Optimal pooled testing

Abstract

This paper first characterizes three pooled testing techniques for generic testing conditions. It then applies these methods to demonstrate potential costs savings for universal HIV screening in the United Stated and Thailand. The success of these techniques in general is shown to be dependent on the prevalence as well as the disparity in prevalence between high and low risk groups. Further limitations on effectiveness include the potential impact of dilution on sensitivity. This and other limitations are addressed in the second section which focuses on applications of the techniques. Cost reductions are demonstrated to be feasible at prevalence of up to 30% and increasing dramatically at lower rates. Applying optimal pooled testing to universal HIV screening would result in an annual savings of over $1.4 billion in the United States and over $130 million in Thailand. Recent calls for universal HIV testing are based on the cost effectiveness of routine screening. These assessments may substantially overstate the cost of routine screening which strengthens the argument for universal screening.