American Journal of Clinical Nutrition, Vol 15, 149-157, Copyright © 1964 by The American Society for Clinical Nutrition, Inc.

Sources of Variation in Repeated Measurement of Plasma Free Fatty Acids in Rested Fasting Subjects

¹ From the Veterans Administration Hospital, Palo Alto, California, and the Research Computational Center, University of Texas Medical Branch, Galveston, Texas

Plasma free fatty acids (FFA) were measured repetitively in different samples of rested, fasting subjects during six separate clinical studies. Statistical analysis of the sources of variance in these repeated measurements of FFA revealed the greatest contribution to be from intraindividual variation. Mean variance from this source averaged 62.5 per cent for the series of six studies, ranging between 39.7 and 75.9 per cent in individual studies. The remainder of the variance was attributable to differences between subjects; variance due to technical error was minimal. Mean coefficients of variation ranged between 18.6 per cent and 29.3 per cent, the usual trend being toward a reduction in variation with a decreased interval between measurements. Nonetheless, variation was high even when samples were taken at an interval of 1 hour (18.6 per cent) or under identical basal conditions 24 hours apart (23.9 per cent). This high degree of intraindividual variation of FFA levels must be considered in clinical studies using this measure, as many of the factors which affect FFA levels are beyond ordinary experimental control.