To optimize two Tibetan medicine intervention schemes for rheumatoid arthritis (RA) to improve the transparency, standardization, and scalability of their formulation.

Based on a systematic review of relevant randomized controlled trial (RCT) evidence, appropriateness scales were developed for each Tibetan formula. A two-round research and development corporation (RAND)/University of California, Los Angeles (UCLA) appropriateness method (RAM) consensus process (independent scoring followed by face-to-face discussion) was employed to finalize the intervention plan. The clinical efficacy of the two final schemes was assessed in RCTs using a prespecified performance goal of 60%.

Twenty-eight RCTs informed the scale for RAM round 1. The results of the scale showed that 13 items (56.5%) showed disagreement. Round 2 finalized schemes A and B. In patients with RA (disease activity score 28: 2.6–5.1), scheme A enrolled 59 patients (15 dropouts), and scheme B enrolled 62 patients (8 dropouts). At week 12, using intention-to-treat non-responder imputation, the American College of Rheumatology 20% (ACR20) response rates were 54% and 65% for schemes A and B, respectively; per-protocol analysis yielded 66.9% and 70.8%, respectively. Scheme B met the industry's minimum efficacy criterion, whereas scheme A required more rigorous data owing to higher dropouts.

We propose an objective, transparent, and repeatable methodological paradigm for traditional Chinese medicine complex interventions and verify that the final scheme has good therapeutic effects.