Study scorecard: High blood pressure from BPA in canned drinks?

In 2009, the National Institute of Environmental Health Studies (NIEHS), which has invested more than $170 million on BPA-related research, including approximately $30 million in government stimulus funding, determined that much of that research was not relevant to the regulatory agencies charged with monitoring the chemicals in food packaging.

In the interest of deploying its tax dollars more effectively, the NIEHS created a set of criteria to identify studies that provided the greatest impact for assessing human health effects. These criteria range from the types of animals studied (some species are less sensitive to estrogen), to what they were given (some diets interfere with sensitivity to BPA), to how they were dosed (the chemicals need to be delivered orally, or in a way that mimics oral delivery), to the way the results are described (chemical effects must be linked to the exposure).

Using these criteria as a guide, the North American Metal Packaging Alliance Inc. (NAMPA) regularly assesses how BPA studies of interest measure up to NIEHS standards.

At Can Science, we plan to regularly post NAMPA’s study scorecards in the hope of helping readers who are interested in evolving science engage in the process of asking questions and weighing evidence before simply accepting all scientific claims as BPA facts.

Here, NAMPA examines a recent study published in Hypertension magazine and conducted by the department of preventative medicine at Seoul National University’s College of Medicine.

BPA study report card

The criteria for this report card were set by NIEHS for studies it was funding. In 2010, NIEHS stated that it had spent millions of taxpayer dollars on research that had no relevance to regulatory agencies that regulate BPA in food packaging.

Key: P = Meets criteria, C = Partially meets criteria, F = Fails criteria

Criteria Grade Comments
Diet must not interfere with the sensitivity of the model to BPA F Soy milk is high in phytoestrogens
Species and strain of animals (must be sensitive to estrogenic chemicals at low doses) P Humans
Sufficient sample size 60 individuals tested
Internal dose: Total and free BPA (1) should be measured in blood and if possible also in urine throughout the study F No blood or urine sampling of free vs. total BPA sampling
Dose responses (single-dose experiments are not acceptable) F One dose of canned soy milk
Result: Endpoint must be an actual phenotype (2), disease/dysfunction not just toxicity F Showed only a short-term increase in blood pressure
Entire litter (3) unit for developmental exposures must be used as statistical Not applicable
Route of exposure should be oral or justified to provide similar blood levels as oral route P Oral exposure
Male and female subjects should be used when feasible F 56 of 60 participants were female
Molecular targets and mechanisms should be assessed when possible, including gene expression, receptor binding and epigenetic studies. These effects should be linked to the exposure and the disease/dysfunction endpoints. F Very poor statistical relation to actual effect described; study used average blood pressure of control and test groups, while range of blood pressure value showed significant overlap between groups, indicating no real difference

Study: Exposure to Bisphenol A From Drinking Canned Beverage Increases Blood Pressure; authors: Sanghyuk Bae and Yun-Chul Hong


1. Free BPA is BPA found in the bloodstream that has not been metabolized. Total BPA represents all the BPA in a subject’s system, including free BPA and BPA that has been metabolized. Generally, chemical effects on the body occur before the chemicals are metabolized and passed out of the body.

2. A phenotype is simply an outward, observable manifestation of a trait.

3. Developmental toxicity studies that use animals should dose the mother and test the entire litter of offspring, never a sampling.