Results of the multivariable ordinal logistic regression using the proportional odds model
| Ordinal regression (proportional odds model) | OR | 95% CI |
| General | ||
| Age ≤65 years | 1.0 | Reference |
| 65 years<age≤75 | 2.6 | 2.0 to 3.6 |
| Age >75 | 3.6 | 3.0 to 5.0 |
| Male sex (vs female) | 1.7 | 1.3 to 2.1 |
| Comorbidities | ||
| Hypertension alone or CVD alone | 1.5 | 1.2 to 2.0 |
| Hypertension and CVD | 2.4 | 1.7 to 3.6 |
| Chronic lung disease | 2.0 | 1.5 to 2.6 |
| Chronic kidney disease | 1.8 | 1.2 to 2.5 |
| Diabetes mellitus | 1.3 | 0.9 to 1.8 |
| Rheumatic disease | ||
| Rheumatoid arthritis | 1.0 | Reference |
| Systemic lupus erythematosus | 0.5 | 0.2 to 1.1 |
| Vasculitides | 1.1 | 0.7 to 1.5 |
| Other connective tissue diseases | 0.9 | 0.6 to 1.5 |
| Psoriatic arthritis | 0.5 | 0.3 to 0.7 |
| Spondyloarthritides | 0.8 | 0.5 to 1.3 |
| Other rheumatic diseases (not IJDs/CTDs/vasculitis) | 1.0 | 0.6 to 1.8 |
| Medication* | ||
| Methotrexate (monotherapy) | 1.0 | Reference |
| No DMARD therapy | 0.9 | 0.7 to 1.4 |
| Leflunomide | 0.8 | 0.5 to 1.4 |
| Antimalarials | 0.7 | 0.4 to 1.3 |
| Sulfasalazine | 1.1 | 0.6 to 2.1 |
| Immunosuppressants | 2.2 | 1.3 to 3.9 |
| TNF inhibitors | 0.6 | 0.4 to 0.9 |
| Abatacept | 1.3 | 0.5 to 3.0 |
| Rituximab | 5.4 | 3.3 to 8.8 |
| IL-6 inhibitors | 0.7 | 0.3 to 1.5 |
| IL-17/IL-23/IL-12+23 inhibitors | 0.9 | 0.4 to 1.9 |
| JAK inhibitors | 1.8 | 1.1 to 2.7 |
| Disease activity and glucocorticoids | ||
| Remission/low DA, no GCs | 1.0 | (Reference) |
| Remission/low DA, GCs 1–10 mg/day | 1.6 | 1.2 to 2.0 |
| Remission/low DA, GCs>10 mg/day | 4.6 | 1.9 to 11.4 |
| Moderate/high DA, no GCs | 2.0 | 1.3 to 3.1 |
| Moderate/high DA, GCs 1–10 mg/day | 2.4 | 1.5 to 3.7 |
| Moderate/high DA, GCs >10 mg/day | 5.3 | 2.5 to 10.9 |
Ordinal outcome of COVID-19 severity was defined as (1) not-hospitalised, (2) hospitalised but not invasively ventilated and (3) invasively ventilated/deceased.
Missing values imputed via multiple imputation. Effects significant at level α=0.05 are marked in bold. N=2222. Compared with table 1, the following numbers of patients were excluded: 27 patients receiving IL-1 inhibitors, 11 patients receiving belimumab, 8 patients receiving apremilast, 6 patients with non-systemic JIA, 1 patient receiving multiple bDMARDs/tsDMARDs.
*Patients receiving multiple csDMARDs or immunosuppressants (except glucocorticoids) were grouped according to the following hierarchy: immunosuppressants>sulfasalazine>antimalarials>leflunomide>methotrexate. Patients receiving a bDMARD/tsDMARD alone or in combination were considered solely in the bDMARD/tsDMARD group.
bDMARD, biological disease-modifying antirheumatic drug; csDMARD, conventional synthetic disease-modifying antirheumatic drug; CTD, connective tissue disease; CVD, cardiovascular disease; DA, disease activity; DMARD, disease-modifying antirheumatic drug; GC, glucocorticoid; IJD, inflammatory joint disease; IL, interleukin; JAK, Janus kinase; JIA, juvenile idiopathic arthritis; TNF, tumour necrosis factor; tsDMARD, targeted synthetic disease-modifying antirheumatic drug.