POOLED TESTING FOR HIV PREVALENCE ESTIMATION - EXPLOITING THE DILUTION EFFECT

We study pooled (or group) testing as a method for estimating the prev alence of HIV; rather than testing each sample individually, this meth od combines various samples into a pool and then tests the pool. Exist ing pooled testing procedures estimate the prevalence using dichotomou s test outcomes. However, HIV test outcomes are inherently continuous, and their dichotomization may eliminate useful information. To overco me this problem, we develop a parametric procedure that utilizes the c ontinuous outcomes. This procedure employs a hierarchical pooling mode l and estimates the prevalence using the likelihood equation. The like lihood equation is solved using an iterative algorithm, and a simulati on study shows that our procedure yields very accurate estimates at a fraction of the cost of existing procedures. (C) 1998 John Wiley & Son s, Ltd.