The group testing strategy, first introduced by Dorfman (1943) to test for the syphilis antigen in U.S. army recruits, tested blood samples from an assortment of different subjects. Those who tested negative were declared free of infection, whereas subjects who tested positive required retesting. This strategy, in its many subsequent variations, has been employed in both disease screening and prevalence estimating, and has proven to be cost-effective, especially when the disease prevalence is low. The current COVID-19 pandemic has further vitalized this method, which some countries have used to test for infection of the virus.
Zhang, W., Liu, A., Li, Q., & Albert, P. S. (2020). Nonparametric estimation of distributions and diagnostic accuracy based on group tested results with differential misclassification. Biometrics, 1-10. PMID: 32083733
Zhang, W., Liu, A., Li, Q., & Albert, P. S. (2020). Incorporating retesting outcomes for estimation of disease prevalence. Statistics in Medicine, 39(6), 687–697. PMID: 31758594
Li, Q., Liu, A., & Xiong, W. (2017). D-optimality of group testing for joint estimation of correlated rare diseases with misclassiﬁcation. Statistica Sinica, 27(2), 823-838.
Zhang, Z., Liu, C., Kim, S., & Liu, A. (2014). Prevalence estimation subject to misclassification: the mis-specification bias and some remedies. Statistics in Medicine, 33(25), 4482-4500. PMID: 25043925
Liu, C., Liu, A., Zhang, B., & Zhang, Z. (2013). Improved confidence intervals of a small probability from pooled testing with misclassification. Frontiers in Public Health, 1, 1-39. PMID: 24350208
Liu, A., Liu, C., Zhang, Z., & Albert, P. S. (2012). Optimality of group testing in the presence of misclassification. Biometrika, 99(1), 245–251. PMID: 23049137