Geographic Variation in Health Indicators across North Carolina Counties and Regions: A Comparison of Direct and Synthetic Estimates from the North Carolina BRFSS: (323702004-001)

Content Provider	Semantic Scholar
Author	Gizlice, Ziya Herrick, Harry Fultz-Butts, Kristi
Copyright Year	2003
Abstract	Objectives: The objectives of this study are: 1) to investigate if county and regional differences in selected health indicators are statistically significant before and after age adjustment and 2) to compare these crude and age-adjusted direct estimates with synthetic estimates (county-level estimates based on statewide rates). Methods: Survey data from the 2001 North Carolina Behavioral Risk Factor Surveillance System (BRFSS) were used for this study. The BRFSS is a random-digit-dialed telephone survey of non-institutionalized adults ages 18 and older. In 2001, the North Carolina BRFSS sampling frame consisted of the 10 most populated counties and three regions that contained the remaining 90 counties. Three domains of health were compared: chronic diseases, high-risk behaviors, and access to health care. Within each domain, four key indicators were assessed across the 10 counties and three regions of North Carolina. All analyses were performed with the SUDAAN software, designed for the analysis of complex sample designs such as the BRFSS survey. Results: Based on chi square tests, crude rates differed significantly across the 10 counties and three regions for all but two health outcomes, diabetes and current asthma. Similarly, age-adjusted rates also differed significantly across the 10 counties and three regions for all health indicators except current asthma. The Eastern North Carolina region had either the highest rate or a rate not significantly different from the highest rate for most health indicators. The synthetic estimates differed substantially from both the crude and age-adjusted rates and there was much less variation among the synthetic estimates. The synthetic estimates were overestimates in counties where direct survey estimates were significantly lower than the state rate, and were underestimates in counties where direct survey estimates were significantly higher than the state rate. Conclusions: Health indicators vary significantly across North Carolina. A county should generally use crude rates for their own public health planning purposes and age-adjusted rates to compare with other counties and regions on the relative level of health risks. An underlying factor contributing to the geographical health disparities in North Carolina is access to health care, which is strongly correlated with economic conditions. The use of synthetic estimates does not capture the variation across counties in health risks and behaviors and can be misleading. Our study provides strong evidence for the importance of direct local-level sampling to detect sub-state geographical variation in health indicators.
File Format	PDF HTM / HTML
DOI	10.1037/e323702004-001
Alternate Webpage(s)	http://www.schs.state.nc.us/SCHS/pdf/SCHS137.pdf
Language	English
Access Restriction	Open
Content Type	Text
Resource Type	Article