Premium
Varicella infection is not associated with increasing prevalence of eczema: a U.S. population‐based study
Author(s) -
Li J.C.,
Silverberg J.I.
Publication year - 2015
Publication title -
british journal of dermatology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.304
H-Index - 179
eISSN - 1365-2133
pISSN - 0007-0963
DOI - 10.1111/bjd.14028
Subject(s) - chickenpox , medicine , chickenpox vaccine , confidence interval , odds ratio , logistic regression , demography , vaccination , multivariate analysis , pediatrics , population , immunology , varicella vaccine , environmental health , immunization , virus , antigen , sociology
Summary Background Chickenpox infection early in childhood has previously been shown to protect against the development of childhood eczema in line with the hygiene hypothesis. In 1995, the American Academy of Pediatrics recommended routine vaccination against varicella zoster virus in the United States. Subsequently, rates of chickenpox infection have dramatically decreased in childhood. Objectives We sought to understand the impact of declining rates of chickenpox infection on the prevalence of eczema. Methods We analysed data from 207 007 children in the 1997–2013 National Health Interview Survey. One‐year prevalence of eczema and ‘ever had’ history of chickenpox were analysed. Associations between chickenpox infection and eczema were tested using survey‐weighted logistic regression. The impact of chickenpox on trends of eczema prevalence was tested using survey logistic regression and generalized linear models. Results Children with a history of chickenpox compared with those without chickenpox had a lower prevalence [survey‐weighted logistic regression (95% confidence interval, CI )] of eczema [8·8% (8·5–9·0%) vs. 10·6% (10·4–10·8%)]. In pooled multivariate models controlling for age, sex, race/ethnicity, household income, highest level of household education, insurance coverage, U.S. birthplace and family size, eczema was inversely associated with chickenpox [adjusted odds ratio (95% CI ), 0·90 (0·86–0·94), P < 0·001]. The prevalence of eczema significantly increased over time (Tukey post‐hoc test, P < 0·001 for comparisons of survey years 2001–13 vs. 1997–2000, 2008–13 vs. 2001–04 and 2008–13 vs. 2005–07). In multivariate generalized linear models, the odds of eczema was not associated with chickenpox in 2001–13 ( P ≥ 0·06). Conclusions These findings suggest that lower rates of chickenpox infection secondary to widespread vaccination against varicella zoster virus are not contributing to higher rates of childhood eczema in the U.S.