An approach to evaluating the reliability of diagnostic tests on pooled groups of infected individuals

Abstract

An experimental design and statistical analysis providing information on the reliability of pooled test procedures is described. It involves estimating the relationship between the probability of a positive pooled test result (dependent variable) and the expected number of infected individuals in a pool (explanatory variable). The intercept is an estimate of the proportion of false positives (1−pooled specificity) and pooled sensitivities can be estimated for indicative prevalences of infected individuals. Simulations for a theoretical infection are used to investigate the advantages and limitations of the approach. The approach is used to evaluate the reliability of a virus isolation and qRT-PCR test procedure detecting Salmonid alphavirus the pathogenic agent necessary for the development of Pancreas Disease in Atlantic salmon (Salmo salar).