Assessing to the Impact of Differential Response Rates Across National Health and Nutrition Examination Survey (NHANES) Locations

• Activity Number: 311 - SPEED: Environment and Health, Governmental Policies and Population Surveys, Part 2
• Type: Contributed
• Date/Time: Tuesday, July 30, 2019 : 9:25 AM to 10:10 AM
• Sponsor: Government Statistics Section
• Author(s): Te-Ching Chen* and Jennifer Parker and Tala Fakhouri
• Companies: CDC/NCHS and CDC/NCHS/OAE/SPB and CDC/NCHS
• Keywords: NHANES; Response rate; Modeling; Localized response rate

Abstract

The National Health and Nutrition Examination Survey (NHANES) collects participants’ information through in-home screenings and interviews and physical examinations in mobile exam centers (MECs). To collect high quality data in a cost effective manner, each year NHANES surveys 5,000 participants in 15 locations (typically defined by county), with 6-8 weeks of operations at each location. To ensure representation of the U.S. noninstitutionalized civilian population, there are almost 90 sex-race/Hispanic origin-age income domains, some of which are oversampled. An assumption behind the NHANES sample design is response rates are consistent across locations. However, differential response rates across locations, especially among certain subgroups of the population, have been observed. Using NHANES 2011-2016 data, this study investigates the relationship of the response rates and the study locations’ demographic and geographic characteristics. Further, we model the relationship between localized response rate and selected health outcome, controlling for locations’ demographic and geographic distributions.