Objective: This study quantitatively analyze the evaluation indicators used in premenstrual disorder clinical trials, establishes a placebo response model, observes its influencing factors, and provides a basis for the design of future clinical trials. Methods: A comprehensive search was conducted in PubMed, EMBASE, Cochrane Library, and PsychINFO for randomized controlled trials (RCTs) related to premenstrual disorders. The time course of the placebo responses were modelled and analyzed for common evaluation indicators. Furthermore, the placebo response for each indicator under various conditions was simulated to provide an external control and inform sample size estimation. Results: In the 47 studies included in the analysis, it includes 5406 patients, five common evaluation indicators were identified: total symptom score, depression symptom score, anxiety symptom score, irritability symptom score, and physical symptom score. The maximum placebo response ranged from 20.9% - 38.4%, with the efficacy plateau being reached within 2 - 3 months. The total symptom score is recommended as the primary outcome due to its minimal placebo response size, rapid onset of action, and low inter-trial variability. Using this model as an external control, the study suggests that the efficacy of natural S-equol was not significantly different from that of a placebo. Furthermore, the placebo response model was found to provide accurate effect sizes for sample size estimation under various conditions. Conclusion: The placebo response models of common evaluation indicators in clinical trials of drugs for premenstrual disorders have been established, providing an important basis for the optimized design and decision-making of subsequent clinical trials.
Yang et al. (2026) studied this question.