Tobacco smoke exposure as a risk factor for human papillomavirus infections in women 18-26 years old in the United States

Abstract

Background Although tobacco smoke has been associated with many infections, little is known of its association with human papillomavirus (HPV) infections among young adult women. The aim of the study was to explore the association of tobacco smoke exposure on HPV infections in young adult women in the United States. It was hypothesized that tobacco smoke exposure (both active and passive) as objectively measured by cotinine levels was associated with increased HPV infection in a national sample of 18 and 26 year-old women in the United States. Study methods and findings The NHANES 2007-2012 dataset was used in the analyses. A national representative sample of women 18 to 26 year old (N = 1,414) was included in the study. Infection with any HPV was determined by PCR while tobacco smoke exposure was determined by measuring serum cotinine levels. Women with cotinine levels <0.05 ng/mL were considered unexposed and those with levels > = 0.05 were considered as exposed. Exposed women were further categorized as passive smokers (cotinine levels 0.05-<10 ng/mL, while active smokers were those with cotinine levels = > 10ng/mL). Data were analyzed by descriptive statistics and multiple logistic regression analysis. Exposed women (passive smokers with cotinine levels > = 0.05ng/mL-10ng/mL) were almost 2 times (64% vs 35%) more likely to be infected with any HPV type than their unexposed counterparts (cotinine levels <0.05ng/mL). Also women who were active smokers (cotinine levels > = 10 ng/mL) were more than twice more likely (75%) to be infected with the virus than the unexposed counterparts. The relationship held true even after controlling for various socio-demographics. Indeed in the multiple regression analyses controlling for the various confounders, compared to their unexposed counterparts, women exposed to second hand smoke were more than twice more likely to have HPV infections (OR: 2.45; 95% C.I = 1.34-4.48). When compared to their unexposed counterparts, actively smoking women were more than 3.5 times more likely to be infected with HPV (OR = 3.56; 95% CI 1.23-10.30). Finally, a strong dose-response relationship was further demonstrated with increasing risk of HPV with each quartile of cotinine levels even after controlling for various confounders. The risk of HPV was lowest in the lowest quartile (Referent OR = 1) and increased steadily with each quartile of cotinine levels until the risk was highest among women with cotinine levels in the 4th quartile (OR = 4.16; 95% C.I. = 1.36-12.67).