Establishing the cut-off value of a screening method

Generally, the cut-off point of a screening assay is defined through a one-sided prediction limit obtained from a given sample of blanks. Depending on the assumptions one is willing to make about the underlying data distribution, different types of prediction limits can be employed. In this paper, Monte Carlo simulations are used to illustrate the coverage performance of normal, lognormal and nonparametric prediction limits under normal and non-normal conditions.