Comparison of objective and patient-reported hot flash measures in men with prostate cancer

Hot flashes are one of the bothersome symptoms frequently experienced after endocrine treatments for breast and prostate cancers. Many studies have evaluated interventions for hot flashes, but results are obscured by methodologic limitations. We compared the performance of three techniques to measure hot flashes over 48 hours among 47 patients with prostate cancer undergoing androgen deprivation therapy to determine the feasibility and accuracy of each measure. Sternal skin conductance, electronic event marking, and twice daily diaries identified 478, 410, and 285 hot flashes, respectively. Diaries produced the lowest hourly hot flash rate (M = 0.17), which was significantly different from the rates of the objective profile (M = 0.28) and event marks (M = 0.23).The sensitivity and positive predictive value of the three measures demonstrated that the diaries underperformed, but these values did not exceed moderate levels for any measure (% = 28-59). Study results suggest the combined use of sternal skin conductance and event marking for the measurement of hot flash frequency in pathophysiologic studies and clinical trials with prostate cancer patients. Conversely, the use of retrospective diaries may be adequate for clinical practice to determine clinically significant changes in salient events.