ObjectiveTofacitinib is an oral JAK inhibitor that is used for the treatment of rheumatoid arthritis (RA). In previous clinical trials of tofacitinib, a Disease Activity Score in 28 joints (DAS28)–based analysis was used to assess outcomes. In this study, remission rates according to various remission criteria were evaluated across 5 phase III randomized controlled studies.MethodsIn all 5 studies, tofacitinib was administered at a dosage of 5 mg twice daily or 10 mg twice daily, either as monotherapy or with background methotrexate or other conventional synthetic disease‐modifying antirheumatic drugs. One of the studies included adalimumab 40 mg once every 2 weeks. In addition to the 4‐variable DAS28 using the erythrocyte sedimentation rate (DAS28‐4[ESR]), a primary efficacy variable used in the phase III studies, disease activity was assessed post hoc by the 4‐variable DAS28 using the C‐reactive protein level (DAS28‐4[CRP]), the Clinical Disease Activity Index (CDAI), the Simplified Disease Activity Index (SDAI), and Boolean‐based assessment.ResultsA total of 3,306 patients were analyzed (1,213 of these patients received tofacitinib 5 mg twice daily, 1,212 received tofacitinib 10 mg twice daily, 679 received placebo, and 202 received adalimumab 40 mg every 2 weeks). Remission rates varied according to the criteria used, with higher rates in the active‐treatment groups for the DAS28‐4(CRP) than for other scores. At month 3, remission rates with tofacitinib 5 mg twice daily were 18–22% using the DAS28‐4(CRP), 5–10% using the DAS28‐4(ESR), 4–7% using the SDAI, 5–6% using the CDAI, and 2–7% using the Boolean‐based method. In contrast, the remission rates with placebo varied from 0% to 7%, with small differences between the DAS28‐4(ESR) and the DAS28‐4(CRP).ConclusionAlthough tofacitinib at dosages of 5 mg twice daily and 10 mg twice daily was effective compared with placebo in achieving disease remission, regardless of the disease activity measure, remission rates were substantially higher when the DAS28‐4(CRP) was used. The presence or absence and type of acute‐phase reactants in remission criteria were significant contributors to remission rates across treatment groups. This finding has important consequences for trial design and clinical practice.