TY - JOUR
T1 - Geographic Variation in Heart Failure Mortality and Its Association With Hypertension, Diabetes, and Behavioral-Related Risk Factors in 1,723 Counties of the United States
AU - Liu, Longjian
AU - Yin, Xiaoyan
AU - Chen, Ming
AU - Jia, Hong
AU - Eisen, Howard J.
AU - Hofman, Albert
N1 - Funding Information:
This study was prepared using Data from the U.S. Centers for Disease Control and Prevention (CDC) surveillance and health survey systems, and the National Center for Health Statistics (NCHS). The opinions expressed in this paper are those of the authors and do not necessarily reflect the views of CDC and NCHS. The authors thank Miss Sarukkalige Shanika De Silva, BS and Miss Yuna Kim, BS, who are MS students from Drexel University Dornsife School of Public Health, for their help with preliminary data collection and analysis.
Publisher Copyright:
© Copyright © 2018 Liu, Yin, Chen, Jia, Eisen and Hofman.
PY - 2018/5/7
Y1 - 2018/5/7
N2 - Background and objectives: Studies that examined geographic variation in heart failure (HF) and its association with risk factors at county and state levels were limited. This study aimed to test a hypothesis that HF mortality is disproportionately distributed across the United States, and this variation is significantly associated with the county- and state-level prevalence of high blood pressure (HBP), diabetes, obesity and physical inactivity. Methods: Data from 1,723 counties in 51 states (including District of Columbia as a state) on the age-adjusted prevalence of obesity, physical inactivity, HBP and diabetes in 2010, and age-adjusted HF mortality in 2013–2015 are examined. Geographic variations in risk factors and HF mortality are analyzed using spatial autocorrelation analysis and mapped using Geographic Information System techniques. The associations between county-level HF mortality and risk factors (level 1) are examined using multilevel hierarchical regression models, taking into consideration of their variations accounted for by states (level 2). Results: There are significant variations in HF mortality, ranging from the lowest 11.7 (the state of Vermont) to highest 85.0 (Mississippi) per 100,000 population among the 51 states. Age-adjusted prevalence of obesity, physical inactivity, HBP, and diabetes are positively and significantly associated with HF mortality. Multilevel analysis indicates that county-level HF mortality rates remain significantly associated with diabetes (β = 2.7, 95% CI: 1.7–3.7, p < 0.0001), HBP (β = 3.6, 2.1–5.0, p < 0.0001), obesity (β = 0.9, 0.6–1.3, p < 0.0001), and physical inactivity (β = 1.2, 0.8–1.5, p < 0.0001) after controlling for gender, race/ethnicity, and poverty index. After further controlling obesity and physical inactivity in diabetes and HBP models, the effects of diabetes (β = 1.0, −0.3 to 2.3, p = 0.12) and HBP (β = 2.4, 0.9–3.9, p = 0.003) on HF mortality had a considerable reduction. Conclusion: HF mortality disproportionately affects the counties and states across the nation. The geographic variations in HF morality are significantly explained by the variations in the prevalence of obesity, physical inactivity, diabetes, and HBP.
AB - Background and objectives: Studies that examined geographic variation in heart failure (HF) and its association with risk factors at county and state levels were limited. This study aimed to test a hypothesis that HF mortality is disproportionately distributed across the United States, and this variation is significantly associated with the county- and state-level prevalence of high blood pressure (HBP), diabetes, obesity and physical inactivity. Methods: Data from 1,723 counties in 51 states (including District of Columbia as a state) on the age-adjusted prevalence of obesity, physical inactivity, HBP and diabetes in 2010, and age-adjusted HF mortality in 2013–2015 are examined. Geographic variations in risk factors and HF mortality are analyzed using spatial autocorrelation analysis and mapped using Geographic Information System techniques. The associations between county-level HF mortality and risk factors (level 1) are examined using multilevel hierarchical regression models, taking into consideration of their variations accounted for by states (level 2). Results: There are significant variations in HF mortality, ranging from the lowest 11.7 (the state of Vermont) to highest 85.0 (Mississippi) per 100,000 population among the 51 states. Age-adjusted prevalence of obesity, physical inactivity, HBP, and diabetes are positively and significantly associated with HF mortality. Multilevel analysis indicates that county-level HF mortality rates remain significantly associated with diabetes (β = 2.7, 95% CI: 1.7–3.7, p < 0.0001), HBP (β = 3.6, 2.1–5.0, p < 0.0001), obesity (β = 0.9, 0.6–1.3, p < 0.0001), and physical inactivity (β = 1.2, 0.8–1.5, p < 0.0001) after controlling for gender, race/ethnicity, and poverty index. After further controlling obesity and physical inactivity in diabetes and HBP models, the effects of diabetes (β = 1.0, −0.3 to 2.3, p = 0.12) and HBP (β = 2.4, 0.9–3.9, p = 0.003) on HF mortality had a considerable reduction. Conclusion: HF mortality disproportionately affects the counties and states across the nation. The geographic variations in HF morality are significantly explained by the variations in the prevalence of obesity, physical inactivity, diabetes, and HBP.
UR - http://www.scopus.com/inward/record.url?scp=85069835941&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85069835941&partnerID=8YFLogxK
U2 - 10.3389/fpubh.2018.00132
DO - 10.3389/fpubh.2018.00132
M3 - Article
AN - SCOPUS:85069835941
SN - 2296-2565
VL - 6
JO - Frontiers in Public Health
JF - Frontiers in Public Health
M1 - 132
ER -