The importance of scientific debate

The peer review of a study does not end with its publication. An important form of the post-publication peer review process is repetition of a study, which does not happen often, if at all. Another form is the letter to the editor which allows another, perhaps different, examination of the study’s results. For that reason, it is our view that there should be a low threshold for publication of such letters as they provide a useful contribution to the continuing development of scientific knowledge.

Recently, I submitted, with co-author Nicholas Druar, a letter to the editor of Critical Reviews in Toxicology (CRT). Our letter critiqued a 2012 article by Donovan et al. published in that journal. Dr. Roger McClellan, CRT’s editor of, refused to publish the original version and subsequent revisions we made to this letter. Because we believe in the importance of scientific debate, IJOEH is publishing our response to Donovan as a “Special Contribution” in this issue. We welcome and will publish any response that Dr. McClellan or Donovan et al. may have to this editorial and letter.

In our view, Dr. McClellan based his rejection of our letter on arguments which run counter to the principle of the exchange of scientific views, a key component of the peer review process. More importantly, other editors’ adoptions of his approach would completely stifle the peer review process and critical review. We take this opportunity to open a discussion on the censorship of scientific dialog. We outline and critique the reasons Dr. McClellan provided for his refusal to publish our response to Donovan et al.’s review of para-occupational (take home) exposure to asbestos, published for use in asbestos-related litigation.

Dr. McClellan claims that the primary focus of our letter should be on the ‘science’ of the article.

The majority of our response addressed the scientific issues relevant to the original review. On the other hand, Donovan et al.’s article begins and ends with references to litigation: “In recent years the number of legal cases involving alleged para-occupational or ‘take home’ asbestos exposure has increased dramatically.”¹ The conclusion of the paper focuses solely on litigation: “In recent years, toxic tort litigation involving asbestos claims has reached record levels; an increasing number of these cases involve claims of para-occupational exposure. Our analysis indicates that, to date, the vast majority of household cases reported in the literature occurred among individuals living with one or more family members who worked in industries characterized by high exposures, nearly always to amphibole fibers, and frequently during the 1930s–1960s.”¹

Further, following the publication of the original paper, ChemRisk, the authors’ employer, sent a letter soliciting litigation consulting to a “valued client” and wrote, “Please pay particular attention to the ‘asbestos take home’ paper. It represents a major commitment by our firm.”² The disclosure published with the article states that it was funded by ChemRisk and not written at the request of any organization. ChemRisk’s self-funding of the paper appears to be an effort to make the paper more acceptable in the litigation context. In its 1995 Daubert decision, the US Supreme Court specified that a study was less credible if it was funded or conducted for the sole purpose of influencing litigation.³

Furthermore, Donovan et al. indicated that their paper arose from questions related to litigation, not science. Their declaration of interest read:

The gathering of articles that were read and analyzed was associated with work done to prepare other published papers or for assembling information to be presented in courtrooms, an effort that was supported by a number of companies that have been, and are, involved in asbestos-related litigation. No organization asked the authors to write this paper, and funding for doing so was provided by the authors’ employer, ChemRisk, LLC. One of the authors (DJP) has served as an expert in asbestos litigation, including claims regarding para-occupational exposure.¹

Finally, the question of exposure to household asbestos has no scientific or policy implications apart from litigation. OSHA regulations related to eliminating such risk went into effect in 1972.⁴ Household exposures, which asbestos companies never measured, cannot be used to estimate potency or dose–response relationships or any other scientific issue. Indeed, the authors do not suggest any scientific implications of their “findings.”¹ In our view, the social and political implications of a paper are appropriate subject matter for a letter to the editor, particularly in response to review articles. Furthermore, it is absolutely appropriate subject matter in response to an article which arose from litigation-related issues.

Dr. McClellan objected to our use of non-peer reviewed documents (e.g. internal memos of companies).

Donovan et al. relies on unpublished corporate documents. Donovan et al. note:

More recent literature has utilized data from well-established mesothelioma and cancer registries in various countries, and even from records from law firms participating in asbestos litigation.¹

They describe this in their methods section:

Using the same search terms described above, a targeted review was also conducted of the Claims Resolution Management Corporation (CRMC) Asbestos Claims Research Facility in Aurora, Colorado, which houses a collection of more than 32,000 boxes of documents, 7200 rolls of microfilm, and 5000 subject-related and witness files turned over to the facility by the Johns-Manville (JM) Corporation. The search results yielded many different types of documents, including scientific articles published in the peer-reviewed literature, product specifications, US Navy memoranda, safety handbooks, internal JM documents, and other surveys, reports, and letters. Approximately 80 boxes of health, safety, and environment documents, as well as two boxes of microfilm files, each containing 42 rolls of film, were reviewed in order to locate pertinent information. This search, however, did not result in any new information that was not already publically available.¹

Literature that is more recent has utilized data from well-established mesothelioma and cancer registries in various countries as well as records from law firms participating in asbestos litigation.^5–11

While they claim their search did not uncover any ‘new information,’ they did omit contradictory non-peer reviewed corporate documents that were in the JM archives. It is necessary to rely on other unpublished contradictory corporate documents and studies to address the issues Donovan et al. raise in the corporate documents they cite. Corporate knowledge is a component, often the most important component, of knowledge concerning para-occupational exposures. The companies who utilized asbestos had the earliest and most extensive knowledge in this area because their workers’ spouses and children were the first to develop disease as a result of take home exposure. Since Donovan et al. rely on non-peer reviewed literature, including a large collection of JM documents, we do not understand why he asked us to omit similar or identical (Manville Trust) sources.¹

Dr. McClellan also asked that the letter be less than 1000 words, but we are unable to find any such requirement on the journal’s website or within the journal’s instructions to authors. Indeed, at least one letter published by the journal is over 1500 words.⁵ The original Donovan et al. paper is approximately 10 000 words.

Finally, Dr. McClellan insisted that I use my business affiliation and email address rather than my academic address, citing the correspondence between the other author and himself that had included a copy to my business address. I appropriately identified my potential conflict related to my participation in asbestos litigation at the request of both asbestos manufacturing companies and workers with asbestos-caused diseases. It is the solely up to an author and his institution to determine if work is to be identified with academic pursuits or a private consulting arrangement. As with any other academic paper or editorial like this one, the views expressed are those of the authors and not the University. I considered the letter to the editor to be part of my academic activities. The CRT editor has no basis to claim otherwise.

Peer review of a scientific paper is a continual process which does not end with its publication. We sent previous versions of this editorial to Dr. McClellan and our letter to Dr. Donovan and we invited them to publish any response to these communications. Although they declined, our offer remains open. Particularly in the age of the Internet, increased access invites and facilitates critical examination of the scientific literature, helping foster a higher standard in research and analysis. Limiting the publication of critical comments based on arbitrary criteria unnecessarily restricts scientific dialog. Editors, the gatekeepers of these publications, and thus, the state of scientific knowledge at large, should encourage critique and discussion of all aspects of the research they publish, rather than search for reasons to limit scientific discourse.

Disclosure: The author has served as an expert witness at the request of plaintiffs and defendants in asbestos-related litigation.