Fig. 4

Assessment of change in CRP level: crude mean (± SD) change in CRP level from baseline to month 6 (A). Adjusted mean change (beta coefficient ± 95% CI) in CRP level from baseline to month 6, comparing IL-6Ri with TNFi and JAKi separately (B). Proportions of patients with high month 6 CRP level, proportions who changed from normal baseline to high month 6 CRP level, and proportions who changed from high baseline to normal month 6 CRP level (C). Odds of having a high month 6 CRP level, changing from normal baseline to high month 6 CRP level, or changing from high baseline to normal month 6 CRP level, comparing IL-6Ri with TNFi and JAKi separately (D). *p < 0.05; **p < 0.01; ***p < 0.001. High CRP was defined as ≥ 0.8 mg/dL. Adjusted model covariates were baseline CRP, age, duration of rheumatoid arthritis, EuroQol-5 Dimension score, Health Assessment Questionnaire score, sex, prior use of ≥ 2 TNFi, white race, history of hyperlipidemia, and CDAI (baseline CDAI and 6-month CDAI). BL, baseline; CDAI, Clinical Disease Activity Index; CI, confidence interval; CRP, C-reactive protein; IL-6Ri, interleukin-6 receptor inhibitor; JAKi, Janus kinase inhibitor; M, month; OR, odds ratio; SD, standard deviation; TNFi, tumor necrosis factor inhibitor