In a study published on August 16, 2016 in the peer-reviewed journal Environmental Health Perspectives, Radhika Dhingra and colleagues from Rollins School of Public health, Emory University, Atlanta, U.S. examine the influence of reverse causation bias on the outcomes of cross-sectional epidemiological studies with perfluorooctanoic acid (PFOA, CAS 335-67-1).
Earlier this year, the same authors reported that no associations between PFOA serum levels and either early menopause or chronic kidney disease could be observed in a longitudinal epidemiological study, although positive associations for both outcomes were found previously by cross-sectional studies. This discrepancy could be due to the bias introduced by reverse causation, the authors suggest, because PFOA levels in serum could be changing due to an altered excretion occurring in menopause or as a result of renal dysfunction.
In the present analysis, the associations with menopause or reduced kidney function were examined for both measured and modeled serum PFOA levels, the latter estimated retrospectively based on residential history and plant emissions. Menopause history was self-reported, and estimated glomerular filtration rate was measured in blood samples as a marker of kidney function. The authors reasoned that, in case of a true association existing between PFOA levels and either menopause onset or kidney function, it should be observed for both measured and modeled estimates (given that the modeled metric is correct). However, if no true association exists, measured PFOA could still show an association if PFOA excretion is influenced by the menstrual pattern or renal function.
Both menopause and the marker of kidney function were found to be significantly associated with measured but not modeled serum PFOA levels. Moreover, an increase in measured PFOA levels was observed in the same subjects for at least the first seven years after menopause, providing additional evidence that a reverse causation relationship exists between measured PFOA and menopause.
Dhingra and colleagues conclude that “in prior studies, early menopause and reduced kidney function are the causes rather than the results of increased measured serum PFOA.” Their study also demonstrates that the interpretation of cross-sectional studies relying on serum biomarkers of chemical exposure should be approached with caution. Since in cross-sectional studies both the biomarker and the health outcome are measured concurrently, the possibility of a reverse causation arises, particularly when the chemical’s pharmacokinetics can be impacted by the outcome of interest.
Due to its water-repellent and grease-proof properties, PFOA has been used in food packaging, such as popcorn bags, pizza boxes and fast food wrappers, as well as in a variety of other products (FPF reported). A systematic review study conducted in 2014 found sufficient evidence demonstrating adverse effects of PFOA on fetal growth. This compound has also been linked to immune disorders (FPF reported) and several cancers (FPF reported). Despite several initiatives to restrict the production and use of PFOA and some other per- and polyfluoroalkyl substances (PFASs), the ammonium salt of PFOA (APFO, CAS 3825-26-1) is still authorized for use in plastic food contact materials (FCMs) in Europe, and more than a dozen other PFASs are used in various FCMs in both Europe and the U.S. (see FPF dossier).
Dhingra, R., et al. (2016). “A study of reverse causation: Examining the associations of perfluorooctanoic acid serum levels with two outcomes.” Environmental Health Perspectives (published August 16, 2016).
Dhingra, R., et al. (2016). “Perfluorooctanoic acid exposure and natural menopause: A longitudinal study in a community cohort.” Environmental Research 146: 323-330.
Dhingra, R., et al. (2016). “Perfluorooctanoic acid and chronic kidney disease: Longitudinal analysis of a Mid-Ohio Valley community.” Environmental Research 145: 85-92.
Johnson, P., et al. (2014). “The Navigation Guide – evidence-based medicine meets environmental health: Systematic review of human evidence for PFOA effects on fetal growth.” Environmental Health Perspectives 122: 1028-1039.