Estimated differences in Clinical Disease Activity Index (CDAI) between type of tocilizumab (TCZ) treatments at various times
| Time (months) | TCZ+MTX vs TCZ Estimate (95% CI) | TCZ+MTXplus vs TCZ Estimate (95% CI) | TCZ+other vs TCZ Estimate (95% CI) |
|---|---|---|---|
| 2 | 0.44 (−0.94 to 1.81) | 1.54 (−0.49 to 3.57) | 1.90 (−0.01 to 3.80) |
| 6 | 0.30 (−0.97 to 1.57) | 1.38 (−0.51 to 3.27) | 1.62 (−0.14 to 3.38) |
| 12 | 0.10 (−1.10 to 1.30) | 1.15 (−0.64 to 2.93) | 1.22 (−0.44 to 2.87) |
| 18 | −0.10 (−1.35 to 1.14) | 0.91 (−0.92 to 2.74) | 0.81 (−0.87 to 2.51) |
| 24 | −0.31 (−1.70 to 1.08) | 0.68 (−1.34 to 2.69) | 0.41 (−1.47 to 2.29) |
Estimated differences and 95% Wald-type CIs for each combination treatment versus monotherapy based on a covariate-adjusted longitudinal mixed effects analysis are shown. A positive difference means that CDAI under monotherapy is estimated lower than under the respective combination treatment at this time point. The p values (from F-tests) for an effect of type of TCZ treatment were 0.16 for the initial linear decrease over 2 months and 0.46 for the subsequent linear phase. All 1428 eligible patients with information on CDAI and complete covariate information were included. The distribution of patients between the four TCZ treatments was comparable to the whole population. Overall, 281 patients lacked a baseline CDAI and 242 provided only one CDAI value (for 176 of these this was the baseline value). Of note, all Swedish patients were excluded due to lack of a global physician's assessment of disease in this registry. All patients from the Netherlands were excluded due to incomplete data.
MTX, methotrexate.