Surveillance studies on change point in incidence rate

Surveillance is a major issue in the present day world. The concern for developing adequate surveillance strategies is genuine in many spheres of human activities. In this dissertation we focus on some specific problems of surveillance and discuss some statistical issues related to them. They deal with a possible change in the incidence rate of an event [to a higher value] when a system is studied at discrete time points. We first propose an approach to estimate the product exposure within a time frame by taking the delay between shipping and application of the product, product effective exposure time into consideration. Then we apply the exponentially weighted moving average methods for detecting an increased incidence rate per exposure unit of an event by monitoring an overdispersed Poisson process. Different measures of evaluation and their analytical bounds, suitable in different types of applications, are studied to give a broad picture of the features of the methods. Results from simulations indicate that the analytical bounds perform fairly well when the weight parameter is large. When observing and monitoring the process in a moving window, the CUSUM method, the Shewhart method, the Shiryaev-Roberts method and the exponentially weighted moving average method are extended and compared to detect an increased incidence rate for an overdispersed Poisson process.