Biological and analytic components of variation in long-term studies of serum constituents in normal subjects. II. Estimating biological components of variation.

Standard statistical analysis of variance has been used on data from weekly determinations of 15 blood constituents in each of 68 normal subjects, our purpose being to isolate and estimate biological components of observed variation. These components include intra-individual, or “personal” variation, and inter-individual, or “group” variation. The most critical problem, that of separating personal variation from long-term analytic deviations in each individual’s data record, was resolved with the aid of concurrent analysis of a constant “pooled” serum. Comparison of isolated personal variation with inter-individual variation indicates that many common blood tests-e.g., magnesium, calcium, total protein, uric acid, serum enzymes, and cholesterol-could contribute to an individually distinctive blood “profile” if substantial improvements were made in analytic precision. Presently, only cholesterol (among older women) and uric acid (in males) appear to possess strong capabilities for distinguishing among normal individuals.