Estimation of placebo effect in randomized placebo-controlled trials for moderate or severe vasomotor symptoms: a meta-analysis.

OBJECTIVE: The Food and Drug Administration guidance recommended that for the indication of vasomotor symptoms (VMS), studies should enroll participants with minimum of seven to eight moderate to severe VMS per day at baseline, and coprimary endpoints should be the mean change in frequency/severity at weeks 4 and 12. This study aimed to estimate placebo effects in randomized controlled trials (RCTs) fulfilling this guidance. METHODS: PubMed was searched using the following terms: "vasomotor symptom or hot flash or hot flashes or hot flash or hot flashes [title/abstract]," "menopause or climacteric," "RCT or randomized controlled or randomized controlled," "placebo [title/abstract]," and "frequency or severity." Inclusion criteria were as follows: (1) placebo controlled RCTs, (2) enrolling women with moderate or severe VMS with a minimal frequency more than seven to eight times per day or 50 times per week, and (3) efficacy measurements including mean change in VMS frequency from baseline at week 12. A random-effects model was used in the meta-analysis. RESULTS: Seventeen studies were included for the estimation of VMS frequency reduction and 13 studies for severity. Estimated change of VMS frequency in placebo arms was -5.44 times per day (95% CI, -5.81 to -5.07 times per day) at week 12. For VMS severity, the estimated change at week 12 was -0.36 (95% CI, -0.46 to -0.27). CONCLUSION: Substantial and consistent placebo effects were observed in RCTs for VMS treatment. These data suggest a reduction of 5.44 times per day in frequency and 0.36 in severity might be observed as a placebo effect.