Author: Benmei Liu (Division of Cancer Control and Population Sciences, National Cancer Institute, MD, USA)
National health surveys, such as the National Health Interview Survey (NHIS), the Behavioral Risk Factor Surveillance System (BRFSS), and the Tobacco Use Supplement to the Current Population Survey (TUS-CPS), have been used to collect data on cancer screening and smoking related measures in the U.S. noninstitutionalized population. These surveys are designed to produce reliable estimates for the national and/or state level. However, policy makers, cancer control planners and researchers often need county level data for cancer surveillance and related research. In such case, model-based small area estimation (SAE) techniques have to be used to provide estimates with adequate precision. This study reviews several SAE research projects conducted at the National Cancer Institute (NCI). In all projects, Bayesian methods are developed to combine information from one or two national surveys and the relevant sources such as census or administrative records and generate estimates with increased precision. The model-based SAE techniques represent an effective means of generating estimates where there is small (or zero) state or county sample. The SAE results, which are released and disseminated at several NCI’s websites including the state cancer profiles website and the Surveillance, Epidemiology, and End Results (SEER) data base, provide a useful resource for the broad cancer surveillance society to fulfill multiple needs.