Defining the Rural HIV Epidemic: Correlations of 3 Definitions—South Carolina, 2005-2011
Version of Record online: 15 DEC 2013
© 2013 National Rural Health Association
The Journal of Rural Health
Volume 30, Issue 3, pages 275–283, Summer 2014
How to Cite
Weissman, S., Duffus, W. A., Vyavaharkar, M., Samantapudi, A. V., Shull, K. A., Stephens, T. G. and Chakraborty, H. (2014), Defining the Rural HIV Epidemic: Correlations of 3 Definitions—South Carolina, 2005-2011. The Journal of Rural Health, 30: 275–283. doi: 10.1111/jrh.12057
- Issue online: 1 JUL 2014
- Version of Record online: 15 DEC 2013
- rural definition
To gain a better understanding of the HIV epidemic in rural South Carolina (SC) by contrasting 3 definitions of rural and urban areas.
The sample included newly diagnosed HIV cases aged ≥18 years in SC between January 1, 2005, and December 31, 2011. Each individual was assigned a rural or urban status as defined by the Office of Management and Budget (OMB), Census Bureau (CB), and Rural Urban Commuting Area (RUCA) classifications. Descriptive statistics were conducted to compare sociodemographic characteristics, CD4 counts, viral loads, and time to AIDS diagnosis between rural and urban populations. Kappa statistics measured the agreement between the 3 definitions of rurality.
Depending on the definition used, the proportion of newly diagnosed HIV cases in rural areas varied from 23.3% to 32.0%. Based on the OMB and RUCA definitions, rural residents with HIV were more likely to be older, women, black, and non-Hispanic, report heterosexual contact, and have an AIDS diagnosis within 1 year of their HIV diagnosis. The OMB and RUCA definitions had a nearly perfect agreement (kappa = 0.8614; 95% CI = 0.8457, 0.8772), while poor agreements were noted between the OMB and CB or the RUCA and CB definitions.
When examining the rural HIV epidemic, how “rural” is defined matters. Using 3 definitions of rurality, statistically significant differences were found in demographic characteristics, timing of HIV diagnosis and the proportion of rural residents diagnosed with HIV in SC. The findings suggest possible misclassification biases that may adversely influence services and resource distribution.