Comparison of four enzyme immunoassays for detection of immunoglobulin M antibodies against rubella virus

We evaluated four tests for the detection of rubella virus-specific immunoglobulin M antibodies. Primarily, consecutive serum samples were tested by two different assays. Selected panels of sera from patients with proven or likely recent rubella and false-positive and true-negative results in the two primary assays were further tested with two recently developed, fully automated techniques. The four tests were comparable in overall accuracy, but their dynamic ranges may differ considerably. Ways to optimize the predictive values are discussed. We conclude that automated assays may be used without causing significant changes in diagnostic accuracy or distortions in notifications of the incidence of rubella compared with the use of established tools.