The Voice of Epidemiology


    Web EpiMonitor

► Home ► About ► News ► Job Bank Events ► Resources ► Contact
New Controversy Swirls Around IARC’s Categorization Of Herbicide Glyphosate As “Probably Carcinogenic”

Epidemiologist Accused Of “Withholding” Data Showing
No Increased Risk of Cancer

The International Agency for Research on Cancer (IARC) is in the news again in connection with its determination in 2015 that glyphosate, an ingredient in the widely used Roundup weed killer by Monsanto, is “probably carcinogenic”. Different issues have arisen since the IARC determination was made. The reason for the latest stir up is a special report published in June 2017 by Kate Kelland, a Reuters journalist, declaring that epidemiologic data relevant to the IARC’s review of glyphosate were not considered.


The stakes surrounding this controversy are high since Monsanto’s  popular weed killer has many benefits and advantages described in a Science report as “cheap, highly effective, and is generally regarded as one of the safest and most environmentally benign herbicides ever discovered.”

The IARC conclusion has triggered lawsuits against Monsanto alleging that exposure to Roundup has caused non-Hodgkins lymphoma, and the release of previously confidential documents associated with the lawsuits (the “Monsanto papers”) has given rise to allegations of cover up and catalyzed other accusations on both sides of the dispute.

Current controversy

The data not considered by IARC were updated results from the large NIH-sponsored Agricultural Health Study which reportedly show no connection between glyphosate and non-Hodgkins lymphoma. Earlier but more limited data from the same study also showed negative results. Those published negative data were reviewed and considered by IARC, but not the updated negative findings which include additional cases and follow-up from the same study.

The Reuters report accused Aaron Blair, former NCI investigator and chair of the IARC panel, of withholding the latest information which it claims could have changed or downgraded the IARC’s risk categorization of glyphosate. To be more precise, Blair and the panel did not technically “withhold” data since by the rules of the IARC process only published, in press, or readily available findings such as government reports are suitable for consideration.


There are a number of issues raised by the various media accounts of this controversy. The first is whether or not relying only on published data to conduct IARC reviews is a good policy. Asked about this question, Blair told one reporter “The rule is you only look at things that are published. What would it be like if everyone on the working group whispered things they knew but weren’t published and made decisions on that?” This view contrasts with others described in media accounts which consider the ban on unpublished data as unjustified or even “absurd”.

However, the IARC has defended its policy saying the cancer evaluation program “does not base its evaluations on opinions presented in media reports”, but on the “systematic assembly and review of all publicly available and pertinent scientific studies, by independent experts, free from vested interests.” This has not satisfied everyone and other criticisms of the IARC process have been voiced, including the selection of workgroup members and even the quality of the data included in the review and the interpretations made.

Blair Reactions

In an interview with The Epidemiology Monitor, Blair made clear that any data must be considered provisional until published. There is always the possibility that analyses could change or be interpreted differently until the investigators are finished with it, he told the Monitor.

Other reasons for considering only published results are that they often contain a fuller description of the study and its findings and have undergone peer review. Adopting a policy to consider unpublished and non-reviewed data would raise a host of thorny practical problems such as how to locate the unpublished work and how to avoid being biased by the direction of the findings. It is sound policy according to Blair for the IARC or any review group to consider only published data or completed papers “in press”. 


Some media accounts report that Blair stated that the unpublished data would have changed the IARC’s classification had they been considered, while another article reports Blair saying in his leaked deposition that nothing has changed his opinion about glyphosate and non-Hodgkins lymphoma. Asked directly about this by the Epi Monitor, Blair clarified that nothing contained in the totality of the evidence in the IARC review has changed his mind about glyphosate. He did not address  specifically whether the unpublished AHS data would change his mind since for Blair those data are not final and therefore not qualified to be considered at this time. He emphasized that change is always possible before the final analyses are completed and that relying exclusively on “finished” work is the best approach. It does not preclude reconsideration of an issue after significant new relevant information has been published.

Another member of the Blair IARC panel, John McLaughlin, an epidemiologist at the University of Toronto, told EcoWatch that the unpublished work did not alter his view about the validity of the IARC conclusion on glyphosate. It is not clear how he learned of these findings or in what form.

            Hard to Fathom

The contrasting media and other accounts illustrate how complicated it is for consumers to find truthful and reliable information about health risks related to chemical exposures. The Reuters reporter Kate Kelland, who first surfaced the accusations against Blair, has been criticized by Carey Gillam from U.S. Right to Know for cherry-picking information and being susceptible to the public relations efforts of large companies such as Monsanto. According to Gillam who is referring to all the publicity surrounding the Reuters article, “Follow up the hand-fed story with a press release from an industry-funded outlet and calls for an investigation from the industry group American Chemistry Council and you have propaganda gold. What you don’t have is the truth.”

Big Picture Perspective

Asked for his perspective as an epidemiologist on these disputes, Blair told the Monitor that investigators working on similar topics must be prepared for the heat of battle.  He takes a big picture view that we live in a democratic society where information must be made public and regulators and representatives must decide what precautions are needed, if any. Sometimes decisions are made to tolerate risks for the benefits conferred, and he cited the more than 30,000 deaths per year due to motor vehicles. Society must go through the process of debating these risks and benefits as it has for asbestos, tobacco, and other exposures. Eventually a decision we are willing to live with as a society is arrived at. We are still in that process regarding glyphosate, he said. ■


Reader Comments:
Have a thought or comment on this story ?  Fill out the information below and we'll post it on this page once it's been reviewed by our editors.

  Name:        Phone:   



      ©  2011 The Epidemiology Monitor

Privacy  Terms of Use  Sitemap

Digital Smart Tools, LLC