This study aimed to elucidate which bio-naïve patients with rheumatoid arthritis (RA) are suitable for treatment with CTLA4-Ig. This study enrolled 953 patients with RA who were administered their first biological disease-modifying antirheumatic drug (CTLA4-Ig, n = 328; tumour necrosis factor inhibitor [TNFi], n = 625) from July 2013 to August 2022. The primary outcome was the Clinical Disease Activity Index (CDAI) remission rate at week 24 in each group, adjusted using Propensity Score-based Inverse Probability of Treatment Weighting (PS-IPTW). After minimizing selection bias using PS-IPTW, the CDAI remission showed no significant difference between the CTLA4-Ig and TNFi groups (p= 0.464). Multivariable logistic regression analysis identified low baseline Health Assessment Questionnaire-Disability Index (HAQ-DI) scores as a contributing factor to the CDAI remission rate at week 24 in both groups, along with high baseline anti-citrullinated peptide antibody (ACPA) levels in the CTLA4-Ig group. However, among patients with high baseline HAQ-DI scores and low baseline ACPA levels (≦57.2), the CDAI remission rate was significantly higher in the TNFi group (29.8%) compared with the CTLA4-Ig group (5.9%, p< 0.0001). Among patients with high baseline HAQ-DI scores and ACPA levels (>57.2), the CDAI remission rate was significantly higher in the CTLA4-Ig group (35.6%) compared with the TNFi group (22.1%, p= 0.0057). Bio-naive RA patients with low HAQ-DI scores showed high treatment efficacy with no significant difference between CTLA4-Ig and TNFi. Among patients with high baseline HAQ-DI scores, TNFi and CTLA4-Ig were more likely to be effective in those with lower and higher baseline ACPA levels, respectively.