Cincinnati, Ohio - KidsChemicalSafety.org provides up-to-date health information on chemical hazards and safe use of chemicals around children, so we asked Dr. Michael Dourson, of Toxicology Excellence for Risk Assessment (TERA), to comment on a recent article, Eng et al. (2013) that found higher levels of bisphenol A were associated with several measures of obesity in children. Specifically, children exposed to higher levels of bisphenol A had increased odds of having a body mass index in the 95th percentile (i.e., greater than 95% of all children) and a waist circumference to height ratio (WC) greater than a value of 0.5. However, several other measures of obesity were unaffected, such as abnormal body fat percentage, BMI and WC thresholds below 95th percentile, and other laboratory measurements of cardiovascular disease and diabetes (e.g., total cholesterol, HDL, LDL, fasting triglycerides, insulin resistance, and fasting glucose).
So does this association mean that bisphenol A causes obesity?
The short answer is not definitively. Associations do not imply causality.
To give a simple example, the sale of chocolate ice cream is positively associated with the number of soccer games played outdoors. So does the sale of chocolate ice cream cause people to play soccer? No. Both the sale of chocolate ice cream and the playing of soccer games, however, are associated with higher outdoor temperatures. Higher temperature may, in fact, be causing both of these events.
Eng et al. (2013) give a number of important caveats with their research. In their own words:
"We also acknowledge limitations of our study. We used a cross-sectional study design; therefore, reverse causality cannot be excluded."
Reverse causality means that obesity may in fact be causing the increase in bisphenol A found in these children, perhaps by causing a slower elimination of bisphenol A. Eng et al. (2013) go on to say:
"It is unclear if a single measure of BPA [which they used] would be indicative of long-term exposure to BPA, as humans metabolize and excrete BPA relatively rapidly (half-life of 6 hours with nearly complete urinary excretion by 24 hours)."
The authors association is based on a single measurement of bisphenol A in each child. However, this single measurement only reflects the concentration of bisphenol A within a day or so. The authors must assume that this measurement reflects a much longer exposure to bisphenol A in order to be associated with obesity, since obesity and related complications take time to manifest. The authors did point to one study that indicates a single measurement is moderately sensitive (0.64) for predicting high BPA levels estimated by multiple measurements (i.e., 64% of the time a high measurement of BPA, as determined by multiple samples, will be predicted by a single high measurement), however a longitudinal study, one where multiple samples and measurements are taken over time, would be better able to determine causality between BPA and obesity.
Eng et al. (2013) go on to say:
"Another limitation [of our study] is the lack of dietary information included in the analyses. BPA exposure presumably can occur via oral ingestion; thus, it is possible that those who are obese are merely eating more BPA-containing food."
The authors are stating that their association of obesity with increase in bisphenol A concentrations may simply be due to obese people eating more food, specifically BPA-containing food. BPA is found in can linings and food packaging, which means people who consume processed and/or canned foods and sodas ingest more BPA than people who do not. Eng et al. (2013) did adjust their results for soda consumption, but this data was incomplete.
Eng et al. (2013) conclude that additional research is needed to elucidate the relationships between BPA exposure and the development of obesity and chronic disease risk factors in children in order to inform future policy regulating children’s consumer products. Based on a review of this paper, we agree that additional research is needed, specifically a study that is designed in a way that allows for determination of a causal link between BPA exposure and obesity, not just correlation.