A plastic ban for dummies

Patrick Basham and John Luik, co-authors of Diet Nation: Exposing the Obesity Crusade, say health chiefs are falling for poisonous stories about a chemical widely used in plastics.

The chemical bisphenol-A (BPA), used to make the plastic in baby and water bottles, food containers and thousands of everyday products, was mislabelled this week as a health menace.

A draft report from the US government’s National Toxicology Program (NTP), released on Tuesday, along with stories that Canada will soon become the first country to declare BPA toxic, have prompted a predictable wave of health hysteria. Canadian shops are already removing plastic water bottles and food storage containers from the shelves in anticipation of the government’s BPA decision.

Environmental and health activists, such as Greenpeace and the Environmental Working Group, have long argued that BPA represents a significant health threat given its association with low sperm counts, early onset of puberty, behavioural problems, impaired immune function, and prostate cancer. Last year a ‘consensus’ statement signed by 38 scientists claimed that the chemical might be harming infants.

In 2006, San Francisco banned the manufacture, distribution, and sale of toys and children’s care products made with or containing BPA. While the legislation was later withdrawn, similar legislation was introduced in California, Maryland, and Minnesota. The NTP’s findings may be the first step toward a global regulatory re-evaluation of BPA.

Given that BPA has been around for over 70 years, and in commercial use for more than 50 years, the scientific literature is enormous. The NTP looked at 500 studies of BPA in lab animals during its 18-month review. Indeed, the relevance of BPA effects in animals and the way in which BPA is typically administered in animal experiments, are two of the core issues in the BPA debate.

BPA finds it way into people through their diet. BPA in food and drink accounts for the vast majority of human exposure to the chemical. Typically, the BPA migrates into food through contact with food containers that are made of polycarbonate plastic. Biomonitoring studies have shown that BPA exposure is extremely widespread. Using data from the National Health and Nutrition Examination Survey (NHANES III) from 2003-4, the US Centres for Disease Control and Prevention found detectable amounts of BPA in 93 per cent of urine samples of those six years and older.

The crucial risk management question is whether the BPA levels found in human populations represent a significant, scientifically demonstrated threat.

Since there are important differences between the effects of BPA in humans and animals, as well as a lack of BPA studies in human populations, there is considerable controversy around what should count as an answer to this question but also what the answer should be.

There is no compelling scientific evidence that current levels of BPA exposure adversely affect human reproduction or development, though one would never discover this from this week’s scary headlines about BPA.

Equally true, however, is the fact that animals exposed to high doses of BPA during pregnancy or lactation can have reduced survival rates, low birth weights, and delayed onset of puberty in both males and females. The bottom line is that there is no evidence of harm at the current levels of human BPA exposure, and clear evidence of animal harm at high levels of BPA exposure.

Therefore, the crucial scientific question is whether low-dose BPA exposure in animals provides compelling evidence of harm in humans, particularly in young children. Here, the evidence has been controversial.

The NTP report references several studies of low-dose BPA effects, but also notes that they have been less than definitive because of ‘insufficient replication by independent investigators, questions on the suitability of various experimental approaches, [and] relevance of the specific animal model used for evaluating potential human risks’. More crucially, many of the low-dose studies of BPA have failed to establish a clear link between the dose of BPA and an adverse health impact, a fact that clearly undermines the claim that low doses of BPA in animals have any relevance to humans.

For example, even where they have been reports of health problems in animals exposed to low doses of BPA, there has been no evidence that these effects persist later in life. Despite, for instance, widespread headlines about BPA and breast cancer, the NTP report concludes that the evidence is ‘not sufficient’ to conclude that BPA is a risk for breast cancer.

Scepticism about the relevance of low dose animal effects for humans is also found in the assessments of other agencies. A comprehensive BPA risk assessment by the European Union found no convincing evidence that BPA affected development or was carcinogenic. The Japanese Ministry of Environment, which conducted extensive BPA research, including animal tests, also concluded that there was no clear evidence that low doses of BPA had endocrine-disrupting consequences. Germany’s Institute for Risk Assessment concluded that it did not ‘recognise any health risk for babies that are fed from baby bottles’ containing BPA.

Similar findings are contained in major risk evaluations of low-dose BPA effects in animals by Harvard’s Center for Risk Analysis and in a just-published study that looked at the effect of low doses of BPA on reproductive organ weights in rats.

Nevertheless, even if low-dose animal findings were relevant to human risks from BPA, even if they were numerous rather than limited, and even if they documented significant adverse effects, the crucial question remains the exposure levels.

Are infant and young children exposed to BPA levels anywhere near even the low-dose exposure levels in animal experiments? The answer is no. The NTP panel estimated that the doses that produced behavioural and developmental effects in animals were significantly higher than the BPA exposure levels found in infants and young children.

The NTP report notes ‘some concern’ for possible neural and behavioural effects in fetuses, infants, and children based solely on the ‘limited’ evidence from low-dose animal studies. In several cases, this evidence consists of a single study out of the 400 considered. This is very, very different from the scary headlines shouting that BPA causes breast cancer, developmental problems, and behavioural issues, and from the hasty calls to throw away every plastic product containing BPA.

A careful reading of the BPA science presents a very different picture to that gleaned from anti-BPA activists’ rhetoric. In truth, a very few studies have been used to justify a headline-grabbing finding. Given that there have been over 400 BPA studies since the last NTP review, if there really was a significant problem there would be more than limited evidence (that is, only a handful of studies) to justify something weightier than ‘some concern’ about BPA.

Patrick Basham and John Luik are co-authors, with Gio Gori, of Diet Nation: Exposing the Obesity Crusade, a Social Affairs Unit book. (Buy this book from Amazon(UK).)

Sources:

Health Canada Bisphenol A Announcement Imminent, CBC News, 15 April 2008

Bisphenol A. Summary Risk Assessment Report, Institute for Health and Consumer Protection, European Commission Joint Research Centre, 2003

‘Weight of the Evidence Evaluation of Low-Dose Reproductive and Development Effects of Bisphenol A’, G. Gray et al., Human and Ecological Risk Assessment (2004) 10: 875-921.

‘Gestational and Lactational Exposure to Ethinyl Estradiol, but Not Bisphenol A’, K. Howdeshell et al. Toxicological Sciences (2008) 102: 371-382.

US Cites Fears on Chemical in Plastics, Washington Post, 16 April 2008

Draft NTP Brief on Bisphenol A, National Toxicology Program, US Department of Health and Human Services, 14 April 2008