by Dick Clapp

The publication of my article on mortality among IBM workers was the culmination of a two and a half year process. I obtained the data, which included information on the deaths of nearly 32,000 former workers who had died between 1969 and 2001, when I served as an expert witness in a lawsuit brought against IBM on behalf of employees who had developed cancer after working at the company’s San Jose facility. I found that among the workers, the death rates from several cancers—including cancers of some digestive organs, kidneys, brain and central nervous system, melanoma of the skin, and non-Hodgknin’s lymphoma —were particularly high when compared to the national averages.

IBM won the San Jose jury trial and then settled the lawsuits with the remaining plaintiffs, but I still thought that it was important to publish the study to make others aware of the occupational health risks in these manufacturing activities. IBM’s lawyers derided my work (one said it gave “junk science a bad name”) and asserted that I couldn’t publish the results because the data was confidential under a court order. I had to undergo a lengthy process in order to publish this study, but I am convinced that it was worth it now that I see that it has helped bring greater attention to the occupational health risks related to computer chip manufacturing.

My first attempt to get a manuscript published was thwarted by the publisher of Clinics in Occupational and Environmental Medicine, which claimed that the article would not be considered because it was original data and they only published review articles.  This led to a boycott by other authors and widespread publicity about interference with academic freedom, suspected meddling by IBM lawyers, and similar stories. 

After a year or so, lawyers for the plaintiffs in a New York lawsuit filed a request for a ruling by the court regarding confidentiality of the data on which the article was based and restraining IBM from interfering with publication.  The lawyers for IBM opposed this, with accompanying affidavits from consulting epidemiologist Jack Mandel, from Exponent (a company known for its product defense activities) and Emory University, and others employed by IBM.  Mandel argued that the article was based on a method that could not estimate risk of death, was based on death certificate information that was required to be kept confidential, and that another study done by consultants for IBM (Beall, et al., 2005) used the only valid methodology.  He further said that if he were a reviewer for a journal, he would not recommend the study for publication.

New York Judge Joan Lefkowitz entered her decision on February 22, 2006 and concluded that IBM had failed to show good cause for sealing the manuscript and enjoined IBM from interfering in its publication.  The ruling provided for a ten-day period during which IBM could file an appeal, which their lawyers did not pursue.  As a result, I submitted the manuscript to a medical journal in May.  The journal immediately responded that it did not publish articles of this type and declined to review it.

Subsequently, I expanded the manuscript and submitted it to Environmental Health, an open access on-line journal.  The editors chose three peer reviewers and forwarded their comments to me for consideration.  All three recommended publication, and two requested revisions and clarifications.  One of the peer reviewers was a subcontractor and co-author of the IBM study published a year earlier.  He had seen my presentation of the results of the two studies in a class at the Harvard School of Public Health and was aware of the similarities and differences in results.  I incorporated all of the peer reviewers’ comments into a final version of the manuscript and the editor accepted it for publication in late August.

Because the article appeared to be newsworthy, the editor asked for a delay in posting it on-line while a press release was prepared.  The manuscript was embargoed and a press release was sent to various news outlets.  On October 19, 2006, the article was published, and William Bulkeley, a Wall Street Journal reporter who had requested an interview before the release date, wrote a long article that appeared the same day.  A Reuters news story also appeared on the same day, followed by a large number of stories in local newspapers, radio and television stations in New York state, Vermont and elsewhere.  There were stories in scientific journals and newsletters, computer industry trade newsletters, the Financial Times, the Times of London, Voice of America and many other outlets.

The on-line journal tracks the number of times articles are accessed and posts this on the website.  In the first month, the article was accessed over nine thousand times and reached the second position among the 100 articles published since the journal began in July, 2002.  As of Nov. 23, the article is approaching 10,000 accessions and has been the topic of email and listserv conversations throughout the health and safety and environmental networks.  I presented it on a panel at the American Public Health Association meeting in early November and many individuals have expressed thanks and congratulations.  Needless to say, this is all very satisfying and makes the two-year struggle to get the article published seem worthwhile.

IBM, for its part, commented on the published article in a letter.  Their spokesperson said it was based on flawed methodology and woefully inadequate data.  He also contended that the company’s study, which was much larger, showed decreased mortality overall and from cancer.  My response, delivered on Vermont Public Radio, was that IBM was wrong on all three counts.  I also pointed out to several reporters that one of IBM’s consultants in their study was a peer-reviewer of my manuscript; he approved it for publication. 

I have been asked by several reporters if there is anything else I would add that they didn’t ask about.  In every instance, I say that this study is about worker protection and the opportunity to prevent exposures to harmful chemicals so that future deaths from workplace toxins can be avoided.