Occurrence and relative risks of COVID-19-related events and other outcomes in individuals with chronic inflammatory joint diseases (rheumatoid arthritis, ankylosing spondylitis, psoriatic arthritis, other spondyloarthropathies and juvenile idiopathic arthritis), 1 March through September 2020, according to DMARD treatment status 1 March
| Outcome | Cohort | N events | Crude risk (%) | HR (95% CI)* |
| Hospitalisation, all causes | csDMARD | 2805 | 8.4 | 1 (ref) |
| TNFi | 1288 | 5.8 | 0.99 (0.89 to 1.10) | |
| Abatacept | 115 | 8.7 | 0.94 (0.69 to 1.26) | |
| Tocilizumab | 79 | 7.6 | 0.92 (0.64 to 1.33) | |
| Rituximab | 272 | 12.5 | 1.25 (1.02 to 1.53) | |
| JAKi | 146 | 8.5 | 0.93 (0.67 to 1.27) | |
| All b/tsDMARDs | 1900 | 6.7 | 0.99 (0.90 to 1.10) | |
| Hospitalisation due to COVID-19 | csDMARD | 207 | 0.6 | 1 (ref) |
| TNFi | 67 | 0.3 | 1.05 (0.67 to 1.64) | |
| Abatacept | 5 | 0.4 | 0.49 (0.15 to 1.59) | |
| Tocilizumab | 4 | 0.4 | – | |
| Rituximab | 24 | 1.1 | 1.03 (0.58 to 1.81) | |
| JAKi | 18 | 1.0 | 2.72 (1.14 to 6.47) | |
| All b/tsDMARDs | 118 | 0.4 | 1.08 (0.73 to 1.58) | |
| Admission to intensive care due to COVID-19 | csDMARD | 21 | 0.1 | 1 (ref) |
| TNFi | 8 | 0.0 | 2.05 (0.70 to 6.06) | |
| Abatacept | 1 | 0.1 | – | |
| Tocilizumab | 0 | 0.0 | – | |
| Rituximab | 2 | 0.1 | – | |
| JAKi | 1 | 0.1 | – | |
| All b/tsDMARDs | 12 | 0.0 | 1.74 (0.63 to 4.84) | |
| All-cause death | csDMARD | 412 | 1.2 | 1 (ref) |
| TNFi | 73 | 0.3 | 0.71 (0.49 to 1.03) | |
| Abatacept | 16 | 1.2 | 1.12 (0.50 to 2.48) | |
| Tocilizumab | 7 | 0.7 | 1.11 (0.41 to 3.02) | |
| Rituximab | 43 | 2.0 | 2.52 (1.56 to 4.07) | |
| JAKi | 16 | 0.9 | 1.30 (0.52 to 3.26) | |
| All b/tsDMARDs | 155 | 0.5 | 0.91 (0.67 to 1.24) | |
| Death due to COVID-19 | csDMARD | 52 | 0.2 | 1 (ref) |
| TNFi | 7 | 0.0 | 1.03 (0.40 to 2.61) | |
| Abatacept | 1 | 0.1 | – | |
| Tocilizumab | 2 | 0.2 | – | |
| Rituximab | 9 | 0.4 | 3.20 (1.19 to 8.57) | |
| JAKi | 5 | 0.3 | 10.03 (2.35 to 42.76) | |
| All b/tsDMARDs | 24 | 0.1 | 1.26 (0.60 to 2.64) |
*HR from propensity score-weighted Cox regression, adjusted for oral steroids and csDMARD co-medication. Separate models for individual drugs and for all b/tsDMARDs.
b/tsDMARD, biologic/targeted synthetic DMARD; csDMARD, conventional synthetic DMARD; DMARD, disease modifying antirheumatic drug.