Variance components of an enzyme-linked immunosorbent assay for detection of IgG antibodies in milk samples to Mycobacterium avium subspecies paratuberculosis in dairy cattle.

Milk samples from 120 cows were tested up to 10 times in an enzyme-linked immunosorbent assay (ELISA) for detection of antibodies to Mycobacterium avium ssp. paratuberculosis. The purpose of the study was to estimate variance components of the assay attributable to laboratory factors using mixed model theory. Because of significant interaction between the between-run, between-day and between-plate variables, the ELISA-plate variable was nested in run-number and run-number was nested in day-number. The nested variable accounted for 68% of the laboratory variability (P < 0.001), whereas the intraplate variability accounted for only 0.04% of the laboratory variability (P > 0.9). Therefore, it was concluded that the intraplate variability could be ignored whereas the variability from the combined run-day-plate variable should be considered in any analyses based on the ELISA.