忘记用户 ID? or 忘记密码?

Not a Member?


Menu 返回 Poster-Presentations-Details

P207: The Comparison of BMI Scores Between Japanese and Caucasian Healthy Volunteers in Clinical Trials and its Implications

Poster Presenter

      Nozomu Miyawaki

      • Senior Manager, Clinical Research Division


This study was conducted to examine BMI distributions for Japanese and Caucasian healthy subjects for both sexes screened at our site with the aim of proposing more appropriate BMI reference ranges for the enrollment of these ethnic populations in clinical trials.


A retrospective analysis of our screening data obtained from January 2016 to June 2020 was conducted. The BMI scores of Japanese and Caucasian healthy subjects aged 20-49 were examined. The Japanese and Caucasian BMI data were stratified by sex, and 95% CIs were estimated for each ethnic group.


The number of healthy subjects whose BMI scores were examined for the purpose of this study was 12,195. The breakdown of the number of the study subjects by race and sex is as follows: Japanese males: 10,245; Japanese females: 1,039; Caucasian males: 813; Caucasian females: 98. The two-way ANOVA of the collected BMI data indicated a race-sex interaction with a significant difference (p<0.01). Accordingly, the BMI data were stratified by sex, and the 95% confidence intervals of the mean differences between the ethnic groups were estimated. The mean BMI of the Japanese female subjects was 1.11kg/m2 lower than that of the Caucasian female subjects. However, this difference was not statistically significant (p=0.09). The mean BMI of the Japanese male subjects was 2.38kg/m2 lower than that of the Caucasian male subjects. This difference was statistically significant (p<0.01). The current study suggests that the degrees of ethnic differences in BMI distribution differ between males and females.


Received an increasing number of Caucasian study requests at our site in Japan, and in response to growing Japanese study needs from overseas sponsors, we have noticed that the same BMI reference range is often used for both ethnic groups. However, based on our over 36-year clinical trial experience, we believe it is reasonable to set more appropriate BMI reference ranges which reflect BMI distributions of each ethnic group. Although the National Statistics Data on BMI reference range are often used, it is also pointed out that it does not necessarily reflect the actual BMI distribution of study participants in clinical trials. Therefore, in order to identify and enroll the study subjects more effectively, the BMI distributions of Japanese and Caucasian subjects screened at our site were investigated. The results of this study have shown that the BMI distribution differs between Japanese male subjects and Caucasian male subjects, thus supporting the use of unique BMI reference ranges for each ethnic group. No statistically significant differences were observed between the BMI distribution of Japanese female subjects and that of Caucasian female subjects. This may be because the number of available female data was small, especially for Caucasian females (N=98). Further accumulation of female subject data is necessary to have better statistical power. Setting appropriate BMI reference ranges for the enrollment of different ethnic groups will facilitate efficient subject recruitment.