0.09, 0.10) | −0.02 (−0.13, 0.09) | |||
| Moderate | 0.29 (0.11, 0.46) | −0.06 (−0.10, −0.03) | |||
a : This is the variance of the random slope.