Work

Association between the use of traditional Chinese medicine patterns as an inclusion criterion and statistical significance in primary outcomes of randomized placebo-controlled trials of Chinese herbal formulas

Year: 2026

Type: article

Abstract: Randomized, placebo-controlled trials of Chinese herbal formulas (CHFs) are commonly used to evaluate efficacy, but a proportion of such trials fail to demonstrate statistically significant effects on primary outcomes. Some researchers have pointed out that excluding traditional Chinese medicine (TCM) pattern differentiation from inclusion criteria may reduce the alignment between treatment and enrolled patient characteristics, which could, in turn, affect observed efficacy. This study analyzed 378 randomized, placebo-controlled trials of CHFs with clearly defined primary outcomes that were published between January 1, 2020, and October 21, 2025, and identified through searches of eight databases. A total of 92,691 participants were included in the primary outcome analyses. Of these, 324 trials (85.7%) reported statistically significant results on primary outcomes, and 197 (52.1%) used TCM patterns as an inclusion criterion. Multivariable Poisson regression revealed that trials using TCM patterns were more likely to report statistically significant results (adjusted risk ratio [aRR] = 1.14; 95% CI: 1.06–1.23; P = 0.001). The association was generally consistent across prespecified subgroups and sensitivity analyses; however, only 52 studies (13.8%) were evaluated as low risk of bias according to the Cochrane Risk of Bias 2.0 tool. Because this study included only published trials and few published pattern-based trials reported statistically nonsignificant primary outcomes, the observed association should be interpreted cautiously and may be influenced by publication bias, selective outcome reporting, and contextual differences in research and reporting practices. Further high-quality, prospectively registered, methodologically rigorous, and better-standardized trials are needed to evaluate this association.

Source: Acta Pharmaceutica Sinica B

DOI: 10.1016/j.apsb.2026.07.033