The National Academies of Sciences, Engineering and Medicine
Office of Congressional and government Affairs
At A Glance
: Gulf War and Health -- Institute of Medicine Reports
: 07/30/2009
Session: 111th Congress (First Session)
: Lynn Goldman

Professor, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, Maryland; and Chair, Committee on Gulf War and Health: Review of the Medical Literature Relative to Gulf War Veterans Health and Chair, Committee on Gulf War and Health: Fuels, Combustion Products, and Propellants and Member, Committee on Gulf War and Health: Insecticides and Solvents, Board on Health Promotion and Disease Prevention, Institute of Medicine, The National Academies

: House
: Veterans' Affairs Committee, Subcommittee on Oversight and Investigations


Gulf War and Health – Institute of Medicine Reports

Testimony of

Lynn Goldman, MD, MPH
Professor, Bloomberg School of Public Health
Johns Hopkins University
Baltimore, Maryland
Chair, Committee on Gulf War and Health: Review of the Medical Literature Relative to Gulf War Veterans Health
Chair, Committee on Gulf War and Health: Fuels, Combustion Products, and Propellants
Member, Committee on Gulf War and Health: Insecticides and Solvents
Board on Health Promotion and Disease Prevention
Institute of Medicine
The National Academies

before the

Subcommittee on Oversight and Investigations
House Committee on Veterans' Affairs
U.S. House of Representatives

July 30, 2009

Good morning Mr. Chairman and Members of the Subcommittee. Thanks to Congressman Mitchell and members of the Subcommittee on Oversight and Investigations, House Committee on Veterans' Affairs for your concern about veteran’s health.

My name is Lynn Goldman. I am a professor of environmental health sciences and epidemiology at the Bloomberg School of Public Health at Johns Hopkins University in Baltimore and chair of our program in applied public health. Prior to joining Hopkins in 1999 I served for six years at the US Environmental Protection Agency (EPA) as Assistant Administrator for the Office of Prevention, Pesticides and Toxic Substances. My primary training is in pediatrics and epidemiology. I also have served as Chair of two Institute of Medicine (IOM) Gulf War and Health committees: the committee that worked on the report Gulf War and Health: Review of the Medical Literature Relative to Gulf War Veterans Health, and the committee that produced the report Gulf War and Health: Fuels, Combustion Products, and Propellants. Additionally, I was a member of the committee that produced Gulf War and Health: Insecticides and Solvents. I am here before you today because of my experience as a volunteer serving on those IOM committees and as an elected member of the Institute of Medicine.

I will focus on four main points in my testimony. First I will discuss the overall study process, including the review process, for the Gulf War series of reports and how that process compares to the study process for the IOM Agent Orange reports, including the report review process. Second I will discuss the categories of association used by the Gulf War & Health committees to classify the likelihood that exposure to a given agent is related to a given health effect, and how those categories compare to those used by the Agent Orange committees. Third, I will discuss how scientific studies are used by the Gulf War and the Agent Orange committees, with a focus on animal studies. Finally, I will discuss what is known about exposures during the Gulf War and how that affects the committees’ work.

Let me begin with the IOM study process. The IOM is a division of The National Academies, a non-governmental institution originally chartered by President Lincoln to provide independent scientific advice to the nation. That scientific advice is usually in the form of consensus reports produced by expert, unpaid committees. In the case of the Gulf War and Health and the Agent Orange studies, the committees usually comprised ten to twenty members with expertise in epidemiology, toxicology, exposure assessment and relevant areas of clinical medicine. The members are usually from universities, nonprofit organizations, and consulting firms. The reports are developed through an established study process designed to ensure committees and the reports they produce are free from actual or potential conflicts of interests, are balanced for any biases, and are independent of oversight from the sponsoring agency. At no time during a committee’s deliberations or during the preparation and review of an IOM report is the sponsor allowed to participate in the process or have access to any part of the report. In cases where a committee asks the sponsor for information, any such information is made public.

Committees review relevant literature, hear from experts, and deliberate. Once the committee has reached its consensus, but prior to the report being released, the draft report is subjected to a formal, peer-review process. External reviewers are nominated by a broad range of individuals including IOM and National Academy of Sciences (NAS) members, committee members and other interested parties. The list of reviewers’ is approved by a review oversight body, the National Academies Report Review Committee, which ensures the reviewers have the necessary expertise. The reviewers read the draft report and individually provide comments on: 1) whether the committee has addressed its charge; 2) the strength of the evidence for and the validity of the committee’s conclusions; and 3) the technical aspects, clarity and flow of the report. Comments of the reviewers are provided anonymously so that committee members and the study staff do not know the source of the review comments when they receive them. In the case of the Gulf War and Health and the Agent Orange studies, 10 to 15 experts in various scientific fields reviewed the reports. The committee must respond to each comment from each reviewer and indicate what revisions were made to the report to address the comment or provide a detailed explanation why the suggested revision was not made. After all the comments have been addressed, each study committee member must “sign off” on the revised report. The report is then sent to the Review Monitor, who is a member of the National Academies Report Review Committee, and a Review Coordinator, who is assigned by the IOM executive office. Those two individuals assess the committee’s response to reviewers’ comments and ensure that the committee has adequately addressed every comment. Only when they are satisfied is the final report released to the public on their recommendation. A courtesy copy of the final report is sent to the sponsor immediately prior to public release. The sponsor is not provided an opportunity to review the report or any portions of the report, or to suggest changes to the IOM report prior to its release. This stringent and established process was followed for both the Gulf War & Health and the Agent Orange reports.

In addition to those general procedures that are required by The National Academies, each committee also has procedures it follows in reviewing the data and drawing its conclusions. Each committee begins its deliberations by discussing and developing an approach to the committee’s statement of work. This statement of work has been approved by The National Academies governing body and has been included in the contract between the IOM and the study sponsor. However, in general these statements of work do not detail the specific approach to be used to complete the work, allowing the committee to use its expertise to identify the best approach. For the Gulf War & Health and Agent Orange reports committees needed to consider not only the statements of work but also the requirements of the legislation mandating the studies in developing approaches to how the committee would gather, review and evaluate the information it collects.

I can tell you from personal experience that the members of the IOM committees take their responsibility to assess the scientific data in a fair and unbiased manner very seriously. For each Gulf War and Agent Orange report, the expert committee members reviewed, evaluated and interpreted literally thousands of scientific publications that were identified through comprehensive searches of electronic databases such as those of the National Library of Medicine. On the basis of their analyses and deliberations, the committees reached consensus conclusions. Each committee prepared a consensus report outlining its findings which includes descriptions of the methods it used, the scientific information it reviewed, and the rationale for its conclusions.

By direction of the U.S. Congress, most IOM Gulf War studies have looked at chemical or biological agents or other possible deployment exposures and have drawn conclusions about what adverse health outcomes could be associated with or caused by those exposures. Similarly, the Veterans and Agent Orange studies look at specific chemical agents (Agent Orange and other herbicides) used during the Vietnam War and draw conclusions about what adverse health outcomes could be associated with or caused by those exposures. The conclusions are based on categories of evidence. In both cases, the legislation requests that the IOM committees make conclusions on the strength of the evidence for an association between exposure to certain agents and potential health outcomes. Successive Gulf War committees have decided to use the following five categories of association to describe the weight of the evidence and to make conclusions:

  • sufficient evidence of a causal relationship between an exposure and a health outcome,
  • sufficient evidence of an association between an exposure and a health outcome,
  • limited/suggestive evidence of an association,
  • inadequate/insufficient evidence to determine whether an association exists, and
  • limited/suggestive evidence of no association.

Those categories evolved from the categories used by the Agent Orange committees, which in turn were adapted from established categories of evidence used by the International Agency for Research on Cancer when it ranks evidence for chemicals that may cause cancer. The Agent Orange categories have gained wide acceptance over more than a decade by Congress, government agencies, researchers, and veterans groups.

The major difference between the categories used by the Gulf War committees and the ones used by the Agent Orange committees is the addition of the category of sufficient evidence of a causal relationship for all but one of the Gulf War committees. The additional category makes causation explicit and includes evidence beyond that found just in epidemiologic studies. Although association and causation are often used interchangeably they have different meanings scientifically. To demonstrate an association, the evidence simply must indicate that as exposure to an agent increases, the occurrence of an adverse outcome also increases. That an association is not the same as causality can be understood using the following example: fire trucks are associated with fires but they do not cause fires. For causation, the evidence must demonstrate that the exposure leads to the health outcome. For example, the influenza virus causes a person to get influenza. Therefore, the categories of evidence used by the first and subsequent Gulf War committees explicitly distinguish between causation and association.

One other change the Gulf War committees made was to clarify the definitions of Limited/Suggestive Evidence of an Association and Sufficient Evidence of an Association. The committee added the phrase “in human studies” to those definitions where they discuss “chance and bias, including confounding”. Chance, bias and confounding are much more significant problems in human epidemiology studies than in animal studies (which are more controlled). The addition of the statement about human studies simply clarifies that point. Although this phrase has been read to mean that the IOM studies have only addressed human studies, in reality both the Agent Orange studies and the Gulf War studies evaluate animal studies. This is quite evident when you read the reports and review the references that have been cited. At the same time, the IOM has put more weight on the human studies than on the animal studies. The Gulf War and Health volumes simply clarified that point, but conduct their studies in the same manner as the Agent Orange studies.

This leads me to address the issue of how animal data have been used by the Gulf War and the Agent Orange committees and why human studies have been given more weight. First, as might be expected, the published studies that are potentially relevant to the exposures evaluated by the Gulf War committees include studies that are conducted in animals. The committees looked at all relevant animal studies, including published reviews of the animal studies. However, many of the chemicals reviewed by the Gulf War committees have been tested in animals for decades in hundreds of studies and have well-established effects in animals that are described in basic toxicology text books. In such cases, committees have sometimes determined that it was not necessary to review all the individual animal studies that support those established effects but instead to cite reviews that summarize these specific well-established effects. Even in those instances where the health effects of an agent are well known, however, the committee still reviewed and described in their reports all of the animal studies that are critical to the committee’s conclusions.

Animal studies have been relevant and important but, there are limitations when drawing conclusions in humans on the basis of data in animals, which is why those studies were given less weight than human studies. Animal studies sometimes provide very different information than studies in humans. For example, vinyl chloride causes cancer in different organs in animals than in human; arsenic is a known human carcinogen but animals do not show similar tumors; and saccharine causes bladder tumors in male rats but not in humans. Using animals to look at human health effects is especially problematic for symptoms for which there are no diagnostic tests. A person can tell you that he or she has a headache, is tired, or just doesn’t feel very well, but a rat or mouse can not; by definition, such symptoms only can be seen in human studies. Therefore, the Gulf War committees have relied more on human studies, including epidemiologic and clinical studies, to reach conclusions regarding the association between an exposure to an agent and a health outcome. Animal data, when available, provide support for those conclusions.

Next I would like to briefly discuss what we know about the exposures in the Gulf War, and how that has affected the work of committees. The legislation that led to the Gulf War and Health studies lists a number of chemical and biological agents that the IOM was asked to consider. The number and diversity of those agents precluded all of the agents being reviewed by a single committee in a single report. The IOM held an open meeting with veterans and veteran service organizations to help identify the agents the veterans were most concerned about. On the basis of that meeting, the agents were prioritized for review.

All of the Gulf War committees have grappled with the issue of exposure and the lack of information, not only on how much of a chemical a person was exposed to, but even the specific chemicals a person might have been exposed to. For example, the committee could not find any information on which vaccines or medications, or the amount of a medication, that a specific person took during deployment. The committee members heard from veterans about being given a vaccination, for example en route to the combat arena, but they did not know what the vaccination was for, and the committee was told by the DOD that there are no records of who received what vaccinations. In other cases, when asked, veterans reported being exposed to a multitude of agents such as pesticides, pyridostigmine bromide, kerosene heaters, and oil well fire smoke during their deployment, but the levels of exposure to specific agents have not been determined and possibly never will be. This lack of information on exposure makes it very difficult to link a given health effect in veterans to a specific exposure.

Although most of the Gulf War committees looked at the health effects of the potential exposures, one of the committees was charged, as directed by the attached legislation, with evaluating Gulf War veterans’ health. This committee reviewed the published studies conducted on the Gulf War veterans themselves and made conclusions on the prevalence of health outcomes in the veterans. Because of the lack of exposure information, however, that report does not link health outcomes to specific exposures. An updated review of the literature on Gulf War veterans published since the preparation of that report is currently underway.

With that, I would once again like to thank you for inviting me to testify before this Subcommittee. I appreciate the work of this Subcommittee on Oversight and Investigations of the House Committee on Veterans' Affairs. On behalf of all IOM Gulf War committee members past and present I thank you for your trust in our ability to assist you with this important work for our nation’s veterans. I know from my service on these committees that the nation’s scientists are happy to serve, and look to you for guidance on how we can be of most assistance to you and the VA in assessing health impacts of Gulf War deployment. I look forward to answering any questions you might have.


An archived webcast of the hearing can be found on the House Veterans' Affairs Committee's Web site.