Sex ratio, poverty, and concurrent partnerships among men and women in the U.S

Background: Social and economic contextual factors may promote concurrent sexual partnerships, which can speed population HIV/STI transmission and are more common among African Americans than among US Whites. Methods: We used the 2002 National Survey of Family Growth (NSFG) (N=12, 571) and its Contextual Database to examine concurrency prevalence during the preceding 12 months in relation to two features of the county of residence: the sex ratio (#men / #women) and the percentage in poverty (both within the respondent's racial/ethnic group. Analyses incorporated sample weights and accounted for the complex survey design. Results: Concurrency was more prevalent in (1) counties with low (<0.9) sex ratios compared to counties with balanced (0.95-1.05) sex ratios (OR 2.30; 1.75, 3.02) and (2) counties with higher poverty percentages (OR 2.0; 1.31, 3.06 for >21% vs 6.6%). These analyses could not be controlled for race/ethnicity because of the limited overlap in the racial/ethnic-specific distributions of sex ratio and poverty percentage: 83% (weighted) of Blacks lived in counties with sex ratios < 0.9, compared to 8% of Hispanics and 5% of Whites; more than three-quarters of Blacks, one-half of Hispanics and less than 5% of Whites lived in counties with poverty percentages above 21% for the respondent's racial/ethnic group. Conclusion: The differences in context of life between African Americans and other large racial/ethnic groups in the US are so stark as to preclude a direct test of whether sex ratios or community poverty can explain racial/ethnic differences in concurrent partnerships.