Fig. 3

Differences from comparator in mTSS (month 6 and month 12) according to baseline prognostic factors. ^aMedian baseline mTSS value was 13.1 for ORAL Scan and 4.0 for ORAL Start. LS mean differences from placebo (ORAL Scan) or MTX (ORAL Start) with 95 % CIs of each tofacitinib group vs comparator are presented; a CI that does not contain 0 indicates that the difference is statistically significant (p < 0.05). The ANCOVA model used was the same for each subgroup and included effects for treatment, geographic location, and baseline value of mTSS. The ANCOVA model for the ORAL Start study initially included a categorical variable for duration of RA at baseline. Missing values were imputed by linear extrapolation. Across both studies and tofacitinib doses the number of patients in each subgroup ranged from: 234–298 for anti-CPP+; 42–70 for anti-CCP-; 250–346 for DAS28-4(ESR) >5.1; 22–34 for DAS28-4(ESR) ≤5.1; 206–301 for RF+; 60–72 for RF-; 160–237 for CRP >7 mg/L; 109–131 for CRP ≤7 mg/L; 101–205 for erosion score <3; 162–193 for erosion score ≥3;140–177 for seropositive and erosion score ≥3; 116–164 for not seropositive and erosion score ≥3; 108–122 for CRP >7 and erosion score ≥3; 169–258 for not CRP >7 and erosion score ≥3; 137–181 for baseline mTSS > median; and 140–187 for baseline mTSS ≤ median. ANCOVA analysis of covariance, BID twice daily, CCP cyclic citrullinated peptide, CI confidence interval, CRP C-reactive protein, DAS28-4(ESR) Disease Activity Score in 28 joints (erythrocyte sedimentation rate), LS least squares, mTSS van der Heijde modified total Sharp score, MTX methotrexate, PBO placebo, RF rheumatoid factor