24.12 THIS IS HOW WE DO IT: Would you like your receipt? (Maybe not.)

Humans are masters of altering environments and creating goods to make life safer and easier. But as we create more chemicals—often in pursuit of these goals—we must take care to monitor and evaluate their effects. Exposure to some of these chemicals, such as DDT, PCBs, phthalates, and TBTs, can disrupt normal hormone functioning and have harmful consequences for a wide variety of organisms—including ourselves.

Concerns over one such endocrine disruptor, bisphenol A (BPA), which can mimic human estrogen, have led to a ban on its use in reusable food and beverage containers in some U.S. states, and others are considering similar bans. In several hundred studies, including more than 50 in humans, exposure to concentrations of BPA that people might encounter in everyday life has been associated with reduced reproductive functioning in adults and impaired brain development in children exposed to BPA during gestation.

BPA is commonly used to line soup cans, plastic bottles, and some food containers because of its ability to keep the contents fresh. And BPA typically gets into our bodies when we eat canned food. But it is found in a huge number of other products as well; each year, more than eight billion pounds of BPA are produced worldwide.

In fact, BPA shows up in one item almost everyone in developed countries has contact with: printed receipts. More than half of all thermal paper—which is used for printing receipts from cash registers and ATMs—contains a layer of BPA that produces color when it is heated. Could BPA get into our bodies just from coming in contact with our skin?

Here’s one approach to answering that question. Researchers measured BPA levels in 389 pregnant women, and those who routinely handled receipts had higher-than-average levels of BPA in their bodies. And while the average concentration of BPA in urine was 1.9 micrograms per liter, or μg/L (SE = 1.0; SE is a measure of variation across study participants), the women who worked as cashiers had the highest concentrations of BPA, at 2.8 μg/L (SE = 1.1). Teachers and industrial workers, for comparison, had levels of 1.2 μg/L and 1.8 μg/L, respectively.

These results are consistent with the hypothesis that contact with receipts leads to higher levels of BPA in the body. But are there viable alternative hypotheses for these observations? Absolutely. Here’s one: The population of people who become cashiers may differ from the population of those who have other occupations in ways that—other than receipt handling—increase their exposure to BPA.

Some researchers set about testing this idea, reporting their findings in a 2014 article published in the Journal of the American Medical Association. The researchers first measured the BPA levels in the urine of 24 volunteers. Then the volunteers handled receipts continuously for 2 hours, while not wearing gloves, and their BPA levels were measured again. As a control, researchers conducted the procedure a week later, but with one change: participants wore gloves. Again, BPA levels were measured before and after 2 hours of handling receipts.

Here’s what they found:

In the group handling receipts without gloves, BPA levels were more than three times higher than for the group wearing gloves. (When researchers re-measured 12 of the non-glove participants after 8 hours, the BPA levels were five times higher!) Based on these results, the researchers concluded that BPA can and does pass through our skin and get into our bodies after handling receipts. Does the “control” procedure help rule out other means by which BPA on receipts might get into the body? How?

These findings don’t prove that BPA in receipts is harmful to our health. Remember, with careful scientific thinking, our goal is not so much about discovering “yes or no” answers about the natural world. Rather, we strive to increase our confidence about our hypotheses.

In this case, it’s reasonable to have high confidence that BPA can get into the body as a consequence of handling receipts on thermal paper. However, the researchers did not measure any health outcomes. (And, in fact, the increased BPA levels from receipt handling were relatively low. In another study, BPA levels after eating canned soup were considerably higher: 20.8 μg/L.)

First, we should answer this question: What experiment would generate the appropriate evidence for testing the hypothesis that “BPA from handling receipts carries a health risk”? You could recruit large numbers of individuals for a randomized, controlled, double-blind study in which the treatment group would be exposed to BPA for a long period of time so that researchers could detect any increase in negative health outcomes. Can you see any obstacles, ethical or otherwise, to such a study? It’s clear that the most logical or powerful experimental approach for testing a hypothesis may not be possible or desirable. This helps illuminate the value of smaller-scale, indirect approaches to investigating phenomena.

The researchers in the receipt-handling study, as well as other experts, conclude that people may be exposed to much higher levels of BPA than previously thought. They suspect, however, that most people’s exposure to BPA in thermal paper may not be high enough to have significant adverse effects on health. Still, they suggest that people in certain occupations, such as cashiers and bank tellers, who may handle receipts for 40 hours a week, as well as pregnant women, might have more cause for concern.