Generalized Scan Statistics for Disease Surveillance

In applying scan statistics for disease surveillance, it would be valuable to have an integrated model that simultaneously includes environmental covariates and spatial correlation. In this paper, a generalized scan statistics under quasi‐likelihood functions is proposed to address this issue. We use a two‐step estimation process to obtain estimates of coefficients and adapt a bootstrapping method for the minimal p ‐value to address the multiple‐testing problem. Under suitable conditions, the proposed method is consistent and can control the type I error rate. Simulations and applications to real data sets are used to evaluate the method.