Seroprevalence of Epstein-Barr virus infection in U.S. children ages 6-19, 2003-2010.

backgroundEpstein-Barr virus (EBV) is a common herpesvirus linked to infectious mononucleosis and multiple cancers. There are no national estimates of EBV seroprevalence in the United States. Our objective was to estimate the overall prevalence and sociodemographic predictors of EBV among U.S. children and adolescents aged 6-19.

methodsWe calculated prevalence estimates and prevalence ratios for EBV seroprevalence using data from the 2003-2010 U.S. National Health and Nutrition Examination Survey (NHANES) for children aged 6-19 (n = 8417). Poisson regression was used to calculate multivariable-adjusted prevalence ratios across subgroup categories (sex, race/ethnicity, parental education, household income, household size, foreign-born, BMI, and household smoking).

findingsOverall EBV seroprevalence was 66.5% (95% CI 64.3%-68.7%.). Seroprevalence increased with age, ranging from 54.1% (95% CI 50.2%-57.9%) for 6-8 year olds to 82.9% (95% CI 80.0%-85.9%) for 18-19 year olds. Females had slightly higher seroprevalence (68.9%, 95% CI 66.3%-71.6%) compared to males (64.2%, 95% CI 61.7%-66.8%). Seroprevalence was substantially higher for Mexican-Americans (85.4%, 95% CI 83.1%-87.8%) and Non-Hispanic Blacks (83.1%, 95% CI 81.1%-85.1%) than Non-Hispanic Whites (56.9%, 95% CI 54.1%-59.8%). Large differences were also seen by family income, with children in the lowest income quartile having 81.0% (95% CI 77.6%-84.5%) seroprevalence compared to 53.9% (95% CI 50.5%-57.3%) in the highest income quartile, with similar results for parental education level. These results were not explained by household size, BMI, or parental smoking. Among those who were seropositive, EBV antibody titers were significantly higher for females, Non-Hispanic Blacks and Mexican-Americans, with no association found for socioeconomic factors.

conclusionsIn the first nationally representative U.S. estimates, we found substantial socioeconomic and race/ethnic differences in the seroprevalence of EBV across all ages for U.S. children and adolescents. These estimates can help researchers and clinicians identify groups most at risk, inform research on EBV-cancer etiology, and motivate potential vaccine development.