By David Michaels

The controversy continues over NIH’s review of Bisphenol A (BPA), and the agency’s firing of Sciences International. Members of the NIH’s BPA Expert Panel have joined the discussion, through comments to the Pump Handle, assuring the public that their work was not not influenced by any potential conflicts.

In addition, today’s Washington Post and Los Angeles Times both have articles with additional details on the conflicts of interest that triggered the firing, along with comments on the process to date. There is also an interview with a panel member defending the quality of Sciences International’s work.

And a member of the Expert Panel has written a letter to the Post, with a copy to the Pump Handle, clarifying what he believes are mistakes in the Post’s coverage.

First, here’s the text of note the members of the panel sent to us yesterday:

We are members of the Bisphenol A Expert Panel of the National Institutes of Health, Center for the Evaluation of Risks to Human Reproduction (CERHR). CERHR reviews reproductive and developmental toxicity for chemicals of public concern. Recently, concerns were raised regarding the influence of contractors used by CERHR to aid the Expert Panel. We believe that independent review is the cornerstone of good science and we are proud to use our scientific expertise in the service of public health. Expert Panel members are charged to review available data and render an unbiased and scientifically rigorous opinion. We guard our objectivity, independently evaluate the literature, and work together to reach consensus. Scientific integrity is our most valued possession, and we wish to reassure the public that this process has not been, and will not be, prejudiced by outside influences. Simply said, our final product will be based upon the highest quality science.

Today, Marla Cone, who broke the original story in the Los Angeles Times, quoted Anthony Scialli, a Sciences International vice president, who said that NIH’s decision was unfair because the institute had no conflict of interest policy.

“Frankly, we feel we’ve been unjustly treated,” said Scialli, who had been the federal center’s chief scientific investigator.

If the institute “felt it was necessary to have more stringent conflict of interest policies, they should develop them and make sure their contractors comply with them. But to end our contract because we didn’t comply with policies that didn’t exist doesn’t make sense,” he said. “If they want to hold us to standards, they should tell us what those standards are and we will cooperate.”

Scialli said his firm was fired because of political pressure from environmentalists, mainly the Environmental Working Group. Sciences International had one year left in a five-year, $5-million contract.

Institute spokeswoman Robin Mackar said the agency had no comment on the contract issue.

Court documents show that last year, Scialli was hired by 3M as an expert witness to testify in a lawsuit filed by Minnesota residents whose water was contaminated by a chemical that 3M formerly used in Scotchgard. The chemical has been detected in most people’s blood, and some studies suggest it can cause reproductive damage.

Scialli said he notified the federal health center that he was working for 3M but officials seemed unconcerned.

Environmentalists say that points to the need for the institute to develop a formal policy prohibiting conflicts of interest.

“If they think it was OK for their contractor to be an expert witness for 3M, that’s very disturbing,” said Richard Wiles, executive director of the Environmental Working Group.

Sciences International’s main role for the government was preparing drafts of 20 federal reports that summarized scientific data about the health effects of individual chemicals.

Included is a controversial report on bisphenol A, a compound in polycarbonate plastic linked to prostate cancer and other reproductive effects. Some scientists who study the chemical say that the draft report appears to have a pro-industry bias because it omits some findings and downplays its risks.

Some Sciences International employees previously conducted work for BASF and Dow Chemical, two manufacturers of bisphenol A, although that work involved other chemicals. The firm was also paid by three chemical industry trade associations to perform scientific consulting work on three other chemicals that the company reviewed for the government’s reproductive health center.

Wiles welcomed the dismissal of Sciences International and urged the institute to review all 20 of the center’s past decisions on chemicals.

But Scialli said that his role was only to summarize scientific data for the center’s panel of scientific experts and that it was uninfluenced by the chemical industry work.

“What we wrote was very dry and technical. We provided no commentary at all,” he said. The company knew bisphenol A was particularly contentious, so “we were going out of our way to produce a report that was accurate and complete,” he said.

More controversial is Lyndsey Layton’s piece in today’s Washington Post, an article criticized by a member of the BPA Expert Panel. Layton writes

When it was fired on Friday, [Sciences International] the Alexandria-based company was reviewing about 500 scientific studies on bisphenol A, a chemical common in plastics that has been linked to cancer and reproductive problems in animals. NIH had given Sciences International a contract to prepare a summation for a panel of experts responsible for determining whether the chemical poses risks to human fertility or development.

This is a statement the panel took issue with. When Dr. L. Earl Gray, a member of the BPA Expert Panel, read this article, he sent a letter to the Post (with a copy posted here), asserting that the Post was incorrect to suggest that Sciences International reviewed the BPA studies; rather, this was done by panel members. Here’s an excerpt from his letter:

You stated SI was reviewing 500 articles when they were fired. This does not accurately reflect the role SI was playing in the review process. At the time of the first BPA meeting SI was not in the room when the Expert Panel was reviewing the studies in the report. Members of the Expert Panel recently finished the first review of all of the BPA studies relating to the reproductive or developmental effects of this chemical. Neither SI nor CERHR staff were present during the review by Panel members and the papers were evaluated according to rigorous scientific criteria developed solely by the Panel. The Panel also rewrote summaries of each article, as warranted by the panel.

The Panel also rewrote the limitations and strengths and utility sections for each article since this is the first time we had a chance to meet altogether and come to a consensus. The Panel will rereview our utility sections for key papers and then a summary will be prepared. The summary will be based only upon the high quality studies identified by the Panel.

This report is the independent work of the BPA Expert Panel and is not influenced by either SI or CERHR staff. The current draft of our report, with the changes from previous documents shown, is scheduled to be put online later this week by CERHR.

It appears that what happened here (and please correct me if I’m wrong) is that SI prepared a report that included hundreds of studies, and the BPA expert panel conducted their review of their report without the presence or input of SI staff.

If that’s the case, there is still a potential problem. If any bias existed in the preparation of the draft report, the panel’s ultimate recommendations could be affected because the draft report is what they base their work on. In comments she submitted to NIH’s National Toxicology Program about the BPA review, Sarah Janssen of the Natural Resources Defense Council lists examples of recent studies that she thinks should have been included in the draft BPA report but were not (see the post by Janssen and Jennifer Sass here, or a PDF of Janssen’s comments here).

This does not necessarily mean that SI’s draft report was influenced by the fact that some of its staff members have ties to chemical companies, and it is certainly possible that panel members knew or were informed about the studies Janssen listed and were able to take them into account. The issue is, and has been from the start, that NIH (or at least the National Toxicology Program) lacks the policies to deal with conflicts of interest – so when controversies like this arise, it is not possible to assure the public that the process and the scientists involved were unconflicted.

But managing conflict of interest is a difficult task. Follow the rest of the Post’s article:

Sciences International’s corporate clients have included Dow Chemicals and BASF, two companies that manufacture bisphenol A.

Herman Gibb, the president of Sciences International, called the firing “unfair.” He said his company’s work for BASF predated its federal work on bisphenol A, and he described an 11-employee firm where workers assigned to federal jobs were unaware that other employees were working for industry. None of the science was compromised by the firm’s business ties, he said.

I don’t ever believe in my heart of hearts there was a conflict of interest,” Gibb said.

Robert Chapin, the chairman of the expert panel selected by NIH to determine whether bisphenol A poses health risks, said Sciences International is being unfairly tarnished.

On all of the panels of which I’ve been a member, SI has presented nothing but balanced and scientifically rigorous summations,” said Chapin, who works for Pfizer, the pharmaceutical company. “This is all just theatrics. This has to do with a campaign by outside interests to hijack the process. SI was doing a perfectly fine job.”

Gibb acknowledged that his company was working for three chemical trade associations at the same time it was performing federal reviews of two chemicals linked to those groups. He said he learned of those potential conflicts last month when NIH asked him to review the company’s corporate contracts.

Allen Dearry at NIH said he and other federal officials were sufficiently concerned to terminate the bisphenol A contract, but the government will not revisit the company’s past work on other chemicals. “To the extent we could evaluate the work that SI performed, we tried to assess it and were satisfied,” he said.

Dearry said the agency is taking steps to “ensure the integrity of our work and science.” For the first time, it will require all current and future contractors to disclose any potential conflicts of interest regarding their federal work. In addition, the agency will convene an independent panel of scientific experts to assess all contracts let by the National Toxicology Project for conflicts of interest and report its findings by July 1, he said.

snip

Rep. Henry A. Waxman (D-Calif.), chairman of the House Oversight and Government Reform Committee, has begun an investigation into the work the company performed for the government. A committee staffer said Waxman wants an independent review of Sciences International’s work on other chemicals to determine whether the contractor had conflicts of interest and, if so, whether the conflicts affected the federal work.

snip

The federal contract represented about half of Sciences International’s income, and the company will be forced to lay off employees, Gibb said. He said it is unclear whether the company has legal grounds to challenge its dismissal.

Whatever anyone thinks about the quality of the BPA review, its clear that we’d all have greater confidence in both the process and the outcome if conflict of interest issues had been addressed up front.

David Michaels heads the Project on Scientific Knowledge and Public Policy (SKAPP) and is Professor and Associate Chairman in the Department of Environmental and Occupational Health, the George Washington University School of Public Health and Health Services.