Diary of hot flashes reported upon occurrence: results of a randomized double-blind study of raloxifene, placebo, and paroxetine.

Objective: This trial examined diaries of hot flash events reported upon occurrence to assess the test/retest reliability of the diaries and their ability to measure treatment effects on hot flash frequency and severity.

Methods: Forty-two postmenopausal women (aged ≥40 y; 5-50 hot flashes/wk) were randomized (3:3:1) to placebo, raloxifene 60 mg, or paroxetine 20 mg daily for 12 weeks. Diaries of hot flash frequency and severity were evaluated at 1-week intervals (twice before study treatment and thrice during study treatment).

Results: Forty-one women were evaluated. Baseline characteristics were similar between groups (eg, mean, 29.8 hot flashes/wk). Concordance correlation coefficients between screening (week -2) and baseline (week -1) measures of hot flash frequency and severity were 0.73 and 0.71, respectively. After 12 weeks, the mean (95% CI) percent changes from baseline in weekly hot flash frequency were as follows: placebo, -37.4% (-60.9 to -14.0); raloxifene, -14.2% (-37.7 to 9.3); paroxetine, -49.8% (-88.6 to -11.0); the mean (95% CI) percent changes in hot flash severity were as follows: placebo, -39.9% (-69.1 to -10.8); raloxifene, -9.6% (-38.8 to 19.6); paroxetine, -36.6% (-84.7 to 11.5). There were no significant differences in hot flash diary results between treatment groups.

Conclusions: Measures of hot flash frequency and severity show acceptable test/retest reliability between screening and baseline. Reductions in vasomotor symptoms by raloxifene are numerically less than those seen with placebo, but no statistically significant treatment differences have been documented in this small study. The large effect of placebo and the significant reduction in vasomotor symptoms by paroxetine are consistent with other studies. The diary seems to be suitable for use in hot flash clinical trials.