Association between the medications for RA and HZ in the NCC study.
| Medications | OR (95% CI) | p |
|---|---|---|
| Use of TNFi#** | 2.28 (1.09–4.76) | 0.028* |
| PSL dosage (per 1 mg/day increment) ** | 1.13 (1.03–1.23) | 0.008* |
To assess the association of the medications for RA and HZ, use of TNFi, and use of non-TNFi, MTX dosage and PSL dosage were put into the conditional logistic model as dependent variables with a stepwise backward elimination method.
↵* Statistically significant after correction for multiple testing using false discovery rate and Benjamini and Hochberg method28.
↵# Versus no use of biological DMARD.
↵** Use of TNFi during the 90 days preceding the index date was collected. Prednisolone dosages were collected at the index date. The eliminated variable (i.e., use of non-TNFi and MTX) was collected as shown in the legend of Table 2. RA: rheumatoid arthritis; HZ: herpes zoster; NCC: nested case control; TNFi: tumor necrosis factor inhibitor; MTX: methotrexate; PSL: prednisolone; DMARD: disease-modifying antirheumatic drugs.