County-Level Estimates of U.S. Cancer Screening/Risk Factors: the National Health Interview Survey (NHIS) and the Behavioral Risk Factors Surveillance Survey (BRFSS), 2011–2016

• Activity Number: 184 - New Research in Small Area Estimation
• Type: Contributed
• Date/Time: Tuesday, August 4, 2020 : 10:00 AM to 2:00 PM
• Sponsor: Survey Research Methods Section
• Author(s): John Pleis* and Benmei Liu and Eric Feuer and Yulei He and Van Parsons and Bill Cai and Machell Town
• Companies: National Center for Health Statistics and National Cancer Institute and National Cancer Institute and National Center for Health Statistics and National Center for Health Statistics and National Center for Health Statistics and Center for Disease Control and Prevention
• Keywords: Small-area; NHIS; BRFSS; MCMC

Abstract

Small-area estimates for health-related outcomes (e.g., cancer screening) continue to be of great interest for public health researchers, cancer control planners and policy makers. The National Cancer Institutes, collaborating with the National Center of Health Statistics (NCHS), the Centers for Disease Control and Prevention (CDC) and academics, developed novel small area models that combine information from two large national surveys, the NHIS and the BRFSS, to generate estimates with increased precision for several cancer risk factors and screening behaviors for four consecutive data periods covering years 1997 to 2010. The NHIS data was stratified by household phone status. The modeling accounted for the difference between the estimates for the NHIS households with landline phones and the BRFSS. However, since 2011, BRFSS has changed its design by including cellular telephone–only households and using an improved weighting methodology to reduce potential bias. As a result, a modified methodology may be used for producing small-area estimates of 2011 and forward. This paper describes the updated methodology being developed and presents some of the results.