Table 2.
Rate of resolution of swollen joints after start of TNFi in pairwise comparisons of joints (subgroups of patients affected at both sites)
| Site of joint | Number of affected joints/patients | Reference joint(s) | HR (95% CI) | |
|---|---|---|---|---|
| Proximal–distal axis of upper limb | Shoulder | (74/56) | MCP3/PIP3 | 3.27 (1.27; 8.42) |
| Elbow | (99/78) | MCP3/PIP3 | 3.89 (1.80; 8.40) | |
| Wrist | (537/338) | MCP3/PIP3 | 0.75 (0.63; 0.90) | |
| Anterior–posterior axis of the hand (digits 1–5) | MCP1/PIP1 | (490/264) | MCP3/PIP3 | 1.43 (1.14; 1.79) |
| MCP2/PIP2 | (1107/539) | MCP3/PIP3 | 1.24 (1.05; 1.46) | |
| MCP4/PIP4 | (650/352) | MCP3/PIP3 | 2.01 (1.65; 2.44) | |
| MCP5/PIP5 | (431/247) | MCP3/PIP3 | 2.33 (1.83; 2.96) | |
| Knee vs. elbow | Knee | (103/75) | Elbow | 0.55 (0.27; 1.10) |
Estimates of resolution rates (HR and 95% CI) from individual interval-censored mixed-effects Cox proportional hazards models. Confounding is minimized since patient characteristics are equally distributed for both joint sites in each comparison