James C Thomas1, Mary E Gaffield. 1. Department of Epidemiology, University of North Carolina, Chapel Hill, North Carolina 27599-7435, USA. jim.thomas@unc.edu
Abstract
OBJECTIVES: We sought to identify characteristics of counties in the southeastern United States associated with endemically high rates of gonorrhea. In particular, we were interested in aspects of race other than the proportion of Blacks in a population, including the potential influence of under-reporting of infections. DESIGN: The associations between the characteristics of counties in 1990, and the rates of reported gonorrhea from 1986 to 1995, were estimated with multivariable logistic regression. SETTING: 14 states in the southeastern region of the United States. PARTICIPANTS: 835 counties and county equivalents in the 14 southeastern states. MAIN OUTCOME MEASURES: The odds of having an endemically high county-level rate of gonorrhea. RESULTS: The variables with a strong effect on endemically high rates of gonorrhea included racial residential isolation in the absence of low income dualism (odds ratio [OR]: 210.84, 95% confidence interval [CI]: 19.35, 999.00), and Black-White income dualism in communities with few female-headed households (OR: 4.57, 95% CI: 2.68, 7.80). The percentage of Blacks in the population that was Black had little or no association with the rate of gonorrhea. These estimates were relatively robust when subjected to a sensitivity analysis of potential under-reporting of gonorrhea. CONCLUSIONS: Previous studies have demonstrated that a high percentage of Blacks in a population is the strongest predictor of high rates of gonorrhea. We found, however, that when variables measuring aspects of social structure, such as a race-based income distribution, and de facto residential segregation, were included in the model, the proportion of Blacks no longer had an effect on rates of gonorrhea. Progress in lowering endemically high disease rates will require attention being paid to community racial and class dynamics.
OBJECTIVES: We sought to identify characteristics of counties in the southeastern United States associated with endemically high rates of gonorrhea. In particular, we were interested in aspects of race other than the proportion of Blacks in a population, including the potential influence of under-reporting of infections. DESIGN: The associations between the characteristics of counties in 1990, and the rates of reported gonorrhea from 1986 to 1995, were estimated with multivariable logistic regression. SETTING: 14 states in the southeastern region of the United States. PARTICIPANTS: 835 counties and county equivalents in the 14 southeastern states. MAIN OUTCOME MEASURES: The odds of having an endemically high county-level rate of gonorrhea. RESULTS: The variables with a strong effect on endemically high rates of gonorrhea included racial residential isolation in the absence of low income dualism (odds ratio [OR]: 210.84, 95% confidence interval [CI]: 19.35, 999.00), and Black-White income dualism in communities with few female-headed households (OR: 4.57, 95% CI: 2.68, 7.80). The percentage of Blacks in the population that was Black had little or no association with the rate of gonorrhea. These estimates were relatively robust when subjected to a sensitivity analysis of potential under-reporting of gonorrhea. CONCLUSIONS: Previous studies have demonstrated that a high percentage of Blacks in a population is the strongest predictor of high rates of gonorrhea. We found, however, that when variables measuring aspects of social structure, such as a race-based income distribution, and de facto residential segregation, were included in the model, the proportion of Blacks no longer had an effect on rates of gonorrhea. Progress in lowering endemically high disease rates will require attention being paid to community racial and class dynamics.
Authors: Jacky M Jennings; Ralph B Taylor; Rama A Salhi; C Debra M Furr-Holden; Jonathan M Ellen Journal: Soc Sci Med Date: 2012-02-13 Impact factor: 4.634
Authors: Kwame Owusu-Edusei; Harrell W Chesson; Jami S Leichliter; Charlotte K Kent; Sevgi O Aral Journal: Am J Public Health Date: 2013-03-14 Impact factor: 9.308