Usefulness of home blood pressure measurements in assessing the effect of treatment in a single-blind placebo-controlled open trial

Objectives Reproducibility of home blood pressure measurements (HBP), and the placebo effect on HBP, were examined to establish the number of subjects required in order to observe a significant antihypertensive effect of a drug for a comparative study between placebo and active treatment, Methods Reproducibility of HBP (n = 172 for systolic blood pressure and n = 137 for diastolic blood pressure) and the placebo effect on HBP (n = 35 for systolic and n = 42 for diastolic blood pressure) were examined using a semi-automatic electronic device on untreated hypertensive subjects during 21 consecutive days of measurements. From these two studies, the number of subjects required in order to observe a significant antihypertensive drug effect was assessed, In both studies, measurements from the first 3 days were excluded from the analysis, Results Reproducibility, defined as the difference between the initial 5 day average (days 4-8) and the last 5 day average (days 17 - 21) was -1.9 +/- 7.0/ -1,4 +/- 4.8 mm HQ (mean +/- SD), indicating high reproducibility with a minimal regression to the mean effect A placebo was administered after the 8 day run-in period. The difference between the initial 5 day average (days 4-8) and the last 5 day average (days 17-21) was compared, The mean difference was 1,1 +/- 6.2/0.2 +/- 5.7 mmHg, suggesting that there was a minimal, if any, placebo effect. As determined by the power calculations, based on the standard deviations of the tests for reproducibility and the placebo effect, nine of 13 subjects are required to show that a 9/5 mmHg decrease in systolic HBP/diastolic HBP (50% of estimated effective decrease in systolic HBP/diastolic HBP according to the guidelines for clinical trials in Japan) is statistically significant, Conclusions The results suggest that the HBP measurement is highly reproducible and that there is a minimal, if any, placebo effect. These characteristics of HBP contribute to minimizing the number of subjects necessary for assessment of antihypertensive drug effects in comparative studies between an active treatment and placebo. (C) 2001 Lippincott Williams & Wilkins.