2025 Volume 74 Issue 2 Pages 285-292
95% confidence intervals for the coefficient of variation (95%CICV) were estimated by the boot-strap (BS) method. The BS method was independently implemented using Excel VBA. The sample size for BS method was set to 50, and the number of iterations was set to 100. The 95% CICV was estimated from the mean and 2.5 and 97.5 percentile values of the distribution of CV% calculated by the BS method. The 95% CICVs were compared for 12 clinical laboratory tests of healthy reference individuals across eight Asian cities, including Osaka, Hong Kong, and Hanoi. The results showed significant intercity differences in γGT, triglycerides and IgA. Conversely, a significant correlation was found only in ALP between the latitude of each city and the range of the 95% confidence intervals (r = −0.727, p = 0.041). However, this relationship was a pseudo-correlation, with small intestinal ALP, whose activity in serum is increased in blood types B and O, being a confounding factor. Estimation of 95% CICV by the BS method is practical in that it does not require complex calculations and is extremely useful for measurement method evaluation and assessment of interindividual differences.
