[Set-up of standard value by sex and age of biochemical tests derived from blood donors].

We set up the standard value of 7 biochemical tests from the data of approximately 237,000 blood donors who are healthy at least objectively. The abnormal data were excluded by Grubbs Smirnov method, and then converted by the log equation in order to simulate to normal distribution. Then the reference interval was set at mean +/- 2SD. If this failed, 95% percentile was adopted after non-parametric analysis. The data fluctuated much between sex and generations. Especially sex difference was dominant in ALT, AST, gamma-GTP and TP. ALT and AST was greater in males than in females, although age difference was smaller with aging. gamma-GTP peaked in 40s in males and greater in males than in females. TP was greater in males. Age difference was dominant in ALB, CHOL and BUN. ALB decreased with aging, and CHOL and BUN increased with aging. Therefore standard value in each sex and age group should be adopted in these tests. The volume of the data in each age group in this study is suitable for the set-up of the standard value.