SMRFIT: a Statistical Analysis System (SAS) program for standardized mortality ratio analyses and Poisson regression model fits in community disease cluster investigations

Abstract

Increasingly, health departments are being pressed by the public to respond to disease risk with cluster investigations in communities and neighborhoods. This is a direct result of growing concern about the role that the environment may play in disease risk. While extensive analyses directly inputing exposures or numbers at risk are often necessary to thoroughly investigate clusters, it is quite useful to perform an exploratory analysis with existing morbidity and mortality data as a first level of response. To meet this need for timely evaluation, the authors describe a user-friendly Statistical Analysis System (SAS) program called SMRFIT to automate community disease cluster evaluations. The program creates frequency tables for number at risk and number of disease outcomes for the community, balance of parent county, and balance of state. SMRFIT then constructs standardized mortality ratios, with the community compared with balance of county and balance of state referents. Poisson regression is offered as an option for the modeling of community disease rates.