A new report suggests there's more evidence than previously thought that a chemical used to make polycarbonate may harm people when it leaches from plastic bottles and containers.
Several prominent scientists are calling for a thorough government review of the issue.
The report, written by two university professors and published in the scientific journal Environmental Health Perspectives, analyzes 115 studies on the safety of bisphenol A, and found that those funded by governments overwhelmingly reported problems, while those paid for by industry said the chemical was safe.
Industry officials disputed the EHP report and said evidence continues to point to BPA's safety. But several scientists who led or were active in previous BPA reviews said enough new evidence has surfaced in the past few years to warrant a taxpayer-funded, full-scale review to set new safety levels.
Such a review could have far-reaching implications for BPA, which mimics human hormones like those in birth control pills and which some studies have suggested is linked to lower sperm counts. An Environmental Protection Agency spokeswoman said the agency has no immediate plans for a review, but will consider the new study.
Scientists have debated hotly BPA's safety for nearly a decade, in language that is sometimes strong: One author of the EHP report, University of Missouri professor Frederick vom Saal, recently compared the plastics industry's scientific interpretation with that of cigarette makers.
Others have not used quite that language, but say accumulating evidence about BPA is raising questions that need better answers.
``I think there's a strong and compelling justification to have a new risk assessment,'' said George Lucier, former director of the Environmental Toxicology Program at the federal National Institute of Environmental Health Sciences and co-chair of a lengthy 2001 study of BPA conducted by the National Toxicology Program. He participated in an April 13 news conference unveiling the EHP report.
The other co-author of the report, adjunct professor Claude Hughes of East Carolina University, added: ``There's a lot more data now than there was a few years ago, so all that needs to be thoroughly scrutinized.''
Hughes participated in the 2001 NTP study and was one of several authors of a 2004 study from the Harvard Center for Risk Analysis that largely pronounced BPA safe. That study, funded by the American Plastics Council, examined 19 other studies and has been cited heavily by industry.
Hughes said the Harvard study is sound, but did not look at data after 2002 and was limited in scope to examining reproductive health effects. The NTP review, which said studies on both sides were credible, was broader than the Harvard review, both Hughes and Lucier said.
The report in EHP, a journal published by the National Institutes of Health, said that significant effects were found in 94 of the 115 studies, and that in 31 studies of animals, levels of BPA below the predicted safe dose still produced significant effects.
None of 11 industry-funded studies found problems, while conversely, more than 90 percent of the government-funded studies did, the EHP report said.
``Do investigators funded by a government grant strive to get positive results? Maybe so,'' East Carolina's Hughes said. ``Do investigators funded by other sources strive to get results that will support the point of view of those sponsors? Maybe so. One hopes not.''
Steve Hentges, director of the American Plastics Council's polycarbonate unit, said the EHP report is a weak analysis that does not weigh the relative strengths of different studies.
``The sum of weak evidence does not make for strong evidence,'' he said. ``If the literature is properly evaluated, in a weight-of-evidence fashion, the conclusion remains that BPA does not pose a risk to human health at the levels at which people are exposed.''
The two sides sometimes present very different interpretations of data.
Vom Saal said BPA has been shown to harm human cells at about 20 parts per trillion, well below the 1.3 parts per billion found in the blood of 95 percent of people examined in general population studies by the Centers for Disease Control.
But Hentges countered that looking at effects in cell studies alone does not reveal much, because the body is much more complicated than a single cell. He said the average dose people get is 1,000 times lower than the doses in studies, and government assessments in Europe and Japan have not found problems for human health.
However, researchers are concerned that BPA is found so widely in people throughout the developed world, and one key task for a government review is to figure out where exactly it is coming from, Lucier said. BPA is used in polycarbonate products from bottles to compact discs, for example, and also in can liners and sometimes as a dental sealant.