Thou dost protest too much. Let the disclosure chips fall where they may

Earlier this month I wrote about the merits of policies that require conflict of interest disclosures. Last week, two items also about conflicts of interest landed in my in-box. They were just too juicy to not take a bite, and write about here.

First came a commentary from the October 2013 issue of the Annals of Occupational Hygiene written by the journal’s chief editor Noah Seixas, PhD, MS.  The lead paragraph reads:

“On 6 June 2013, a court in New York handed down a decision that calls into question the validity of research that was sponsored by Georgia-Pacific [GP] and published in eight peer-reviewed articles, including two papers published in the Annals of Industrial Hygiene. While the court did not determine that any fraud has occurred in the process of producing these papers, it did state that the research had been conducted under the auspices of the corporation’s counsel and that the counsel, who was not listed as an author, was ‘significantly involved in the prepublication review process.’”

“Further, the court stated that GP had commissioned the program of studies on risks associated with chrysotile in drywall work with the intent to ‘cast doubt on the capability of chrysotile asbestos to cause cancer.’”

The Seixas goes on:

“The court suggests that the research was sponsored with a particular outcome in mind and that the sponsor of the research, without attribution, had significant input to the content and reporting of the results.”

I had to look up these papers for myself, wondering if they were published decades ago when disclosure requirements were not as vigorous as today’s. To my surprise (sort of), they appeared in Seixas’ journal in 2011 and 2012.

It’s been 36 years since the World Health Organization’s International Agency for Research on Cancer (IARC) designated all forms of asbestos as human carcinogens. IARC has reevaluated the evidence numerous times over the years, most recently in 2012. Each time the WHO agency has reiterated the conclusion: all forms of asbestos are human carcinogens, and specifically associated with mesothelioma, and cancers of the lung, pharynx, larynx,esophagus, stomach, colorectum, and ovary.

The chief editor of the Annals of Occupational Hygiene noted that the journal’s disclosure policy conforms with that of the International Committee of Medical Journal Editors. He indicated, however, that the authors of the Georgia-Pacific research apparently failed to comply with that policy. Chief editor Seixas added that the journal will be strengthening its disclosure policy and concluded:

“the editors have the responsibility and therefore retain the right to consider potential conflicts of interest in our decision to publish, or not, a specific piece of research. Clear and complete transparency is necessary to fulfill this roll. We will continue to publish the best science we can.”

Considering the impact of potential conflicts of interest on scientific integrity is an important role for journal editors. Disclosing such information allows readers of the articles to consider potential conflicts on the study’s methods and reported findings.

Last month OSHA said it would be requesting, (but not requiring) similar disclosure from those who submit reports, studies and analyses during the public comment period on the agency’s proposed rule on respirable silica. I applauded the agency’s decision. It didn’t take long, however, for some industry groups, including the U.S. Chamber of Commerce, to squawk about it.

A story in the October 17 edition of Bloomberg BNA’s Occupational Safety and Health Reporter offers reaction from industry attorney David Savardi. (He’s a partner at Keller and Heckman where the firm’s motto is “Serving Business through Law and Science.”)  BNA reporter Stephen Lee writes:

“David Savardi, an industry attorney with Keller & Heckman LLP, said Oct 11 the disclosure request could lead to a form of ad hominem attack that challenges the commenter, not his or her comments. The disclosures are also likely to color an agency’s treatment of the comments, Savardi said. ‘When OSHA sees something that’s been submitted by so and so that’s been supported by such and such organization, inevitably, in the back of their minds, a bias is going to be created against that information,” he said. ‘And their job is to evaluate it on its face—its accuracy, logic, supporting information, whatever—not on who the people are, or how they’re paid.’”

The trial judges in the Georgia-Pacific case clearly disagree.

Here’s the second juicy item from my inbox: Investigative reporter Jim Morris of the Center for Public Integrity (CPI) exposed this week Georgia-Pacific’s secret research program.  It’s the same one about which journal editor Noah Seixas referred to in his commentary. The research program was launched it 2005, as Morris reports, when the paper company was facing nearly $1 billion in liability because of its asbestos-containing joint compound.

Morris’ article is published at Salon.com and CPI’s website. It is one of the best investigative pieces in a long time reporting on how industries manufacture doubt about the serious health consequences of their products. Georgia-Pacific’s (GP) strategy and behavior is especially egregious because during litigation the firm refused to turn over documents related to the studies it commissioned. The players for GP included toxicologist David Bernstein (who cut his teeth for the tobacco industry) and the product-defense consulting firms Exponent and Environ (discussed extensively in the book Doubt is Their Product.)

Morris provides quotes from the trial judge and others.  Some of them are real doozies:

  • Trial judge Sherry Klein Heitler wrote: “Georgia-Pacific cannot use its experts’ conclusions as a sword while at the same time attempting to shield the public from information which affects the veracity of its experts’ conclusions. …The court is concerned that Georgia-Pacific’s attorney would be involved in any discussions concerning the content of these purportedly objective scientific studies by Georgia-Pacific’s consulting experts.”
  • John Childs, GP’s corporate counsel for litigation, instructed GP’s director of toxicology and chemical management on his duties: “Your work will be directed solely by GP’s in-house counsel.”
  • The five-judge state court of appeals upheld Heitler’s demand for disclosure, writing: “None of the [published] articles disclosed that [Georgia-Pacific’s] in-house counsel had reviewed the manuscripts before they were submitted for publication. Two articles falsely stated that [Georgia-Pacific] did not participate in the design of the study, analysis of the data, or preparation of the manuscript.”

And taking the cake is the quote from Harvard University Professor Sheila Jasanoff, JD, PhD,a scholar and expert on the use of science and technology in law, politics and policy.  She remarked to Morris:

“There’s something extremely smelly about claiming attorney-client privilege for something that is being claimed at the same time as good science. Legal confidentiality protections should not be placed around good science.”

There’s a long history of industries, when threatened by litigation or regulation, using science to cast doubt on evidence that their products are harmful to health or the environment. This GP case involved asbestos, but we’ve seen it with tobacco, lead, chromium, benzene, diesel particulate matter, just to name a few.

With the track record of powerful economic interests to manufacture doubt, policy-makers are wise to expect disclosure of potential conflicts of interest. OSHA’s request of commenters to (1) disclose funding sources and (2) the extent to which the research was reviewed by potentially affected parties before publication or release, are sensible and reasonable. They make sense for publications in journals, in litigation and in regulatory proceedings.

I hope that Members of Congress who are quick to forewarn federal agencies to use “sound science,” don’t become hypocrites by voicing opposition to OSHA’s modest disclosure request.