Model-Based Point Estimates and Variance for State-Level Prevalence of Hypertension and Uncontrolled Hypertension in the United States Using NHANES and BRFSS

• Activity Number: 122
• Type: Contributed
• Date/Time: Monday, August 1, 2016 : 8:30 AM to 10:20 AM
• Sponsor: Government Statistics Section
• Author(s): Soyoun Park* and Amy L. Valderrama and Jason Baumgardner and Cathleen Gillespie and Quanhe Yang and Jing Fang and Fleetwood Loustalot and Yuling Hong
• Companies: CDC/NCCDPHP/DHDSP and CDC/OPHPR/DSLR and CDC/NCIPC/DVP and CDC/NCCDPHP/DHDSP and CDC/NCCDPHP/DHDSP and CDC/NCCDPHP/DHDSP and CDC/NCCDPHP/DHDSP and CDC/NCCDPHP/DHDSP
• Keywords: State-level ; hypertension ; uncontrolled hypertension ; prevalence ; variance

Abstract

Hypertension (HTN) affects about 1 of 3 adults; only about ½ of them have their blood pressure (BP) controlled. HTN and uncontrolled BP prevalence in the US are based on measured BP from the National Health and Nutrition Examination Survey (NHANES). For smaller geographic regions, the data to estimate uncontrolled BP are not consistently available. Using the Behavioral Risk Factor Surveillance System (BRFSS), state-level self-reported HTN can be estimated; however, BRFSS lacks the data to estimate uncontrolled BP. To compensate, studies have used regression models in NHANES to predict national prevalence and applied the fitted models to BRFSS to generate sub-national estimates. We updated the state-level prevalence of HTN and uncontrolled BP in 2011-13 and expanded the methods to account for the error in both surveys and the regression models, to produce 95% confidence intervals. In 2011-13, the median prevalence of HTN and uncontrolled BP across states were 32.3% (31.3-33.3) and 47.0% (43.0-51.0), respectively. Sub-national data on measured health conditions is lacking. Modeled estimates from existing sources can guide programs and help identify disparate groups and regions.