The influences of sample distribution and age on reference intervals for adult males.

This investigation was designed to improve reference information and to evaluate the influences of sample distribution and age on the derived reference intervals. Specimens from 127 men were collected after a 12- to 14-hour fast and analyzed by 60 different laboratory procedures. Differences in the reference intervals derived, using three separate statistical methods, appeared to be unimportant clinically, but the percentile method was preferred because it required no assumptions concerning the frequency distribution. Relationships between age and analyte concentrations were examined by linear regression analysis, and the analytes were placed in one of three groups, according to the significance of this relationship: greatest significance (P less than or equal to 0.01), 18 analytes; intermediate significance (0.01 less than or equal to P less than or equal to 0.05), 12 analytes; and least significance (P greater than 0.05), 30 analytes. The age-related changes for each analyte were evaluated according to analytic variation, population dispersion, and clinical relevance. Age-dependent reference intervals for adult males are recommended for albumin, cholesterol, phosphorus, and sedimentation rate.