Varicella in Americans from NHANES III: implications for control through routine immunization.

At the time of varicella vaccine introduction in the United States, an estimated 4 million episodes of varicella occurred annually. This survey of varicella seroprevalence is the first to describe immunity to a vaccine-preventable disease prior to vaccine introduction in the United States population. The objective of this analysis is to describe patterns of naturally-acquired varicella and understand characteristics associated with infection in the varicella vaccine-naive United States population. A nationally representative cross-sectional health examination survey that included venipuncture was conducted among 21,288 U.S. participants aged 6 years and older from 1988 through 1994. Serologic evidence of varicella-zoster virus infection was measured by enzyme immunoassay of varicella-zoster virus-specific IgG antibody. The seroprevalence of IgG antibody to varicella-zoster virus increased from 86.0% in children aged 6 through 11 years to 99.6% in adults aged 40 through 49 years. Immunity to varicella remained at 99% or higher in Americans aged 50 years and older. Among persons aged 6 through 19 years, non-Hispanic black children were 40% less likely to be seropositive compared with white children (odds ratio [OR], 0.6; 95% confidence interval [CI], 0.4-0.8). Among young adults aged 20 through 39 years, women with a history of live birth (OR, 4.3; 95% CI, 2.1-8.7) and married men (OR, 2.7; 95% CI, 1.2-5.7) were more likely to have naturally-acquired immunity to varicella. This study found that, prior to use of varicella vaccine in the United States, age, race, and marital characteristics were independently associated with naturally acquired varicella. Future varicella serosurveys in Americans will provide essential information to interpret the population impact of varicella vaccine.