The purpose of screening programs is to discover at an early point in time serious diseases, such as cervical cancer and colorectal cancer, that can be treated at lower social costs than would be the case if detected at later stages of disease development. In general, three questions are at the forefront of screening programs: Given a specific screening activity, first, what is the optimal age to start screening individuals, and, second, how often should the individuals be screened? Third, given budget constraints, how should resources be allocated and expended across different types of screening programs? In this article, my aim is to derive a mathematical model that can be used to answer these three questions. Of course, to implement the model, one needs empirical information, and so I attempt to characterize here as well the types of data needed to answer the questions. Overall, my presentation builds on the prior analyses and findings of Gyrd-Hansen, Søgaard, and Kronborg (1996) and Gyrd-Hansen (1997).
