Most clinical chemical analytes vary in a random manner around a homeostatic set point. Replicate analyses of a series of specimens collected from a group of subjects allows estimation of analytical, within and between subject components of variation. The preferred experimental procedures and statistical methods for evaluation of data and analysis of variance are described; a detailed example is provided in the Appendix. The many uses of data on biological variation in clinical chemistry are reviewed, including setting analytical goals, deciding the significance of changes in serial results from an individual, evaluating the utility of conventional population-based reference values in patient management, and other applications.

No abstract

Intra- and interindividual components of biological variation have been determined for total thyroxin (TT4), free thyroxin (FT4), total triiodothyronine (TT3), free triiodothyronine (FT3), and thyrotropin (TSH). Calculated analytical goals (CV, %) for the precision required for optimal patient care are: TT4 less than or equal to 2.5, FT4 less than or equal to 4.7, TT3 less than or equal to 5.2, FT3 less than or equal to 3.9, and TSH less than or equal to 8.1. The marked degree of individuality demonstrated for all hormones indicates that, if conventional population-based reference ranges are used uncritically, major changes in hormone concentration may not be correctly identified for some patients because observed values continue to lie within the reference range. At analyte concentrations approximating the mean values found in this study, and for analytical performance meeting the appropriate analytical goal, the differences required for consecutive results to be significantly different (p less than or equal to 0.5) have been calculated as: TT4, 14.7 nmol/L; FT4, 5.7 pmol/L; TT3, 0.6 nmol/L; FT3, 1.3 pmol/L, and TSH, 0.7 milli-int. unit/L.

No abstract

Strategies abound for the setting of analytical goals in clinical chemistry. Many, especially those more recently proposed for particular clinical situations, are concerned with tests used in diagnosis. We suggest a general theory for the setting of goals in situations that specifically involve the monitoring of individuals. Goals are calculated from the formula CVA less than [(delta c 2/2Z2)-CVB2]1/2, where CVA is the analytical imprecision (as coefficient of variation, CV); delta c is the percentage change in serial results that is considered clinically significant; Z is the Z-statistic, which depends only on the probability selected for statistical significance; and CVB is the average inherent within-subject biological variation (as CV). Examples given show applications in hematology and in monitoring diabetes mellitus, chronic renal failure, and hepatitis. The derived goals are for total random analytical error (imprecision and intermittent systematic variation), and provide objective criteria that should be achieved in practice. The effect of analytical variability on both variability in test results and the probability that a stated change can be considered significant should be calculated whether or not the goals are attained.