ABSTRACT
Validation of new diagnostic assays requires the establishment of their performance characteristics such as diagnostic sensitivity and specificity, precision, repeatability, accuracy and reproducibility. These different stages of validation are described in the recent Standard Operating Procedure for OIE Validation and Certification of Diagnostic Assays. This report describes a reproducibility study of a new ELISA to titrate rabies antibodies in vaccinated wild and domestic carnivores. The study was modelled on the proficiency tests which are annually organised by the Community Reference Institute (Afssa Nancy, France) in the frame of international movements of pets. Analyses demonstrated that the five participants provided satisfactory repeatability estimates (variation coefficients generally below 15% for the 20 coded sera of the panel), and concordant status for all serums. A regression analysis performed on standard curves revealed that two different positive standards used in two dilution ranges were titrated similarly by all participants, and that no significant differences were observed by using these two standards. Titres obtained on a dilution range included in the panel demonstrated that all laboratories were consistent with themselves (significant correlation between experimental and theoretical results), and consistent with other laboratories (significant correlation between results of laboratory under test and mean results of all other laboratories).
