The relationship between polyfluorooctanoic acid and overall oral health
MetadataShow full item record
Polyfluorooctanoic acid (PFOA) is a known endocrine disruptor and is associated with several diseases. Because of the health effects of PFOA on other organs, it was hypothesized that the presence of PFOA might be directly, through endocrine disruption, or indirectly, through other diseases or their treatments, associated with overall oral health. This cross-sectional study used data from the 2007-2008 and 2009-2010 National Health and Nutrition Examination Surveys (NHANES) to test this hypothesis. After restricting the dataset to those with both the PFOA serum measurement and the answered oral health question, the sample size for this study was 3,233 individuals. The oral health data is self-reported and comes from one question on the oral health questionnaire, while the serum PFOA concentration comes from the NHANES blood test given during the physical examination. Descriptive and univariate comparisons were analyzed using Excel and SAS to describe and compare PFOA and oral health, along with data on other demographic, health, and social factors that could impact the relationship between oral health and serum PFOA concentration. Additional analyses were performed in SAS V 9.3 including stratification of the PFOA-oral health relationship, and multivariable logistic regression analyses to better assess the effects of confounding and effect modification on the primary hypothesis variables. Results show that above average PFOA subjects were more likely to be males, aged 50 years and older, Whites, in the top 25% of incomes, college graduates, and those who drink alcohol frequently or heavily. Oral health, body mass index (BMI), cholesterol level, diabetes status, and smoking habits, were not statistically associated with serum PFOA concentration. Results also found that individuals with better oral health were more likely to be Whites, aged 20-40 years or 60-69 years, with healthy BMIs, non-diabetics, college graduates in the top 25% of incomes, alcohol drinkers less than 5 times per week, and are not binge drinkers or current smokers. Since other factors might affect the relationship between oral health and PFOA concentration, the stratified results were performed to reveal effect modification and confounding factors. Males, certain age groups, BMI, frequency of alcohol consumption, and smoking habits are all effect modifiers for the relationship between serum PFOA concentration and oral health. This suggests that there could be other biological factors aside from PFOA concentration affecting the oral health outcomes. Finally, a logistic regression model controlling for multiple potential confounders supports the stratified results, concluding that PFOA concentration is not related to self-reported oral health in a statistically significant relationship. In conclusion, based on the consistency of the evidence across the analyses, this study found no statistically significant relationship between serum PFOA concentration and self-reported overall oral health. To verify this conclusion, future studies with more defined oral health measures, rather than the self-reported oral health ratings provided by the NHANES would be useful. These studies should also explore the effect modifiers to determine if there truly are other biological factors affecting the relationship between serum PFOA concentration and self-reported oral health. Finally, a study measuring the temporality between exposure and disease is needed to determine cause and effect between PFOA concentration and oral health relationships. Specifically the study would need to include timing of the oral health outcomes in relation to PFOA concentration to determine whether PFOA concentration changes precede oral health changes (and not the reverse).