101st Congress, 2d Session -

Union Calendar No. 415

- House Report 101-672
THE AGENT ORANGE COVERUP: A CASE OF FLAWED
SCIENCE AND POLITICAL MANIPULATION
TWELFTH
REPORT
BY THE
COMMITTEE
ON GOVERNMENT
OPERATIONS
together with
DISSENTING VIEWS
AUGUST 9, 1990.-Committed
to the Committee
of the Whole House on the
State of the Union and ordered to be printed
U.S. GOVERNMENT
PRINTING
OFFICE
32-302
WASHINGTON:
1990
COMMITTEE ON GOVERNMENT OPERATIONS
JOHN CONYERS, JR., Michigan, Chairman
CARDISS COLLINS, Illinois
GLENN ENGLISH, Oklahoma
HENRY A. WAXMAN, California
TED WEISS, New York
MIKE SYNAR, Oklahoma
STEPHEN L. NEAL, North Carolina
DOUG BARNARD, JR., Georgia
BARNEY FRANK, Massachusetts
TOM LANTOS, California
ROBERT E. WISE, JR., West Virginia
BARBARA BOXER, California
MAJOR R. OWENS, New York
EDOLPHUS TOWNS, New York
JOHN M. SPRATT, JR., South Carolina
BEN ERDREICH, Alabama
GERALD D. KLECZKA, Wisconsin
ALBERT G. BUSTAMANTE, Texas
MATTHEW G. MARTINEZ, California
NANCY PELOSI, California
DONALD M. PAYNE, New Jersey
JIM BATES, California
GARY A. CONDIT, California
DENNIS E. ECKART, Ohio
FRANK HORTON, New York
WILLIAM F. CLINGER, JR., Pennsylvania
AL McCANDLESS, California
HOWARD C. NIELSON, Utah
RICHARD K. ARMEY, Texas
DONALD E. "BUZ" LUKENS, Ohio
J. DENNIS HASTERT, Illinois
JON L. KYL, Arizona
CHRISTOPHER SHAYS, Connecticut
PETER SMITH, Vermont
STEVEN SCHIFF, New Mexico
CHUCK DOUGLAS, New Hampshire
C. CHRISTOPHER COX, California
CRAIG THOMAS, Wyoming
ILEANA ROS-LEHTINEN, Florida
JULIAN EPSTEIN, Staff Director
DONALD W. UPSON, Minority Staff Director
HUMAN RESOURCES AND INTERGOVERNMENTAL RELATIONS SUBCOMMITTEE
TED WEISS, New York, Chairman
HENRY A. WAXMAN, California RICHARD K. ARMEY, Texas
NANCY PELOSI, California PETER SMITH, Vermont
DONALD M. PAYNE, New Jersey ILEANA ROS-LEHTINEN, Florida
Ex OFFICIO
JOHN CONYERS, JR., Michigan FRANK HORTON, New York
JAMES R. GOTrLIEB, Staff Director
MARC SMOLONSKY, Professional Staff Member
PAMELA H. WELCH, Clerk
WAYNE CIMONS, Minority Professional Staff
LETTER OF TRANSMITTAL
HOUSE OF REPRESENTATIVES,
Washington, DC, August 9, 1990.
Hon. THOMAS S. FOLEY,
Speaker of the House of Representatives,
Washington, DC.
DEAR MR. SPEAKER: By direction of the Committee on Govern-
ment Operations, I submit herewith the committee's twelfth report
to the 101st Congress. The committee's report is based on a study
made by its Human Resources and Intergovernmental Relations
Subcommittee.
JOHN CONYERS, Jr., Chairman.
(III)
CONTENTS
Page
I. E xecutive sum m ary ............................................................................................... 1
II. Introduction ............................................................................................................... 3
III. B ackgroun d ............................................................................................................... 4
IV . F in dings ...................................................................................................................... 8
A. The CDC Agent Orange exposure study should not have been canceled
because it did not document that exposure of veterans to the herbi-
cide could not be assessed, nor did CDC explore alternative methods
of determ ining exposure .................................................................................. 8
B. The original protocol for the CDC Agent Orange study was changed to
the point that it was unlikely for the heaviest exposed soldiers to be
iden tified ........................................................................................................... . 14
C. The blood serum analysis, which was used as proof by CDC that an
Agent Orange exposure study could not be conducted, was based on
erroneous assumptions and a flawed analysis ............................................ 21
D. The White House compromised the independence of the CDC and
undermined the study by controlling crucial decisions and guiding
the course of research at the same time it had secretly taken a legal
position to resist demands to compensate victims of Agent Orange
exposure and industrial accidents ................................................................. 27
E. The Federal Government has suppressed or minimized findings of ill
health effects among Vietnam veterans that could be linked to Agent
O range exposure ..............................................................................................
33
V . R ecom m endations ...................................................................................................
. 38
A. Congress should require the Department of Defense to create an Agent
O range exposure index ....................................................................................
38
B. When an adequate exposure index is developed, the Federal Govern-
ment should contract through the National Academy of Sciences for
a private, independent epidemiological study matching the health
outcomes of Vietnam veterans against the exposure index ..................... 38
C. All scientific research conducted by Federal agencies should be done
without interference from Federal components outside their respec-
tive agen cies .....................................................................................................
38
VIEWS
Dissenting views of Hon. Richard K. Armey, Hon. Frank Horton, Hon.
Howard C. Nielson, Hon. J. Dennis Hastert, Hon. Jon L. Kyl, and Hon.
C h uck D ou glas ..............................................................................................................
39
Union Calendar No. 415
101ST CONGRESS
I REPORT
2d Session HOUSE OF REPRESENTATIVES 101-672
THE AGENT ORANGE COVERUP: A CASE OF FLAWED
SCIENCE AND POLITICAL MANIPULATION
AUGUST 9, 1990.-Committed to the Committee of the Whole House on the State of
the Union and ordered to be printed
Mr. CONYERS, from the Committee on Government Operations,
submitted the following
TWELFTH REPORT
together with
DISSENTING VIEWS
BASED ON A STUDY BY THE HUMAN RESOURCES AND
INTERGOVERNMENTAL RELATIONS SUBCOMMITTEE
On August 2, 1990, the Committee on Government Operations
approved and adopted a report entitled "The Agent Orange Cover-
up: A Case of Flawed Science and Political Manipulation." The
chairman was directed to transmit a copy to the Speaker of the
House.
I. EXECUTIVE SUMMARY
The Human Resources and Intergovernmental
Relations Subcom-
mittee conducted a year-long investigation of the studies of Agent
Orange exposure and Vietnam veterans' health conducted by the
Centers for Disease Control (CDC) from 1982 to 1987. The exposure
study was intended to be the first comprehensive effort by the Fed-
eral Government to determine the magnitude of Agent Orange ex-
posure to U.S. military personnel
serving in Vietnam during the
war years. The separate health review is the largest medical study
of Vietnam veterans undertaken
by the Federal Government.
The subcommittee's
investigation
included
an extensive
review of
documents
related to the studies at the CDC Center for Environ-
mental Health and Injury Control in Atlanta,
Georgia. Interviews
were conducted of current and former CDC employees. The records
of a White House panel, the Agent Orange Working Group
(AOWG), were obtained from the U.S. National Archives and the
Reagan Presidential Library, and reviewed by subcommittee staff.
Interviews were also conducted of military records experts, epide-
miologists, and environmental authorities employed by Federal
agencies, veterans service organizations, and universities.
Agent Orange, so named for the orange stripe on the 55-gallon
drums in which it was stored, was a herbicide used by the U.S.
military during the Vietnam War to defoliate forests that con-
cealed the activities of the enemy. It was a mixture of two chemical
ingredients, the n-butyl esters of 2,4-D and 2,4,5-T. One of these
chemicals, 2,4,5-T, contained 2,3,7,8-tetrachlorodibenzo-p-dioxin.
Known simply as dioxin, the chemical is considered one of the most
toxic chemicals every created synthetically, and has proven to be
carcinogenic in animal studies.
The study had been mandated by Congress more than a decade
ago. The Veterans Health Programs Extension and Improvement
Act of 1979 authorized the Veterans Administration (VA) to design
and conduct a study to assess the exposure of Vietnam veterans to
Agent Orange. In 1981, Congress enacted the Veterans Health
Care, Training, and Small Business Loan Act, which expanded the
scope of the VA study to include a health study of Vietnam veter-
ans, using service during the war as an exposure factor, exposure
to Agent Orange notwithstanding.
The CDC Agent Orange Exposure Study was canceled in 1987
when a White House panel, the AOWG, concluded that military
records could not be used to assess exposure to the herbicide. The
health status study was published in 1989, and a final study on se-
lected cancers among Vietnam veterans was released in 1990.
The committee finds that the Agent Orange exposure study
should not have been canceled because CDC did not document that
exposure could not be assessed, nor did it explore alternative meth-
ods of determining exposure. The committee concludes that other
methods of determining exposure were available to CDC, but inten-
tionally disregarded.
The committee's report concludes that the CDC study was
changed from its original format so that it would have been unlike-
ly for the soldiers who received the heaviest exposure to the herbi-
cide to be identified. CDC accomplished this by unjustifiably dis-
crediting the military records provided to it by the Department of
Defense's Environmental Study Group (ESG). The key alteration
was CDC's decision to track troop movements using broad battalion
records, which are less precise than the company-level records the
Agency had originally intended to use to follow troop positions
during the Vietnam War.
After restricting the study by eliminating thousands of veterans
who would have been most likely to have been exposed to Agent
Orange, CDC attempted to validate its flawed exposure definitions
with a test to identify traces of dioxin in the blood of the veterans
who had been studied. When the blood test failed to find elevated
levels of dioxin in the veterans' blood, the White House instructed
CDC to halt the study. The committee found that the blood test
was based on flawed and manipulated estimates of the ability of
dioxin to remain in human blood 20 years after exposure. The
weight of scientific evidence indicates that dioxin cannot be traced
in all veterans who were exposed to Agent Orange two decades ago.
The committee also concludes that the CDC study was controlled
and obstructed by the White House, primarily through its AOWG
and the Office of Management and Budget (OMB), because the
Reagan administration had adopted a legal strategy of refusing li-
ability in military and civilian cases of contamination involving
toxic chemicals and nuclear radiation.
The report's final conclusion is that the Federal Government has
suppressed or minimized findings of ill health effects among Viet-
nam veterans that could be linked to Agent Orange exposure. Fed-
eral studies have found that Vietnam veterans are more suscepti-
ble to diseases such as non-Hodgkin's lymphoma and soft tissue
sarcoma. Studies have also concluded that the offspring of Vietnam
veterans suffer higher incidences of certain birth defects, such as
cerebrospinal malformations, and that Vietnam veterans have
sperm malformations and low sperm counts. The Federal Govern-
ment has consistently refused to acknowledge any link between
these diseases and Agent Orange, even though the maladies are
known consequences of herbicide exposure in the civilian work-
place.
The committee recommends that Congress mandate the creation
of an accurate Agent Orange exposure index, and that the index be
used, through a contract to a private organization, to conduct an
independent epidemiological study of the effects of the herbicide on
Vietnam veterans. The committee also recommends that the White
House be barred from interfering with all Federal scientific re-
search.
II. INTRODUCTION
Under the House of Representatives' Rule X, 2(b)(2), the Commit-
tee on Government Operations is authorized to "review and study,
on a continuing basis, the operation of Government activities at all
levels with a view to determining their economy and efficiency."
The committee has assigned this responsibility, as it pertains to the
Department of Health and Human Services (HHS) and, its compo-
nent, the Centers for Disease Control (CDC), to the Human Re-
sources and Intergovernmental Relations Subcommittee.
Pursuant to its authority, the subcommittee conducted a two-
year investigation of the studies of Agent Orange exposure and
Vietnam veterans' health conducted by CDC from 1982 to 1987 and
other studies of herbicides as a cause of disease. The exposure
study was intended to be the first comprehensive effort by the Fed-
eral Government
to determine the magnitude of Agent Orange ex-
posure to U.S. military personnel
serving in Vietnam during the
war years. The separate health review is the largest medical study
of Vietnam veterans undertaken
by the Federal Government.
The subcommittee's
investigation
included an extensive
review of
documents
related to the two studies at the CDC Center for Envi-
ronmental
Health and Injury Control in Atlanta,
Georgia. Inter-
views were conducted of current and former CDC employees
who
worked on the studies. The records of a White House panel, the
Agent Orange Working Group (AOWG), were obtained from the
U.S. National Archives and the Reagan Presidential Library, and
reviewed by subcommittee staff. Interviews were also conducted of
military records experts, epidemiologists, and environmental au-
thorities employed by Federal agencies, veterans service organiza-
tions, and universities.
The subcommittee's investigation included two public hearings,
which were conducted on July 11, 1989,1 and June 26, 1990.2 The
following witnesses testified at the first hearing: Honorable Lane
Evans, a Representative in Congress from the State of Illinois;
Lawrence B. Hobson, M.D., Ph.D., Director, Office of Environmen-
tal Medicine, Department of Veterans' Affairs, accompanied by
Donald Ivers, acting general counsel, and Fred Conway, special as-
sistant to the general counsel; Vernon N. Houk, M.D., Director,
Center for Environmental Health and Injury Control, Centers for
Disease Control; Michael Kafrissen, M.D., former medical epidemi-
ologist, Centers for Disease Control; Philip J. Landrigan, M.D., di-
rector, Division of Environmental and Occupational Medicine,
Mount Sinai School of Medicine; Ellen K. Silbergeld, Ph.D., direc-
tor, Toxic Chemicals Program, Environmental Defense Fund;
Dennis M. Smith, M.D., former visiting scientist, Centers for Dis-
ease Control; John F. Sommer, Jr., director, National Veterans Af-
fairs and Rehabilitation Commission, the American Legion; Jeanne
M. Stellman, Ph.D., epidemiologist, American Cancer Society;
Steven D. Stellman, Ph.D., Assistant Commissioner for Biostatistics
and Epidemiologic Research for the Department of Health of the
city of New York; and Mary R. Stout, president, Vietnam veterans
of America.
The following witnesses testified at the second hearing: Richard
W. Clapp, M.P.H., Sc.D., director, Environmental Health Studies,
JSI Research and Training Institute; Peter C. Kahn, Ph.D., Associ-
ate Professor of Biochemistry, Rutgers University; Arnold Schecter,
M.D., M.P.H., Professor of Preventative Medicine, State University
of New York Health Science Center; Ellen K. Silbergeld, Ph.D.,
senior scientist, Environmental Defense Fund; Daniel Thau Teitel-
baum, M.D., president, Medical Toxicology Partnership; Shelia
Hoar Zahm, Sc.D., epidemiologist, National Cancer Institute; and
Admiral Elmo R. Zumwalt, Jr., USN Ret., Special Assistant to the
Secretary of Veterans Affairs.
III. BACKGROUND
Ageht Orange, so named for the orange stripe on the 55-gallon
drums in which it was stored, was a herbicide used by the U.S.
military during the Vietnam War to defoliate forests that con-
cealed the activities of the enemy. It was a mixture of two chemical
ingredients, the n-butyl esters of 2,4-D and 2,4,5-T. One of these
chemicals, 2,4,5-T, contained 2,3,7,8-tetrachlorodibenzo-p-dioxin.
1 Hearing before a subcommittee of the Committee on Government Operations, U.S. House of
Representatives, "Oversight Review of CDC's Agent Orange Study," July 11, 1989, hereinafter
referred to as "Hearing, 1989."
2 Hearing before a subcommittee of the Committee on Government Operations, U.S. House of
Representatives, "Links Between Agent Orange and Other Herbicides With Diseases," June 26,
1990, hereinafter referred to as "Hearing, 1990."
Known simply as dioxin, the chemical is considered one of the most
toxic chemicals ever created synthetically, and has proven to be
carcinogenic in animal studies.
According to the Office of Environmental Medicine in the De-
partment of Veterans' Affairs, 15 different herbicides were used in
Vietnam between January 1962 and September 1971. More than 80
percent of the defoliation operations used Agent Orange, which was
sprayed primarily between January 1965 and April 1970. Among
the other herbicides used prominently were Agent White, which
also contained 2,4-D, and Agent Blue, which consisted of an organic
arsenic compound called cacodylic acid. Less than 7 percent of the
total acreage sprayed during the Vietnam War was covered prior
to 1965. Most of the spraying occurred from 1966 to 1969. Overall,
more than 20 million gallons of herbicides were sprayed upon 6
million acres. More than 3.5 million acres were sprayed more than
once.
Herbicides were used for three main purposes during the Viet-
nam War. The principal uses were to defoliate military areas to im-
prove observation, and to destroy crops to diminish the enemy's
food supply. A third purpose was to clear vegetation around fire
bases and such installations, around landing zones, and along lines
of communication. Herbicide usage began in 1962, was expanded in
1965 and 1966, and peaked from 1967 to 1969. Herbicides were
phased out in 1970 when it was learned that mice exposed to the
chemical components of the herbicides bore offspring with birth de-
fects.
Most of the herbicides were sprayed from fixed-wing aircraft in a
project called, "Operation Ranch Hand." But herbicides were also
sprayed from helicopters, trucks, riverboats, and hand applicators.
Spraying occurred in all military zones of Vietnam. The most
heavily sprayed areas included inland forests near the demarcation
zone, inland forests at the junction of the borders of Cambodia,
Laos, and South Vietnam, inland forests north and northwest of
Saigon, mangrove forests on the southern-most peninsula of Viet-
nam, and mangrove forests along major shipping channels south-
east of Saigon. Crop destruction missions were concentrated in
northern and eastern central areas of South Vietnam.
Many Vietnam veterans believe their exposure to Agent Orange
has caused a variety of ailments, including cancer, birth defects,
and emotional disorders linked to neurological dysfunction.
Their
concerns are heightened by what science has discovered about the
possible effects of dioxin exposure but, also, by what is not known
about the chemical's effects on humans.
Dioxin is the useless, unwanted
byproduct
from the production
and use of chlorinated phenols. Environmental
hazards have also
been caused by toxic compounds
related to dioxin, such as polychlo-
rinated biphenyls (PCB's), polybrominated
biphenyls (PBB's), poly-
chlorinated
dibenzofurans
(PCDF's),
polychlorinated
naphthalenes,
and polychlorinated terphenyls.
Dioxin
is omnipresent,
existing
in household
products,
dust parti-
cles, and water. It has been found in significant
levels across the
world. Millions of people have been exposed to it through industrial
accidents,
fly ash from waste incinerators,
herbicide
spraying,
man-
ufacturing
plants,
and even in some edible
fish.
There is no disagreement among scientists that dioxin in its pure
form is the most toxic synthetic chemical in existence. Fortunately,
humans are never exposed to undiluted amounts of the substance.
It occurs as trace quantities in the ranges of parts per billion or
parts per million as a contaminant in other products. Sometimes-
in the case of fish, for example-it occurs in parts per trillion, or,
in the case of water, parts per quadrillion.
Animal experiments have found dioxin toxicity to be dose-relat-
ed. At low doses, there are no observed toxic effects of dioxin in
animals. However, in high doses, various species have developed
such diseases as hepatic necrosis and T-cell depression. Increased
incidences of cancer were observed in most species exposed to
dioxin during laboratory experiments.
Corresponding dose-related research cannot be legally or morally
conducted on humans. However, there is some limited information
about the effects of dioxin on humans as a result of inadvertent ex-
posure from industrial accidents. A recent accident occurred in
Seveso, Italy, when a cloud of trichlorophenol and dioxin spread
over seven square miles of populated territory. In another accident,
transformer fluid was inadvertently mixed with rice oil, affecting
seven thousand people who ingested the toxic oil in Japan and
Taiwan. Similar accidents occurred in West Virginia, New Jersey,
and England. The most severe accidents to date occurred in West
Germany and Czechoslovakia. The most well-known dioxin inci-
dents in the United States probably were the Times Beach, Missou-
ri, and Love Canal, NY, disasters, when dioxin from an industrial
plant had seeped into the soil of residential communities.
The most serious health consequences of these accidents were
neuropathy and liver damage in approximately one-third of the vic-
tims of the incidents in West Germany and Czechoslovakia. There
were similar consequences of the West Virginia and New Jersey ac-
cidents. The most prevalent reaction to dioxin contamination-
which may be a marker of exposure and a warning sign for disease
that can take decades to develop-is the skin condition, chloracne,
a painful and extremely annoying ailment, but not life threaten-
ing. Chloracne has occurred in workers and residents exposed
during most industrial accidents where high levels of dioxin were
released.
Some studies of industrial accidents involving dioxin have found
incidences of soft tissue sarcoma, a rare form of cancer, in those
exposed to the chemicals, but some in the scientific establishment
consider the numbers of study subjects too small to carry the
weight of significance. The U.S. Centers for Disease Control, in a
recent study of Vietnam veterans, found that veterans who had
served in Vietnam had significantly higher incidences of non-Hodg-
kin's lymphoma, another rare illness, but dismissed any possible
connection to Agent Orange exposure, instead claiming it did not
know why the veterans had higher levels of the disease. In March
1990, the CDC report prompted the Department of Veterans' Af-
fairs to declare non-Hodgkin's lymphoma a monetarily compensa-
ble, service-connected disease, but the Department did not link the
illness to Agent Orange exposure. In May 1990, the Department
also declared soft-tissue sarcoma, a rare cancer, to be a service-con-
nected, compensable malady for Vietnam veterans on the basis of a
report by the Department's Veterans Advisory Committee on Envi-
ronmental Hazards, which found that herbicides may have caused
the disease.
The rationale behind the U.S. Government's refusal to accept
any causal effects between Agent Orange and all but a few diseases
stems from the CDC exposure study, which was canceled when the
Agency declared such a study to be scientifically impossible.
The study had been mandated by Congress more than a decade
ago. The Veterans Health Programs Extension and Improvement
Act of 1979 authorized the Veterans Administration (VA) to design
and conduct a study to assess the exposure of Vietnam veterans to
Agent Orange. In 1981, Congress enacted the Veterans Health
Care, Training, and Small Business Loan Act, which expanded the
scope of the VA study to include a health study of Vietnam veter-
ans, using service during the war as an exposure factor, exposure
to Agent Orange notwithstanding.
The VA's efforts to conduct the study were mired in delays. In
1982, three years after passage of the Agent Orange study legisla-
tion, the agency had not even begun the study. The House and
Senate Veterans Affairs Committees decided to transfer the study
from the VA to the CDC, after the Director of the CDC Center for
Environmental Health and Injury Control testified at a hearing
that CDC could do the study better and faster than the VA. Less
than a year later, Congress passed an appropriations bill contain-
ing $54 million, to be provided through the VA's budget, for CDC to
conduct the Agent Orange study. Approximately $43 million of the
appropriated funds were expended before the study was canceled.
CDC, like the VA before it, had two missions. One, to design and
implement a study to measure the exposure of Vietnam veterans to
Agent Orange, and the second to assess the overall health of Viet-
nam veterans. The latter mission was divided into two studies. The
first study involved conducting physical examinations of the sub-
jects and reviewing their medical records. The second part was a
study of the incidence of selected cancers among Vietnam veterans.
In January 1989, CDC published the results of the separate
health assessment, called the Vietnam Experience Study (VES), in
five volumes. CDC claimed the study found little differences be-
tween the two cohorts involved: veterans who served in Vietnam
and veterans who did not during the same time period. In early
1990, CDC published the Selected Cancers Study, which found that
Vietnam veterans were at greater risk of one cancer, non-Hodg-
kin's lymphoma.
The exposure study was intended to be a landmark review. For
the first time, formal scientific research would be conducted to de-
termine which soldiers were in areas sprayed with Agent Orange.
The exposure assessment
study was canceled in September
1987,
after the White House AOWG contended
that it was scientifically
impossible
to assess the exposure
of Vietnam
veterans
to Agent
Orange due to the inadequacy
of military records. The first and
only attempt by the Federal Government
to determine
the expo-
sure to Agent Orange by U.S. military personnel
never reached
a
conclusion,
other than that the study was impossible
to do. The
question
foremost
in the minds of Vietnam
veterans
and the Con-
gress in regard to Agent
Orange
has still not been answered.
IV. FINDINGS
A. THE CDC AGENT ORANGE EXPOSURE STUDY SHOULD NOT HAVE BEEN
CANCELED BECAUSE IT DID NOT DOCUMENT THAT EXPOSURE OF VET-
ERANS TO THE HERBICIDE COULD NOT BE ASSESSED, NOR DID CDC EX-
PLORE ALTERNATIVE METHODS OF DETERMINING EXPOSURE
In its November 1987 final report, Comparison of Serum Levels of
2,,7,8-TCDD With Indirect Estimates of Agent Orange Exposure in
Vietnam veterans, the CDC concluded:
The findings of this study and the conclusions from the
AOWG [Agent Orange Working Group of the White
House] Science Sub-Panel report on exposure assess-
ment * * * do not identify any method for utilizing mili-
tary records or self-reported exposure to distinguish be-
tween U.S. Army ground combat troops who were and
were not exposed to Agent Orange in Vietnam, as would
be needed for a cohort study of possible health effects.
3
This conclusion was used as the basis by the White House and
CDC to cancel the study and end all Federal attempts to study the
link between Agent Orange and the health problems of Vietnam
veterans.
4
The law directing the U.S. Government to assess the exposure of
veterans to Agent Orange was not specific about the method of
study, only that a study be conducted. The final protocol for the
study was left to the discretion of CDC, with the Congressional
Office of Technology Assessment (OTA) and the Institute of Medi-
cine (IOM) serving as peer reviewers. CDC could only abandon the
study if no method to assess exposure could be identified at all.
CDC reached just such a conclusion. The study was canceled when
CDC and the White House Panel decided that no methods of any
kind existed to use military records to distinguish between exposed
and unexposed soldiers who served in Vietnam.
The subcommittee's investigation of the CDC study has estab-
lished that other methods existed to assess exposure of veterans to
Agent Orange. Therefore, CDC erred when it reported that it could
not identify any method of determining exposure. Evidence com-
piled by the subcommittee demonstrates that other methods were
available, but were ignored or dismissed by CDC.
CDC's mistake was described by Dr. Philip J. Landrigan, one of
the foremost epidemiologists in the United States, who stated that
the Agency had ignored a large group of veterans who had been
potentially exposed to Agent Orange. He testified that the attempt
to determine Agent Orange exposure in Vietnam veterans was
analogous to an earlier effort by CDC to determine exposure to
' "Comparison of Serum Levels of 2,3,7,8-TCDD With Indirect Estimates of Agent Orange Ex-
posure in Vietnam veterans," Final Report, Agent Orange Projects, Division of Chronic Disease
Control, Center for Environmental Health and Injury Control, Centers for Disease Control,
Public Health Service, U.S. Department of Health and Human Services, November 1987, pp. 4-
5.
4 Hearing, 1989, pp. 55-57. Letter from James 0. Mason, M.D., Dr. P.H., Assistant Surgeon
General, Director, Centers for Disease Control, to Don Newman, Chair, Domestic Policy Council
Agent Orange Working Group, October 28, 1987.
PBB, a toxic chemical related to dioxin, following an industrial ac-
cident in Michigan.
In response to the disaster, just as in response to the
Agent Orange episode in Vietnam, a series of epidemiolog-
ic studies were undertaken. The first of these studies was
an investigation that was undertaken by our group at CDC
in which I was the coprincipal investigator....
We conducted a survey in which we did physical exami-
nations, took blood samples on a very carefully targeted
list of people in Michigan who we identified through
records of contaminated farms. In theses populations of
people who were very carefully targeted, we found a very
high proportion who had elevated serum levels of PBB.
Their levels were roughly in proportion to the extent of
their exposure.
At about the same time Dr. Irving Selikoff ... took this
group into Michigan and did a statewide survey to deter-
mine the proportion of people in Michigan who had been
exposed to PBB. He found... that only a small proportion
of the people whom he tested had elevated serum levels of
PBB.
Now, Dr. Selikoff could have said on the basis of those
findings that most residents of Michigan are not exposed
to PBB and that statewide records do not constitute a good
basis for identifying people in Michigan who have been ex-
posed to this toxic compound. In other words, he could
have drawn conclusions that would have been parallel in
their content to the conclusions that CDC drew with
regard to the exposure to dioxin in Vietnam.
He didn't do that, though. He recognized the possibility
that there might exist a heavily exposed subgroup within
the population .... and he urged the need for further re-
search. And, in fact that research still continues today.
* * * * *
I think the analogy is obvious and I think the question
that confronts us today . . . is the question of whether
there might exist within the population of millions of sol-
diers and marines and sailors and Air Force men who
went to Vietnam a heavily exposed subgroup comprising
maybe 5 percent, maybe 10 percent of the total, who were
heavily exposed. There are several lines of evidence which
suggest
that
such
groups
exist.
5
In its final report, CDC contended that available military records
regarding troop movements and the spraying of Agent Orange
were insufficient to identify veterans who had been exposed to the
defoliant; the purported inadequacy of the military records was the
primary reason the exposure study was canceled and termed im-
possible to complete.
The final CDC report stated:
SHearing, 1989. Testimony of Philip J. Landrigan, M.D., Director, Division of Environmental
and Occupational
Medicine, Mount Sinai School of Medicine, New York, NY, pp. 221-223.
. * At best, available military data permit only a probabi-
listic assessment of opportunity for exposure. Accurate in-
dividual exposures to Agent Orange cannot be determined
because there is limited knowledge about the initial disper-
sion, biodegradation, and eventual ecologic disposition of
Agent Orange in Vietnam; uncertainty about the absorp-
tion of Agent Orange or TCDD in humans; and the inabil-
ity to determine the precise location of individual soldiers
in relation to Agent Orange applications. Because of these
uncertainties, the November 1983 protocol for the Agent
Orange Study indicated that further pilot testing would be
needed before a decision could be made to proceed with the
full-scale study. . . . military records alone could not be
used to estimate possible exposure of individual men to
Agent Orange. In addition, some reviewers were concerned
that too few men were identified as having "meaningfully
high" exposure opportunity scores to warrant proceeding
with a full-scale
study.
6
The CDC conclusion that military records could not be used to
identify exposed veterans has been disputed by former CDC scien-
tists, Department of Defense records experts, and prominent pri-
vate sector epidemiologists.
Dr. Jeanne Stellman of Columbia University, who conducted an
epidemiologic study for the American Legion, testified before the
subcommittee that available records were quite sufficient to deter-
mine exposure to Agent Orange.
. . We are talking about 11,197,929 gallons sprayed on
Vietnam that we know the locations of pretty well. In ad-
dition, Richard Christian's 7 group then picked up an addi-
tional 13 percent of what is now the total ... tape records
of 1,593,380 gallons of herbicides given off by fixed-wing
aircraft, helicopters, and ground equipment such as back-
packs, and something that everybody's always referring to
[as] the unknown sources, but which, in fact, only repre-
sents 20,000 gallons.
So this data that you heard being disregarded and tossed
out as useless to evaluate veteran's health and well-being,
in fact, is the tape that was developed and used by the Na-
tional Academy of Sciences ... we have here data on more
gallonage and more usage of herbicides than I would say
that we have in anyplace else including all the usage in
the United States which we don't have a good handle on.
8
Dr. Dennis Smith, a scientist who worked on the CDC Agent
Orange Project, was one of the people responsible for evaluating
records provided to CDC by the Department of Defense. He testi-
fied that the CDC staff had developed numerous alternatives to the
approach the agency used to assess exposure, and that he and
others at CDC believed sufficient information existed to compen-
' Op. cit., p. 9. See footnote 3.
7 Richard Christian directed the Department of Defense's Environmental Support Group,
which supplied Agent Orange spray records and military troop movement records to CDC.
' Hearing, 1989, p. 158.
sate for gaps in the military records transmitted by the Pentagon.
But, he testified, his superiors were resistant to exploring the alter-
natives.
We didn't really inform people what we were doing. We
should have publicized it to a greater extent and gotten
more feedback on it. We had tremendous problems with
developing exposure indices but they were not in creating
new indices. We had many new indices and many different
things to try and experiment with. But, we had problems
in selling these indices to our superiors and people at the
Office of Technology Assessment.
9
Smith also testified that he believed CDC officials blamed certain
gaps of information in the military records to cover up the prob-
lems CDC was having in identifying dioxin in the bodies of Viet-
nam veterans.
Considerable concern arose over the scientific problems
that plagued the Agent Orange Study, and an attempt was
made to divert focus from these methodological issues to
other areas that had origin or direction outside of Centers
for Disease Control. The first, and most useful diversion,
was to the fact that improper location data would destroy
the usefulness of the exposure index. Regardless of any
utility the data might have, if it was not "complete" (i.e. a
troop location for every single unit on every single day)
then it would be the troop location data that introduced
the major error into the exposure index. Previous concepts
of gap-filling, imputation, verification of troop locations,
obtaining locations from other sources, etc. were no longer
valid; the utility contained in the data already available
from ESG [Environmental Support Group of the Defense
Department] was ignored, and concepts developed that ini-
tially supported the data, such as "digit preference," were
now directed against the ESG data.'"
At one point during the performance of the study, Dr. Smith
sought data from outside the Environmental Support Group to but-
tress the military records already obtained by CDC. On May 14,
1985, he sent a memorandum
to the Chief Scientist of the CDC
Agent Orange Project, detailing a visit to the Military History In-
stitute and U.S. Army War College in Carlisle, Pennsylvania.
t * The historical
records
available
at Carlisle
appear
to be of tremendous
importance
to our Agent Orange
Projects in respect to several aspects of our interest in the
Southeast
Asia conflict. To wit: (1) Troop location; (2) Her-
bicide Missions;
(3) Medical
Information;
(4) Possible
models of troop movement;
(5) Computer
operations
used
during the time period of interest.
The records provide ex-
panded
knowledge
on these areas and help facilitate
an
understanding
of the situations
in Southeast
Asia that
9 Ibid. Testimony of Dennis M. Smith, M.D., former visiting scientist, Centers for Disease Con-
trol, p. 80.
10 Ibid., pp. 93-94.
might affect our projects. These records are great in
number and generally are not indexed: researching a par-
ticular subject may be quite time-consuming, but in my
"clinical impression,"
may be generally
productive.
1
In 1985, officials at the Defense Department's ESG discovered
that CDC was disputing the accuracy of the military records. The
ESG Chief of the CDC Support Branch, who was responsible for
transmitting the records to CDC, warned that CDC did not under-
stand how to use the records and appeared to be ignoring ESG's
advice. The Chief noted that gaps in military tracking records
could be filled with supplementary documents, such as company
morning reports. In a memorandum to the Director of ESG, the
CDC Support Branch Chief described how CDC had disputed, with-
out justification, some records provided by ESG. Quoting from a
CDC internal report, the Chief wrote:
... ESG claimed that additional documents exist that
might help place companies on a daily basis. ESG has not
claimed anything, this statement is fact. Who are these ex-
perts who make decisions based on a computer distance
comparison? For instance, the example CDC gives for Cu
Chi taken from the morning reports substantiates the ac-
curacy of the APO's that are listed. The grid coordinate
and the APO number show the location as Cu Chi. What
evidence does CDC have that indicates this unit was not at
Cu Chi? This information was taken from official Army ar-
chival records. It must be stated for the record, that a
morning report is a company document not a brigade or
division
record.
'
2
The memorandum also expressed concerns that CDC was ignor-
ing ESG's advice and preparing its own, incorrect, analyses of the
military records. For example, CDC had dismissed the use of com-
pany morning reports to fill in missing locations in the main data
base.
The ESG memorandum states:
. . . Our main concern is that CDC does not keep ESG ad-
vised on these types of analysis until after the fact. Our
expertise is U.S. Army combat unit records and troop
movements, yet we have no idea if the proper grid coordi-
nate locations were being compared. We feel a more de-
tailed analysis needs to be performed on this information
before a decision is made not to use morning report data. 13
Richard Christian, who directed the ESG during the CDC study,
testified that he had informed CDC on numerous occasions that
records were available to address all of CDC's concerns about gaps
11 "Site Visit Summary, Carlisle Barracks: Military History Institute and US Army War Col-
lege," Memorandum from Dennis M. Smith, M.D., visiting scientist, Centers for Disease Control,
to Daniel L. McGee, Ph.D., Chief Scientist, Agent Orange Projects, Centers for Disease Control,
May 14, 1985, p.
1
.
12 "Centers for Disease Control Report to OTA," Memorandum for the Director from Donald
C. Hakenson, Chief, CDC Support Branch, Department of the Army, U.S. Army and Joint Serv-
ices Environmental Support Group, October 25, 1985, p. 1.
" Ibid., p. 2.
in the data base. Yet, CDC reported to OTA that sufficient records
were not available to complete the exposure study. Christian testi-
fied:
I was investigated by the National Academy of Sci-
ences on two occasions. The Science Panel of the White
House Agent
'
Orange Working Group was at my office nu-
merous times. There were visits from the OTA to look at
these records and it's just inconceivable to us and to the
Environmental Support Group that they could say that
those records were not adequate to do an Agent Orange
study.
1 4
Christian's testimony raised questions about the ability of the
CDC staff to sufficiently comprehend and use the military records.
He testified about being astounded when he discovered that CDC
was identifying enemy locations when it thought it had pinpointed
U.S. sites.
At one point, the Centers for Disease Control attempted
to take over the work of ESG from the study. My staff pro-
vided the CDC with copies of daily journals. In a test of
that exercise, the personnel in the Centers for Disease
Control recorded the grid points from the Viet Cong loca-
tions. Certainly we were not interested in the enemy loca-
tions; we were looking for the U.S. locations and the U.S.
grid points. 15
In an interim report in 1985, CDC claimed that the ESG had not
provided gap-filling information, such as morning reports. Chris-
tian testified that the interim report. was false.
... If we traced the record of the Environmental Support
Group back to 1981, I reported to Congress that there were
several collections of records that would provide company
information beginning with daily journals, operational re-
ports of lessons learned, command reports, situation re-
ports, intelligence reports, and morning reports.
Yet in that interim report No. 2 . . .they indicate ...
* that I didn't tell them about morning reports. Well, I don't
know what else I could have done, with testifying before
Congress four times, preparing a report at the Internation-
al Dioxin Conference,
laying out all of those types of
records with 27 meetings with the VA before the CDC took
over.
Certainly
everyone
was aware
of those
collections.
1
6
As the study ensued, the limitations
in the methodology
chosen
by CDC prevented the Agency from identifying Vietnam veterans
exposed to Agent Orange. A further constraint on the outcome was
CDC's decision
to use but one approach
to identify
exposed veter-
ans when others existed.
14 Hearing, 1989. Testimony of Richard Christian, former Director of the U.S. Army and Joint
Services Environmental
Support Group, Department
of the Army, pp. 42-43.
15 Ibid., p. 42.
16 Ibid., p. 41.
The committee
believes that the decision by the White House to
cancel the Agent Orange study was premature,
and failed to give
the Agency the opportunity
to explore alternative
methodologies.
The justification
for cancelling the studies-that
military records
alone could not be used to conduct an Agent Orange exposure
study-was
misleading.
Clearly, better alternatives
had not been
fully considered before the study was canceled.
In particular, CDC did not attempt to identify the group of veter-
ans described in Dr. Landrigan's
testimony as being highly ex-
posed. The Government was aware of situations where soldiers had
been highly exposed, but this information was not considered by
CDC. For example, a highly exposed group of veterans was discov-
ered by the White House in 1981. According to a White House
memorandum:
Newly reviewed data shows that a plane taking off from
the Bien Hoa Base in Vietnam on a date certain, discov-
ered a faulty engine, turned left to go back to the air-strip,
and promptly dumped 500 gallons of Agent Orange on the
base camp from a height of 2,000 feet: the "dump value
rate" of the plane used was 1,000 gallons in 30 seconds and
suggests a high exposure to Agent Orange by a newly de-
fined but as yet unidentified
population.
17
The memorandum states that files in the Office of the Army Ad-
jutant General showed that there were 87 such aborted defoliating
missions in Vietnam, and that 41 of the missions involved Agent
Orange."' The White House was aware of the missions and their
locations, yet CDC did not attempt to include the unknowing veter-
ans who received these highly-concentrated exposures in its study.
B. THE ORIGINAL PROTOCOL FOR THE CDC AGENT ORANGE STUDY WAS
CHANGED TO THE POINT THAT IT WAS UNLIKELY FOR THE HEAVIEST
EXPOSED SOLDIERS TO BE IDENTIFIED
It is important to remember that the Agent Orange exposure as-
sessment study was designed to identify soldiers who were exposed
to the herbicide, not to link disease to exposure. That would have
come later had the study not been canceled. The only records avail-
able to CDC were the troop movement records and the Agent
Orange spray data in the files of the Department of Defense.
Therefore, CDC had to design the best method of using troop loca-
tion information to match against the spray records.
But from the start, CDC disregarded the records of soldiers who
would have been among the most likely to have been exposed. The
original protocol for the CDC Agent Orange Study contained nu-
merous restrictions which would have the effect of arbitrarily lim-
iting the number and type of veterans selected for the study. The
protocol states:
CDC proposes to limit this study to draftees and single-
term enlistees in the non-officer ranks who served in the
Army (grades El through E5 only); selection will be fur-
17 "New Information Regarding Agent Orange," The White House, Memorandum from Shan-
non Fairbanks to Martin Anderson, September 9, 1981, p. 1.
18 Ibid.
ther limited to those who had only one tour of duty in
Vietnam. Exclusion of officers is based primarily on a
desire to make the groups as homogeneous as possible with
respect to pre-existing demographic factors which could in-
fluence health. In addition, the inclusion of officers might
require substantially increased record review to assess her-
bicide exposure potential . . . because of multiple tours of
duty in Vietnam. 19
CDC eliminated from the study all military service branches
other than the Army. It excluded all officers and anyone who
served more than one tour of duty in Vietnam. The reason for
these exclusions was to reduce the amount of record review. Clear-
ly only a small percentage of available records were reviewed by
CDC.
The study group selection was also limited only to veterans serv-
ing in III Corps, thus eliminating six other combat corps stationed
in Vietnam during the war.
20
The study group was also limited to
the time period,
1967 to 1968.21
Importantly, the original protocol called for soldiers to be
tracked according to company records because the data is the most
precise and also because companies were the smallest units for
which data was available. The use of company records would have
enabled CDC to track individual soldiers. CDC recognized the im-
portance of company records and was aware early on that the
records could be incomplete. Still, the Agency's scientists believed
the problem could be surmounted. CDC stated in the original proto-
col "that it is not necessarily wise to exclude a unit simply because
some of its records are missing-units with missing records could
have had more or less exposure to herbicides than units with com-
plete
records."
22
The protocol reveals that the limitations in subject selection are
due to expediency, particularly to prevent lengthy record reviews.
For example, in deciding to exclude Marines from the study group,
the protocol states:
The desire to omit the Marine Corps from this study can
now be more easily understood.
If Marines were included,
the records review and other selection tasks to this point
would have to be done separately for them because they
were largely stationed in I Corps, and this would cause
delay.
23
The original protocol documents
that CDC intended to look only
at a small portion'of
the records available,
not because the data
was unusable,
but because Agency officials
thought looking at all
the records would require too much time. Yet, based on only a lim-
ited review of available records, the study was canceled in the end
because CDC found military records could not be used to identify
19 "Protocol for Epidemiologic
Studies of the Health of Vietnam veterans," Centers for Dis-
ease Control, Public Health Service, U.S. Department
of Health and Human Services, November
1983, p. 10.
2o Ibid., p. 11.
21 Ibid.,
p. 12.
22 Ibid.
23 Ibid., p. 14.
veterans who had been exposed to Agent Orange. The committee
questions CDC's conclusion that the records could not be used be-
cause CDC did not consider all the records.
The protocol also suggests that the study would have certain lim-
itations beyond the restrictions CDC had placed on the subjects to
be selected for cohorts. "Unavoidable limitations of the proposed
studies, or indeed any other studies which could be done, will pre-
clude describing the results as 'definitive"' the protocol cau-
tioned.
24
The protocol stated that one important limitation was the
observational, as opposed to experimental, nature of the study.
... Another general caveat is that it is not possible to
prove a negative-that is, it will never be possible to say
with certainty that herbicide exposure or some other
factor connected with Vietnam service did not cause any
adverse health effects.
* * * * *
. .Moreover, even in the absence of exposure misclassifi-
cation, the studies will have low power for rare diseases
and/or low increases in risk, or for increases in risk limit-
ed to those veterans with prolonged and/or heavy expo-
sure to herbicides or some other harmful factor. Thus, an
overall finding of no increase in risk might "hide" a real
increase for specific disease categories or special groups of
veterans.
2 5
After preparing the original protocol, with all its limitations and
caveats, CDC diluted the study even further by adding more re-
strictions to the criteria for the inclusion of subjects. A February
1985 internal, interim report reveals that CDC made four major
modifications in the original protocol. The most serious modifica-
tion was that:
Battalions-rather than companies-will serve as the
basis for cohort selection and unit location.
Current data indicate that daily location information
is not consistently available for units smaller than battal-
ions. In other words, we believe that the location of indi-
vidual battalions can be placed with reasonable accuracy
for most of the 731 days but the locations of the compo-
nent companies of a battalion cannot be as reliably placed
due to lack of data.
2 6
The effect of this change would be to severely diminish the
study's ability to identify veterans most likely to have been ex-
posed to Agent Orange. The interim report admits as much. "...
because battalions are larger, more dispersed units than compa-
nies, the location of men in relation to herbicide applications will
be known with less precision than envisioned in the protocol." 27
24 Ibid., p. 37.
25 Ibid.
21 "Agent Orange Projects Interim Report: Agent Orange Study Exposure Assessment: Proce-
dures and Statistical Issues," Agent Orange Projects, Center for Environmental Health, Centers
for Disease Control, February 1985, p. 4.
27 Ibid., p. 1.
CDC's decision to use battalion data was based on its conclusion
that the military records were not accurate for the purpose of
tracking companies. OTA concurred in this decision at the time.
However, as noted previously, testimony and documents provided
to the subcommittee indicate that CDC's decision was disputed, not
only by the Department of Defense, which had greater expertise in
evaluating military records, but by scientists working on the Agent
Orange Project itself.
The second major change in the protocol was that the ".
Ranking of unit exposure likelihoods will not determine selection
of men into the study." 28 Originally, CDC planned to rank 125
companies according to time and distance proximity to herbicide
applications, before any individuals from the units were selected
for the study. The interim report states this approach was aban-
doned because CDC had identified numerous transfers of soldiers
between companies in III Corps.
2 9
Therefore, unit exposures would
not necessarily correlate to individual exposures.
The committee finds this change to be another example of CDC's
dilution of the study for reasons of convenience and expedience.
The individual soldiers could have been tracked, adjustments made
for transfers, but because of the amount of time and work involved,
and due to its own perception that the study had to.be completed
post-haste, this more precise method of tracking individual compa-
nies and soldiers was disregarded. The committee recognizes that
pursuing alternatives would have been a complicated and arduous
task; however, it was feasible. Because CDC diluted the study with-
out doing the necessary work to compensate for tracking problems,
the committee finds no support for the final conclusion that the
Agency could not identify any method for using military records to
assess exposure.
The interim report discussed two additional changes in the origi-
nal protocol: Individual military records would be used to buttress
unit morning reports to track troop movements and the study
would be restricted to men serving only in infantry or artillery
-units. In explaining the reason for the latter alteration, CDC re-
vealed -that -many soldiers likely to have been exposed were arbi-
trarily eliminated from the study because their backgrounds might
not conform to the backgrounds of the study comparison group who
did not serve in Vietnam. According to the interim report:
. . For the Agent Orange Study, our goal is not to select
a "representative"
sample of Army men in III Corps, but
rather to select men from III Corps who are as comparable
as possible, except for differing likelihoods of Agent
Orange exposure.
We believe that men serving in artillery
and infantry
battalions
will resemble
each other more
,closely
in -terms of baseline
characteristics
and general
military experience
than they would resemble
men from
cavalry,
armor, and engineering
units.
30
28 Ibid., p. 6.
29 Ibid.
30 bid., p. 9.
The effect of these modifications on the study was noted in the
interim report.
.. . because unit exposure likelihoods do not appear to
correlate highly with individual men's exposure likeli-
hoods, we will not be able to exclude men from units with
intermediate exposure likelihoods from the study; there-
fore the spread or range of exposure likelihoods between
the groups being compared will likely be reduced.
3 1
[Further,] The misclassification of Agent Orange expo-
sure inherent in this study means that the true magnitude
of exposure-disease associations will be underestimated.
3 2
By the end of 1985, CDC scientists realized that the restrictions
in the criteria described previously had so limited the scope of
their research, they were unable to identify enough subjects for in-
clusion in the study. To compensate, CDC changed the criteria
again, allowing a larger number of soldiers to be included, but re-
ducing the numbers of soldiers with the greatest likelihood of expo-
sure to Agent Orange. On December 8, 1985, new criteria were es-
tablished in a second, interim report.
33
First, CDC reduced the
number of days served in combat companies for selection in the
study from 9 months to 6 months. The Agency also broadened the
time period for study subjects from veterans who served only in the
time period, 1967-1968, when the herbicide spraying was heaviest,
by adding an additional six months.
3 4
The Director of the CDC Center for Environmental
Health and
Injury Control testified that the reduction in number of days
served eliminated veterans who had greater opportunities
for expo-
sure to Agent Orange. The following exchange occurred between
the Subcommittee Chairman and the Director.
Mr. WEISS. If you reduced that 9-month number, you
would be eliminating the Veterans with the greatest oppor-
tunity for exposure, would you not?
Dr. HOUK. If we reduced the number, right.
Mr. WEISS. The 9-month number.
Dr. HOUK. If we had made it 6 months?
Mr. WEISS. Yes.
Dr. HOUK. We would have less likelihood of exposure.
Mr. WEISS. That's right.
Dr. HOUK. Right.
3
5
In addition to its complaints about the Defense Department's
troop movement records, CDC also questioned data it had received
regarding the dispersion of the Agent Orange after being sprayed.
The December 8, 1985, interim report discussed test missions flown
over a sampling grid in 1970.
" Ibid., p. 1.
3 Ibid.,
p. 2.
. "Exposure Assessment for the Agent Orange Study, Interim Report Number 2, Supplemen-
tal Information," Agent orange Projects, Division of Chronic Disease Control, Center for Envi-
ronmental Health, Centers for Disease Control, December 8, 1985.
34 Ibid.,
p. 5.
35 Hearing, 1989. Testimony of'Vernon N. Houk, M.D., Director, Center for Environmental
Health and Injury Control, Centers for Disease Control, p. 47.
* . .While many conditions such as the spray system, alti-
tude, and aircraft were similar to those which prevailed in
-Vietnam for the Ranch Hand missions, other conditions
(e.g. the number of aircraft used and vegetative cover)
were different, so that results can only be used as rough
estimates of herbicide dispersion. In a crosswind, the con-
centration decreased...
... unfortunately estimates of dose would depend on
knowledge of bioavailability, absorption rates, clothing
worn, as well as behavioral factors such as amount of time
spent in contact with vegetation, soil, and grasses, and con-
sumption of local food and water. Data are insufficient to
estimate
dose.
3 6
At the time CDC was diluting the study, using the rationale that
the Defense Department's records were inaccurate or missing, a
team of expert scientists from the National Academy of Sciences'
Institute of Medicine (IOM) made a site visit to the Department's
Environmental Support Group (ESG). The scientific team found the
ESG records important and useable, and was critical of CDC's per-
formance. The IOM final report stated that they were "satisfied
that the ESG is capable of determining locations and fillings gaps
using a contextual approach, and notes that the ESG exhibits a
high degree of competence in recording data gathered from these
activities." 37 The IOM also said it was "satisfied with the ESG's
documented Standard Operating Procedure to fill in gaps, and was
also satisfied with the methods used by teams or pairs to resolve
.questions which arose during contextual analysis. . . satisfied with
the ESG's quality control program." 38
ESG personnel informed the IOM team that ESG's "ability to
make determinations on company locations has been hampered by
CDC-imposed constraints. The ESG also pointed out that there is a
considerable loss of numbers of veterans with potential exposure
from the study because of CDC's stringent eligibility require-
ments." 39 ESG identified numerous cases of veterans who had re-
ceived 10 or more exposures to Agent Orange, but had been ex-
cluded from the CDC study. For example, ESG pointed out, because
CDC has eliminated. all headquarters personnel from the study, if a
veteran "had 170 days of combat and ... then moved into a head-
quarters unit they were not included in the study..." 4o
After reviewing the work of ESG, the IOM team concluded that
CDC had wrongly restricted the study. For example, the team
stated that it was:
concerned about the use of discrete categories to define ex-
posure (e.g., exposure vs. non-exposure),
and instead favors
measuring
exposure
as a continuous
variable . .. when-
ever possible. The use of discrete categories
in data collec-
36 Op. cit., p. 17. See footnote 33.
3 7
"Site Visit to the U.S. Army Environmental Support Group," Advisory Committee on the
CDC Study of the Health of Vietnam veterans, National Academy of Sciences, Institute of Medi-
cine, March 7, 1986, p. 6.
36 Ibid., pp. 6-7
39 Ibid., p. 4.
40 Ibid., p. 5.
tion stages essentially discards valuable information, and
such data can be stratified for subsequent analyses.
4 1
The IOM team also said it "was perplexed about criteria estab-
lished by CDC to select subjects, especially the 180 day cutoff for
combat time and the exclusion of headquarters-based individuals
exposed to perimeter sprays." 4
2
The team also found, "the criteria
used to define exposure and to define who will be included in the
study seem arbitrary and confusing" and concurred with ESG
"that there appear to be many exposed individuals who will be ex-
cluded from the study as it is now designed." 43
The IOM experts were also concerned that a scientific panel of
the White House Agent Orange Working Group (AOWG) "is cur-
rently trying to define 'exposure' " and noted "the thinness of ex-
pertise in certain areas" of the individuals serving on the panel.
4 4
In conclusion, the scientific team reported that "there appear to be
different kinds of exposure, and that individuals were exposed
under a variety of conditions; these issues need to be addressed
more carefully. The [IOM] subcommittee finds the current defini-
tion of exposure to be inadequate." 45
The IOM findings were not reported to the White House. There-
fore, the White House AOWG was misinformed when its Science
Panel held a meeting three months after the IOM site visit to ESG,
and reported:
Pertinent military records have been used appropriately
to locate all known herbicide spraying operations and mili-
tary units and to identify individuals who may have had
opportunities for exposure to Agent Orange. Limitations
on the assessment of exposure opportunities are due to
limitations in the records themselves.
There is unanimous agreement that an epidemiological
study of ground troops' possible exposures to Agent
Orange disseminated by Operation Ranch Hand fixed-wing
aerial spraying, based solely on military records, does not
appear to be scientifically feasible.
4 6
In reaching this decision, the AOWG Science Panel noted that it
"did not bring in independent evidence or experts." In light of the
evidence compiled since the cancellation of the study, the commit-
tee finds that the White House's final decision was based on incom-
plete and erroneous information, and that pertinent information
was purposefully withheld by CDC. The committee further con-
cludes that the dilution of the study was unnecessary, was based on
arbitrary criteria established by a panel controlled by the White
House, and directly led to the study's collapse. The Agency Orange
Exposure study failed as a result of its own design.
" Ibid., p. 7.
4 2
Ibid.
43
bid., pp. 11-12.
4 4
Ibid.,
p. 13.
45
Ibid., pp. 12-13.
46 "Minutes of the Meeting on June 17, 1986," Science Panel of the White House Agent
Orange Working Group.
C. THE BLOOD SERUM ANALYSIS, WHICH WAS USED AS PROOF BY CDC,
THAT AN AGENT ORANGE EXPOSURE STUDY COULD NOT BE CONDUCT-
ED, WAS BASED ON ERRONEOUS ASSUMPTIONS AND A FLAWED ANALY-
SIS
Using its criticism of the ESG records as its excuse, CDC and the
AOWG were inclined against proceeding with the study. However,
before a decision to cancel or go forward could be reached, CDC
needed to test the exposure definition it had developed. The
method it considered to validate the definition involved testing
Vietnam veterans for traces of dioxin. At first, this was not consid-
ered feasible. Although the measurement of dioxin in humans had
been achieved through the use of gas chromatography/mass spec-
trometry to identify the contaminant in fatty adipose tissue, Feder-
al scientists believed there were two obstacles to using this process
in the Agent Orange study. First, there was the difficulty of obtain-
ing adipose tissue, which required surgical extraction, a procedure
considered too invasive and unwieldy for a study of this sort.
Second, the half-life of dioxin in animal studies proved to be less
than a year, which, if extrapolated to humans, would mean the
dioxin had been eliminated from the bodies of Vietnam veterans
long
before
1986.
4 7
During a meeting of the AOWG Science Panel on June 17, 1986,
representatives of CDC proposed using dioxin levels in blood serum
as a surrogate for Agent Orange exposure. The minutes of the
meeting indicate that the Science Panel was divided on the propos-
al; some thought it should be done, while others did not think it
was feasible. The panel concluded:
There is no agreement at this time whether a feasible
and accurate method for validation of individual exposure
status can be devised, and the science panel cannot recom-
mend attempting a verification study.
* * * * *
There is unanimous agreement that if a well-designed
exposure verification study fails to validate individuals' ex-
posures as determined from military records, the Agent
Orange
Epidemiological
Study
should
be discontinued.
4
1
CDC overcame the difficulties of obtaining adipose tissue by de-
veloping, for the first time, a method of identifying dioxin traces in
human blood. This left the second obstacle, the short half-life of
dioxin, as the sole impediment. This was also overcome, according
to CDC:
The second problem of expected short half-life was
changing
as studies
on human 2,3,7,8-TCDD
exposure
yielded results more compatible
with a half-life much
longer than 1 year. Results from a few persons
in Missouri
47 "Estimates of the Half-Life of 2,3,7,8.Tetrachlorodibenzo-p-Dioxin
in Vietnam veterans of
Operation Ranch Hand," James L. Pirkle, Donald G. Patterson, Donald L. Phillips, and Larry L.
Needham, Centers for Disease Control, and William H. Wolfe, Joel E. Michalek, Judson C.
Miner, and Michael R. Peterson, U.S. Air Force School of Aerospace Medicine, Journal of Toxi-
cology and Environmental
Health, Vol. 26, 1989, p. 166.
48 Op. cit., p. 2. See footnote 46.
... and an occupationally exposed worker in Germany
. . suggested, on the basis of last known exposure deter-
mined by history, that 5-8 years was a more reasonable es-
timate of 2,3,7,8-TCDD half-life for humans. In another
study . . . an individual who ingested radiolabeled 2,3,7,8-
TCDD and tracked his urinary and fecal excretion of the
radiolabel
estimated
the half-life at 5-8 years.
4 9
The exposure examples cited by CDC to justify a new half-life of
dioxin involved situations totally unlike Agent Orange spraying in
Vietnam. Yet, based on this new evidence, CDC, in collaboration
with the U.S. Air Force, tested the blood of Ranch Hand veterans,
as well as veterans whom CDC had selected in its exposure study.
Using a half-life estimate for dioxin of 7.1 years, the blood of 646
Vietnam veterans was tested. According to CDC, only 4 percent of
the non-Ranch Hand subjects had elevated levels of dioxin in their
blood, and there was no correlation between exposure and dioxin in
the blood.
50
CDC reached no conclusions about exposure on the
basis of the study and, without further evaluation, found there was
no other method to assess exposure. The final report on the blood
testing found:
The findings of this study and the conclusions from the
AOWG Science Sub-Panel report on exposure assessment
... do not identify any method for utilizing military
records or self-reported exposure to distinguish between
exposed and nonexposed Vietnam veterans, as would be
needed for a cohort study of possible health effects.
5 1
CDC's conclusion that the half-life of dioxin in the human body
is 7.1 years was reached in disregard of warnings from CDC's own
scientists and the peer review committee at IOM that there was
not sufficient evidence to support the longer half-life. IOM in-
formed CDC that, because of the incorrect assumptions about the
half-life of dioxin, the conclusions of the blood study were not sup-
portable. IOM informed CDC of its concerns:
[The IOM site team has] substantial reservations about
the conclusions and recommendations presented in the
CDC pilot study report. The committee believes that the
conclusions as they are now stated are not fully supported
by the evidence provided in this report.
Moreover, the committee urges that the recommenda-
tions as currently stated in the CDC pilot study report be
deleted; they appear to reach well beyond the data pre-
sented and to incorporate information that goes beyond
the scope of the current investigation.
5 2
The IOM report criticized CDC's assumption about the half-life of
dioxin. The IOM was concerned that the study was based on blood
drawn 20 years after exposure to Agent Orange. The report advised
49 Op. cit., pp. 166-167. See footnote 47.
5o Op. cit., p. 43. See footnote 3.
5 1 Ibid.
12 "Review of Comparison of Serum Levels of 2,3,7,8-TCDD with Indirect Estimates of Agent
Orange Exposure in Vietnam veterans," Fifth Letter Report, Advisory Committee on the CDC
Study of the Health of Vietnam veterans, Institute of Medicine, June 26, 1987, pp. 17-18.
that "There are several possible conditions that could have oc-
curred between 1967 and 1987 that might result in the TCDD levels
observed today." 53 The report stated that the veterans studied
could have had different ranges of exposure, which due to decay of
dioxin in the body, could result in the same background levels of
dioxin 20 years later, or that they could have been equally exposed,
but dioxin decayed in their bodies at different rates, due to physi-
cal and environmental
variables.
5 4
The IOM committee was particularly critical of the conclusion by
CDC and the White House that the blood tests proved a complete
exposure study was impossible.
The conclusion that a full-scale cohort study is not feasible
goes well beyond the scope of this study, which was con-
ducted to assess the validity of indirect indices of exposure.
If such an expansive statement is offered, it certainly war-
rants argumentation and documentation. The committee
did not find support for this conclusion in the pilot study
55
The IOM panel was also concerned that blood was not available
for a significant portion of the cohort studies: 52 percent of the vet-
erans who did not serve in Vietnam and only 68 percent of the vet-
erans who did serve in Vietnam.
5 6
The IOM team asked CDC to
provide alternative explanations for the lack of correspondence be-
tween the indirect measures and current dioxin levels, and in
regard to fears about Agent Orange, concluded:
... it is not clear that the findings presented in this pilot
study warrant complete abatement of concern. Little can
be said with certainty about the TCDD levels in men 20
years ago. Moreover, background levels detected some
years later do not provide assurance that no health effects
will ultimately surface. Given the insufficiency of data re-
lating the level of exposure to health outcome in humans,
delayed effects of low or high doses could become apparent
years
after
exposure
occurred.
5 7
IOM's comments, which were largely disregarded in the final
blood serum assay report, served to remind CDC that the purpose
of the blood tests was to validate the exposure definitions and
scores developed by the agency and the AOWG, not to actually
assess exposure. It should have come as no surprise to CDC that
there was no correlation between the blood tests and exposure
scores, because the scores were based on a diluted protocol that
eliminated from the study the veterans who would have been the
likeliest to be exposed. The Director of the CDC Agent Orange
Project was aware of this problem, and warned the Director of CDC
on March 17, 1986:
If only weak correlations
are found between
residual
TCDD levels and the exposure
opportunity
scores (e.g.,
53 Ibid.,
p. 11.
54
Ibid.
55 Ibid.,
p. 19.
56 Ibid., p. 15.
57 Ibid., pp. 19-20.
there is no difference between the "high" and "low" expo-
sure opportunity scores), it will be unknown if the failure
to find a difference was due to limited Agent Orange expo-
sure in Vietnam
or to misclassification
of the scores.
58
The same memorandum also stated that, while the blood serum
assay was a legitimate tool for validating CDC exposure methodolo-
gy, "it should be noted that current TCDD levels could bear an im-
perfect correlation with exposure to Agent Orange because of indi-
vidual variability in both the metabolic half-life of dioxins and in
post-service
exposure
to dioxin containing
compounds."
59
More than a year earlier, in January 1986, the senior statistician
on the Agent Orange Project expressed serious reservations about
using dioxin measurements to validate exposure. The statistician
warned that a TCDD analysis had a substantial likelihood that
there will be essentially no correlations:
... there is a great danger of very poor correlation in adi-
pose measurements separated by a long period of time.
Thus, there is a substantial chance that we will find at
best a very weak relation between exposure score and the
adipose measurement . . . we are asking for trouble if we
do TCDD measurements.
6
"
The Director of the Center for Environmental Health and Injury
Control stated that the senior statistician did not know what he
was talking about. Dr. Houk testified, "I think he has been proven
to be wrong and I discussed with him that it would be inappropri-
ate for a senior laboratory scientist to comment on the inadequa-
cies of a very elegant statistical design and analysis." 61 Dr. Houk's
criticism is curious, given that he had selected this same statisti-
cian to direct the analysis of the study's results.
6 2
The evidence is compelling that CDC was aware that the blood
test study could not be the final word on the Agent Orange study.
In 1987, the scientists working on the blood serum study cautioned
in a provisional report:
The study reported here measured TCDD levels as a
marker of prior exposure to Agent Orange. It was not de-
signed to detect "biologically meaningful" differences in
serum TCDD levels because:
1. Levels of TCDD in man at which adverse health
effects become manifest are not known;
2. Even if toxic levels of TCDD were known, the
tissue half-life of TCDD in man is not precisely known,
so that it would be difficult to extrapolate current
"Evaluation of residual dioxin levels in Vietnam veterans as a validation of Agent Orange
exposure opportunity scores," Memorandum from Peter M. Layde, M.D., M.Sc., Acting Director,
Agent Orange Projects, to James 0. Mason, M.D., Dr.P.H., Director, Centers for Disease Control,
March 17, 1986, p. 3.
59 Ibid.
50 "Trip Report on Meeting with Tavia Gordon, Max Halperin, and Shelby Stanton," Memo-
randum from John M. Karon, Senior Statistician, Agent Orange Project, to Dan McGee, Direc-
tor, Agent Orange Project, January 9, 1986, p. 6.
61 Hearing, 1989, p. 67.
62
Ibid., p. 68.
levels in veterans back 20 years to when exposure may
have occurred; and
3. The component of Agent Orange that may have
been responsible for adverse health effects is not
known. Although attention has focused on the TCDD
contaminant, the two active Agent Orange ingredi-
ents, 2,4-D and 2,4,5-T, also may cause adverse health
effects. . . . Different components of Agent Orange
may have been responsible for different manifesta-
tions.
6 3
One of the scientists on the Agent Orange Project testified that
CDC, despite its public claims to the contrary, "had no idea about
its [dioxin's] half-life. We had some conjectures about the half-life
of dioxin but we didn't have anything certain." 64 He added:
Also, we weren't certain about the dosage that would
produce an effect. No one knew the relative importance of
the various routes of exposure. We didn't know much at
all about the distribution of dioxin within the body.
There's an assumption that it stores itself in fat tissue 11
times more frequently than in serum but this sort of infor-
mation, I think, is very questionable.
6 5
CDC had already diluted the study protocol to the point it would
not identify the veterans most likely to have been exposed, to avoid
a lengthy document review. The decisions to dilute were made for
reasons of expediency. With its manipulation of the half-life of
dioxin, CDC seemed to be trying to justify an intentionally flawed
exposure assessment methodology with the blood serum analysis.
The blood assay test was more a method of convenience than an
exhaustive scientific approach. As Dr. Smith testified:
We damned the historical records with overemphasis on
minor problems with ESG. We ignored other sources of
records and relied too much on unreliable, anecdotal re-
ports. We played with the half-life of dioxin using only
those studies and reports that would advance the concepts
of what the model which we wanted to use would say at
the time....
We ignored most of the medical literature on dioxin. We
had some excellent references but we only focused on a
few papers, especially when the adipose tissue proposal
came along. The focus was only on a few papers that sup-
ported what we had in mind, or, what we had ultimately
come
to.
6 6
Dr. Michael Kafrissen worked on the Agent Orange Projects and
had been asked by CDC to evaluate the half-life of dioxin in the
human body. Although he said there was some preliminary evi-
dence indicating a half-life of 7.1 years for dioxin in humans in
83 "Comparison of Serum Levels of 2,3,7,8-TCDD with Indirect Estimates of Agent Orange Ex-
posure in Vietnam veterans-Provisional
Report," Agent Orange Projects, Division of Chronic
Disease Control, Center for Environmental Health, Centers for Disease Control, July 1987, p. 18.
64 Hearing, 1989. Smith testimony, p. 81.
:- Ibid.
86 Ibid., p. 82.
26
some situations, he stated, "I don't believe at this time there is suf-
ficient information to say that this number obtains for all men at
all times under all physical circumstances." 67 Dr. Kafrissen also
testified that "We did not know all of the critical levels-and I
don't believe we now know-of dioxin that is going to effect a prob-
lem in humans. For that reason, we were concerned with the possi-
bility that men may have been affected, may have been exposed
and that we would no longer be able to judge that based on these
laboratory
specimens."
68
The assumption that dioxin had a longer half-life in the human
body than previously believed was based on studies of individuals
who were directly exposed to high doses of dioxin in laboratory ac-
cidents, one study of a scientist who ingested dioxin in high doses
to evaluate half-life, and tests conducted of veterans who worked in
Operation Rand Hand, directly spraying and handling Agent
Orange in large amounts. These subjects cannot be compared to
Vietnam veterans who had less direct exposure, possibly over
longer periods of time. The large, short-term exposure to undiluted
chemicals as the result of industrial accidents is quite different
from the possible long-term exposure to Agent Orange in which the
contaminants were diluted with other substances. Obviously, dioxin
exposure is different in these subjects and the elimination of dioxin
traces would be entirely different than in the subjects mentioned
earlier. While studying the half-life of dioxin in highly exposed vet-
erans and workers was relevant to the CDC Agent Orange Study,
the committee finds the conclusion that the 7.1 years half-life in
highly exposed individuals can .be extrapolated to all Vietnam vet-
erans to be unsupportable.
Even the method by which CDC determined the half-life-called
first order kinetics-may not have been the correct way of assess-
ing dioxin elimination from the human body. The scientists who
conducted the half-life study cautioned:
With just two serum 2,3,7,8-TCDD values per veteran, it
is not possible to definitively state that a first-order kinetic
model is appropriate to describe the decline in serum
2,3,7,8-TCDD concentrations over time or that a more com-
plex model is needed.
6 9
Based on the complete record of the CDC Agent Orange Studies,
the committee believes there was a two-part course selected by the
Agency that guaranteed the eventual cancellation of the study.
First, CDC ignored the military records experts at the Pentagon,
created its own use of the records to establish exposure, and then
diluted the study to the point that numerous veterans who were
most likely to have high exposure were eliminated from the study.
Second, using the flawed exposure definition it had developed, CDC
then attempted to justify it with the blood serum assay, which,
itself, was based on false assumptions about the chemical's half-life
in the human body. Unsound assumptions about the half-life of
67 Ibid. Testimony of Michael E. Kafrissen, M.D., former epidemiologist, Centers for Disease
Control, p. 107.
6S Ibid., pp. 106-107.
69 Op. cit., p. 170. See footnote 47.
dioxin were used to validate badly-designed exposure assessments.
Given these restrictions, it would have been impossible to validate
exposure.
The committee also finds that, in adopting these courses of
action, CDC disregarded or ignored the warnings from its own sci-
entists, Defense Department officials, and the expert committee as-
sembled by the IOM.
D. THE WHITE HOUSE COMPROMISED THE INDEPENDENCE OF THE CDC
AND UNDERMINED THE STUDY BY CONTROLLING CRUCIAL DECISIONS
AND GUIDING THE COURSE OF RESEARCH AT THE SAME TIME IT HAD
SECRETLY TAKEN A LEGAL POSITION TO RESIST DEMANDS TO COMPEN-
SATE VICTIMS OF AGENT ORANGE EXPOSURE AND INDUSTRIAL ACCI-
DENTS
The subcommittee's review of Executive Branch records demon-
strates that two components of the White House, the Agent Orange
Working Group (AOWG) and the Office of Management and
Budget (OMB), monitored and controlled the CDC Agent Orange
Project. Top administration officials such as Attorney General
Edwin Meese and White House Chief of Staff James Baker had ul-
timate decisionmaking authority for approving and eventually can-
celing the exposure study, but the nuts and bolts of the work were
handled by the AOWG and OMB.
While Agent Orange was the impetus for the establishment of
the White House advisory committee under President Carter, the
panel's mandate was not limited to the war-time defoliant. Its pur-
view included the effect of herbicide contaminants in addition to
Agent Orange.
The first responsibilities of the White House panel were to in-
struct the Air Force to conduct the study of Ranch Hand personnel
and to oversee both the Ranch Hand study and the epidemiological
study
of Agent
Orange.
70
In June 1983, the Reagan Administration decided to revise the
charter of the White House panel by eliminating its mandate to ex-
plore the effects of all herbicides and contaminants and concentrat-
ing the work of the group only on Agent Orange. In a memoran-
dum from Robert B. Carleson, Special Assistant to the President, to
White House aide Jack Svahn, it was decided:
... the task of the Working Group on Agent Orange be
concentrated so that its work can be completed in a timely
manner. Therefore, please ensure that your working group
deals only and specifically with the possible long term ad-
verse health effects of exposure to Agent Orange and Viet-
nam
veterans.
7 1
The AOWG was the successor to the White House Advisory Com-
mittee on Special Studies Relating to the Possible Long-Term
Health Effects of Phenoxy Herbicides
and Contaminants.
The origi-
70 Memoranda to the Secretary of Defense and the Administrator of Veterans Affairs, from
Stuart E. Eizenstat, Assistant to the President for Domestic Affairs and Policy, September 16,
1980 .
72 "Task Assignment of the Working Group on Agent Orange," Memorandum from Robert B.
Carleson, Special Assistant to the President, to Jack Svahn, Chairman, Working Group on
Agent Orange, June 2, 1983.
nal advisory committee was established on December 11, 1979, in a
memorandum drafted by Stuart E. Eizenstat, Assistant to Presi-
dent Jimmy Carter for Domestic Affairs and Policy. The advisory
panel's mission was to assure that all Federal research into herbi-
cides and contaminants provide reliable data, and to provide tech-
nical support to Federal agencies working on the research.
7 2
The original mandate to focus the White House panel on the ef-
fects of all herbicides was abruptly altered by the Reagan White
House. By focusing the work of AOWG on Agent Orange only, the
administration laid the groundwork for manipulating the study to
the point of uselessness.
A possible reason that the White House chose this path is re-
vealed in confidential documents prepared by attorneys in OMB.
The White House was deeply concerned that the Federal Govern-
ment would be placed in the position of paying compensation to
veterans suffering diseases related to Agent Orange and, moreover,
feared that providing help to Vietnam veterans would set the
precedent of having the United States compensate civilian victims
of toxic contaminant exposure, too.
These concerns were included in a series of memoranda prepared
by an OMB attorney about pending legislative proposals to compen-
sate victims exposed to Agent Orange and nuclear radiation. These
proposals, OMB warned, "have enormous fiscal implications, poten-
tially in the hundreds of billions of dollars."
' 7 3
In discussing H.R.
1961, a bill pending in the House of Representatives that would
provide compensation to veterans exposed to Agent Orange and nu-
clear radiation, OMB called the legislation "our first key challenge
on toxic compensation-and
it has significant ramifications for
other, more costly compensation proposals. It is therefore extreme-
ly important that we organize our position and response... 74
The ramifications feared by the White House were delineated in
a second memorandum. A major worry was that the legislation, if
passed, would make it hard for the White House to prevent com-
pensation to victims of other types of contaminants:
The bill will make it far more difficult to stop broader
victims compensation schemes involving hazardous wastes
and substances. Dioxin-the toxic ingredient in Agent
Orange-is a major issue in this area (Love Canal and
Times Beach are largely dioxin exposure cases); we will be
-in the tenuous position of denying dioxin exposure compen-
sation to private citizens -while providing benefits to veter-
ans for in many instances lower levels of exposure. 7
5
The White House also feared that passage of the legislation
would damage its efforts to prevent compensation
for claims involv-
7 2
Charter, Advisory Committee on Special Studies Relating to the Possible Long-Term Health
Effects of Phenoxy Herbicides and Contaminants," as amended May 1984.
" "Compensation for Low-Level Radiation- Exposure and Related Developments," Executive
Office of the President, Office of Management and Budget, Memorandum to Ed Meese, Joe
Wright, Jay Keyworth, Fred Fielding, and M.B. Oglesby, from Mike Horowitz, December 6, 1983,
p.
1
.
14 Ibid.,
p. 3.
1 "Agent Orange/Radiation Compensation Legislation," Executive Office of the President,
Office of Management and Budget, Memorandum to Dave Stockman, Joe Wright, Jay Keyworth,
and Fred Fielding, from Mike Horowitz, January 17, 1984, p. 2.
ing nuclear radiation. 'The radiation portion of the bill will under-
mine the United States' position in pending radiation litigation.
The Allens case, involving $2 billion in claims, is still under advise-
ment before a Utah federal judge." 76
A subsequent memorandum reiterated these concerns and ex-
panded upon them. In regard to nuclear radiation, OMB warned:
Many residents near the test sites received cumulative
exposures far in excess of the single doses to which veter-
ans were exposed. These residents will undoubtedly
demand compensation for these far greater levels of cumu-
lative exposure .... There is understandable concern at
Justice that H.R. 1961 will effect the outcome of that case.
The bill will certainly be used by anti-nuclear extremists
as providing credibility for their alarmist claims on the
dangers to the public from nuclear weapons facilities, nu-
clear reactors, transportration of nuclear materials, under-
ground testing, etc. This may have serious national securi-
ty ramifications.
7 7
So strident was the administration in its belief that the Federal
Government should not be liable for exposure to toxic chemicals
that the Justice Department ordered the Defense Department not
to assist the Special Master overseeing the legal settlement be-
tween the manufacturers of Agent Orange and Vietnam veterans.
On May 22, 1986, the Director of the Defense Department's ESG
asked the Justice Department if he could assist the Special Master
in the evaluation of herbicide spray and troop movement records
already provided to the Special Master. The assistance, the director
wrote, would "prevent duplication, delay, misinterpretation and
costs against the funds set-aside for Veterans' awards." 78
The Justice Department ordered the Defense Department not to
cooperate with the Special Master, advising that "government em-
ployees not be made available to render advice and assistance con-
cerning the allocation of settlement fund proceeds." 79
A month later, the Director of ESG was asked to testify before a
subcommittee of the House Committee on Veterans Affairs about
the results of the work of the Agent Orange Working Group. In a
memorandum to the Defense Department Adjutant General, the
Director reported that much of his testimony was "deleted by
OMB .... Such results are neither forthright or responsive....
the testimony has been rewritten and its guts torn out. .... , 80
In light of these confidential actions by the Justice Department
and OMB, and given the White House's legal concerns that the
Federal Government could be liable in toxic contaminant law suits,
it is apparent why the charter of the AOWG was revised and limit-
I-
Ibid.
|7 "H.R. 1961, Agent Orange and Atomic Veterans Relief Act," Executive Office of the Presi-
dent, Office of Management and Budget, Memorandum to Ed Meese, Jack Svahn, Jay Keyworth,
and B. Oglesby, from Mike Horowitz, February 6, 1984, p. 2.
7
8 Letter from Robert C. Longstreth, Trial Attorney, Torts Branch, Civil Division, Department
of Justice, to Richard S. Christian, Director, U.S. Army and Joint Services Environmental Sup-
port Group, June 3, 1986. (Letter cites Christian letter of May 22, 1986.)
79 Ibid.
so Memorandum
from Director, Environmental
Support Group, to The Adjutant General, July
30, 1986.
ed to Agent Orange only. At the time of the revision, numerous
Federal research projects were being conducted of domestic and ci-
vilian worker exposure to herbicides.
8 1
By revising the charter to
ignore the dozens of Federal studies dealing with herbicides other
than Agent Orange, the White House forced the AOWG to overlook
a wide range of issues, not only related to Agent Orange, but vital
to the health of civilian workers, as well. Thus, the White House
would not have to deal with potential liabilities involving herbi-
cides other than Agent Orange.
The committee believes that, given the White House's predisposi-
tion against compensating victims of toxic contamination, White
House officials should have recused themselves from involvement
in the deliberations about the CDC Agent Orange Studies and that
the AOWG should not have been given the authority to approve or
81 The following studies were reported to the Agent Orange Working Group in a document,
"Federally Sponsored Human Studies Related to Agent Orange," April 11, 1983, National Insti-
tute of Occupational Safety and Health (NIOSH), Investigation of Leukemia Cluster in Madison
County, Kentucky, Allegedly Associated with Pentachlorophenol Treated Ammunition Boxes;
Centers for Disease Control (CDC), Birth Defects and Military Service in Vietnam Study; NIOSH
Dioxin Registry; National Institute of Environmental Health and Safety (NIEHS), Establish-
ment and Maintenance of an International Registry of Persons Exposed to Phenoxy Acid Herbi-
cides and Contaminants; NIOSH Soft Tissue Sarcoma Investigation, National Cancer Institute
(NC), Case Control Study of Lymphoma and Soft Tissue Sarcoma; NCI Study of Mortality
Among Pesticide Applicators from Florida; CDC Epidemiologic Study of Ground Troops Exposed
to Agent Orange during the Vietnam Conflict; Veterans Administration (VA), Vietnam veterans
Mortality Studies, Vietnam veteran Identical Twin Studies, Survey of Patient Treatment File
for Vietnam veteran In-Patient Care, Agent Orange Registry Examinations, TCDD in Body Fat
of Vietnam veterans and Other Men, Retrospective Study of Dioxins and Furans in Adipose
Tissue of Vietnam Era Veterans; Department of Defense (DOD), Epidemiologic Investigation of
Health Effects in Air Force Personnel Following Exposure to Herbicide Orange, Armed Forces
Institute of Pathology Agent Orange Registry of Vietnam veteran Biopsy Tissues; Environmen-
tal Protection Agency (EPA), Report of Assessment of a Field Investigation of Six-Year Sponta-
neous Abortion Rates in Three Oregon Areas of Relation to Forest 2,4,5-T Spray Practices, Na-
tional Pesticide Monitoring Project of Human Adipose Tissue; Department of Agriculture (DOA),
A Case Control Study of the Relationship Between Exposure to 2,4-D and Spontaneous Abor-
tions in Humans, Exposure Measurements of Mixers, Loaders and Applicators of 2,4,-D on
Wheat, Exposure of Forest Workers to Ground Applications of 2,4-D; Department of Health and
Human Services (HHS), Bioassay of Octachlorodibenzo-p-dioxin, Carcinogenesis Bioassay of
2,3,7,8-Tetrachlorodibenzo-p-dioxin in Swiss Webster Mice, Carcinogenesis Bioassay of 2,3,7,8-Te-
trachlorodibenzo-p-dioxin in Osborn-Mendel Rats and B6C3F1 Mice, Bioassay of a Mixture of
1,2,3,6,7,8- and a Mixture of 1,2,3,6,7,8-Hexachloro-dibenzo-p-dioxins for Possible Carcinogenicity,
Comparative Species Evaluation of Chemical Disposition and Metabolism of 2,3,7,8-Tetrachloro-
dibenzofuran in Rat, Monkey, Guinea Pig and Two Strains of Mice, Neurotoxicity of 2,4,-D in
Rodents, Studies of the Chemical Disposition and Metabolism of Octachlorodibenzodioxin, Ef-
fects of Agent Orange Components on Male Fertility and Reproduction, Mutagenicity Studies of
TCDD, Implications of Low Level Exposure of Dioxins, Mechanisms of Toxicity of the Chlorinat-
ed p-dioxins, Research Toward Understanding the Molecular Level Mechanisms of Toxicity of
TCDD and Related Compounds; Synthesis of Selected Tetrachlorodibenzo-p-dioxins and Related
Compounds as Analytical Standards, Matrix Effect and Sub Parts-per-billion Quantitative Anal-
ysis of TCDD by Mass Spectrometry-With Special Reference to Milk, Toxic Actions of Dioxins,
Xenobiotic Induction of Pleiotropic Responses in Liver, Molecular, Biochemical Actions of Chlor-
inated-p-dioxins, Mechanisms for Toxicity of Chlorinated Dibenzo-dioxins, Modular basis of
Dioxin toxicity in Keratinocytes; EPA, Evaluation of Large Scale Combustion Sources, Evalua-
tion of Municipal Waste Combusters, Bacterial Decomposition of TCDD, Investigation of Bioavai-
lability to Fresh Water Fish of TCDD's in Fly Ash, Analysis of Environmental Samples for
PCDD's and PCDF's, DOA, Survey of Phenoxy Herbicide Use by Agricultural Commodity,
Survey of Phenoxy Herbicide Literature, Photolysis of 2,4,5-T, Biological and Economic Assess-
ment of 2,4,5-T and Silvex, TCDD Residue Monitoring in Deer; DOD, Environmental Chemistry
of Herbicide Orange and TCDD; VA, Review of Literature on Herbicides, Including Phenoxy
Herbicides and Associated Dioxins, Urinary 6-Hydroxy Cortisol: Physiological and Pharmacolo-
gic Studies, Effect of TCDD on Lipid Metabolism, Mechanisms of Dioxin Inducted Toxicity Using
the Chloracne Model, Behavioral Toxicity of An Agent Orange Component of 2,4-D, Effects of
2,3,7,8-Tetra-chlorodibenzodioxin on Hepatobiliary Function in Animals, Mechanism of TCDD
Absorption and Toxicity on Lipid and Lipoprotein Metabolism, Metabolism of the Herbicides
Present in Agent Orange and Agent White, TCDD Exposed Rhesus Monkeys: Effects on Behav-
ior and Stress Hormones, Neuromuscular Toxicity of Agent Orange, Effects of Low Dose TCDD
on Mammalian Chromosomes and Liver Cells, Mechanism of Porphyria Caused by TCDD and
Related Chemicals, Effects of Agent Orange on Sleep.
disapprove CDC's activities. The potential to wrongly influence the
outcome of the CDC studies inherent in this conflict of interest was
realized when White House officials directly interceded in the deci-
sion that steered the study in the wrong direction, and ultimately
resulted in its failure.
A July 18, 1986, OMB memorandum provides evidence that it
was the White House, not CDC, that was making decisions during
possibly the most crucial phase of the Agent Orange exposure
study. It was a time when the military records were considered un-
suitable for use in the study, and the entire project was in danger
of being canceled. The cancellation was averted when CDC pro-
posed testing the blood of Vietnam veterans for dioxin traces. The
memorandum presents information about the White House
AOWG's "decision to proceed with the development of a protocol
for testing the dioxin levels in blood for Vietnam ground troop vet-
erans." 82 The memorandum states that the decision on whether to
cancel the study will be influenced by CDC's public testimony
before the House Veterans Affairs Committee on July 31, 1986. Ac-
cording to the memorandum, "It is important that the hearing tes-
timony leave the AOWG with options on the future of the Agent
Orange study. In particular, the testimony should point out that
the blood test may only have a very limited usefulness in studying
individuals
with low levels of dioxin exposure." 83
The memorandum contains an admission that the final decision
was based on the Federal Government's legal culpability, not nec-
essarily the independent work of scientists.
The decision should.take into account the legal implica-
tions of the blood test (the risks associated with testing in-
dividuals and the potential that claims against the govern-
ment would be made by individuals who have higher than
"normal"
dioxin levels in their blood). I have discussed
these issues briefly with Justice attorneys who would be
available to evaluate the pilot study proposal for the legal
risks. Any evaluation should be done prior to the next
AOWG meeting so that the information may be used in
the decision
on the pilot study.
8 4
So fearful was the White House that it might have to compen-
sate Americans exposed to contaminants, it pressured CDC to
refute the blood serum test as a measure of exposure to Agent
Orange, even before it commenced. To avert possible future claims
for compensation from veterans, the memorandum states, "It is im-
portant that testimony and other public comments not associate the
measurement of dioxin in blood with causation. "
8
5 The memoran-
dum recommended that the White House contact the Director of
CDC and encourage him to "be explicit that developing a measure
of dioxin content in blood is neither a link to cause of exposure nor
proof of a cause and effect relationship between dioxin and dis-
ease."
86
s2 "Agent Orange Study," Executive Office of the President, Office of Management and
Budget, to Debbie Steelman from Sarah Ducich, July 18, 1986, p. 1.
s8 Ibid., p. 2.
8
4 Ibid.
8
5 Ibid.
8
6 Ibid.
The memorandum also represents evidence that the CDC study
was doomed to failure by the White House, long before its final
cancellation. Why else would the-conclusions be reached that the
blood dioxin measurement cannot be linked to Agent Orange, even
before the test began? Perhaps most instructive of the White
House's attitude is that it was its panel, the AOWG, that canceled
the Agent Orange study, not CDC.
When the blood test, which the White House had allowed but
classified as inconclusive even before it began, was finally complet-
ed in 1987, apparently showing no higher levels of dioxin in the
blood of Vietnam veterans, the AOWG moved quickly. On August
27, 1987, the Chairman of the AOWG informed the Chairman of
the White House Domestic Pnlicy Council, "it has been concluded
that military records cannot support a valid epidemiological study
of the health effects of Agent Orange exposure on Vietnam veter-
ans . . . I advise you to recommend to the Domestic Policy Council
that the Agent Orange exposure study be cancelled." 87
Two months later, the Director of CDC acknowledged the instruc-
tions to cancel the study. ". . . AOWG has instructed CDC to begin
the process of cancelling the contracts and closing out all activities
related to the Agent Orange Exposure Study." 88
The AOWG had been riding herd over CDC from the beginning
of the study. The AOWG had ordered that all Federal research be
submitted to the White House panel prior to public release. On
September 20, 1984, the panel ordered that ". . . all documents re-
lating to Agent Orange research studies slated for review by any
person or organization outside the Federal Government be submit-
ted first to the Chair, AOWG." 89 After the study was canceled, the
White House maintained its insistence on controlling any Federal
scientific efforts related to Agent Orange. On April 1, 1988, the
Chairman of the AOWG sent a memorandum to members of the
panel, advising them that the future release of any information re-
lated to Agent Orange must be cleared by the AOWG.
The release of.any report, without the review mandated
by the Agent Orange Working Group procedures, could
constitute a serious breach and may undercut our credibil-
ity. Any premature release could cause embarrassment to
the government.
Research findings and conclusions must be submitted to
the AOWG 48 hours prior to release for review, comment
and clearance before going to Congress or the public.
90
W7 Letter from Don M. Newman, Chair Pro Tempore, Domestic Policy Council Agent Orange
Working Group, to Edwin Meese, Chair Pro Tem, Domestic Policy Council, The White House,
August 27, 1987.
18 Hearing, 1989, pp. 55-56. Note to Don Newman, Chair, Domestic Policy Council Agent
Orange Working Group, from James 0. Mason, M.D., Director, Centers for Disease Control, Oc.
tober 28, 1987.
11 "Accomplishing the AOWG Mission," Memorandum From Edward N. Brandt, M.D., Chair,
Pro Tempore, to Members, Cabinet Council Agent Orange Working Group, September 20, 1984.
W0 "Clearance of all studies and press releases by the Domestic Policy Council Agent Orange
Working Group," from Don M. Newman, Chairman, Domestic Policy Council Agent Orange
Working Group, to Members, Domestic Policy Council, Agent Orange Working Group (DPC/
AOWG), April 1, 1988.
E. THE FEDERAL GOVERNMENT HAS SUPPRESSED OR MINIMIZED FIND-
INGS OF ILL HEALTH EFFECTS AMONG VIETNAM VETERANS THAT
COULD BE LINKED TO AGENT ORANGE EXPOSURE
The Agent Orange Exposure Study was one of three parallel
projects assigned to CDC involving the health of Vietnam veterans.
The other two were the Vietnam Experience Study (VES), to deter-
mine the health status of veterans of Vietnam, and the Selected
Cancers Study, conducted to determine if Vietnam veterans had
been susceptible to a group of rare cancers. The key to the White
House's legal strategy of not paying compensation for any type of
toxic contamination was the cancellation of the exposure study.
Once the exposure phase of the studies was canceled, on the
premise that assessing exposure was scientifically impossible, Fed-
eral scientists were able to dismiss any link between diseases and
maladies they discovered and Agent Orange.
Two examples involve the CDC's findings regarding birth defects
involving the offspring of Vietnam veterans and the semen analy-
sis conducted of the Veterans in the VES. CDC found that Vietnam
veterans were more likely than non-Vietnam veterans to report
birth defects. The study also concluded that Vietnam veterans re-
porting exposure to herbicides are at even greater risk of reporting
miscarriages, birth defects, serious health problems, and infant
mortalityY
1
Any possible link between herbicide exposure and the reported
birth defects was dismissed by CDC because of the cancellation of
the exposure study.
The findings from a recent study in which current
dioxin . .. body burdens in Vietnam and non-Vietnam
veterans were assessed suggest that self-reported herbicide
exposure may not be a valid estimate of actual herbicide
exposure. Among Vietnam veterans, there was no evidence
of elevated serum dioxin levels, and no correlation be-
tween average dioxin levels and self-reported exposure to
herbicides in Vietnam. Thus, the herbicide exposure index
used here may reflect the level of concern and anxiety
Vietnam veterans have about the impact of Agent Orange
on their health
and health
of their
children.
9 2
CDC's review of birth records found that the offspring of Viet-
nam veterans were twice as likely to have digestive system birth
defects and were also twice as likely to suffer early neonatal
death.
93
The birth records' review also indicated that the offspring
of Vietnam veterans were more susceptible to cerebrospinal malfor-
mations, such as spina bifida, anencephaly, and hydrocephalus.
CDC explained this problem as an underreporting of the birth de-
fects among non-Vietnam
veterans, rather than an excess among
Vietnam veterans.
9 4
CDC's semen analysis of Vietnam veterans
also found problems:
91 "Health Status of Vietnam veterans, Volume V, Reproductive Outcomes and Child.
Health," Vietnam Experience Study, Centers for Disease Control, January 1989, pp. 21-29.
z Ibid., pp. 41-42.
93 Ibid., pp. 72-73.
94 Ibid.,
pp. 84-85.
Mean
sperm
concentration
was 20% lower
for Vietnam
veterans
than for non-Vietnam
veterans.
* * * **
Specimens
from Vietnam
veterans
had a lower mean
proportion
of morphologically
"normal"
sperm than did
speciments
from non-Vietnam
veterans
. . . specimens
from Vietnam
veterans
were more likely to contain larger
and more tapered sperm. . . . Both the mean cell perime-
ter and the mean length of the major axis of the sperm
cells were significantly
larger for Vietnam
than for non-
Vietnam veterans.
9 5
Although
CDC could not explain definitively
why Vietnam veter-
ans had low sperm counts and altered sperm shapes, it quickly con-
cluded there was no link to Agent Orange:
The possibility
that exposure
to dioxin-containing
herbi-
cides may have affected the sperm of Vietnam veterans is
a potential concern. However, this explanation
seems un-
likely, since in a recent study we found that few Army
ground troops were heavily exposed to dioxin-containing
herbicides.
9 6
The committee finds CDC's conclusion derived from the canceled
exposure study that few ground troops were exposed to Agent
Orange to be a false and misleading interpretation
of its own study.
The -exposure study found that CDC could "not identify any
method for utilizing military records or self-reported exposure to
distinguish between U.S. Army ground combat troops who were
and were not exposed to Agent Orange in Vietnam . . ." 97 The
study did not find that few ground troops were exposed, the reason
used to dismiss any relation between birth defects and low sperm
counts by CDC.
"We never did an Agent Orange study. The Agent Orange expo-
sure component was deemed not to be feasible based upon the vali-
dation study and that study of TCDD exposure was never done,"
admitted Dr. Vernon Houk, Director of the CDC office which con-
ducted the study.
9 8
The canceled CDC study, the study "we never did," according to
Dr. Houk, had become a convenient way for the Federal Govern-
ment to dismiss any link between Agent Orange and health prob-
lems among Vietnam veterans, and deny liability for claims involv-
ing the herbicide. CDC's own Vietnam Experience Study contained
preliminary indications that the liability could be large. CDC's
findings showed that psychological problems such as Post Traumat-
ic Stress Disorder, depression, anxiety, and psychopathology were
more prevalent among Vietnam veterans than non-Vietnam veter-
ans. Vietnam veterans were more likely to have histories of hospi-
talizations, hypertension, benign growths, chloracne, ulcers, hepati-
tis, liver conditions, urinary tract problems, and fertility difficul-
95 "Health Status of Vietnam veterans, Volume III, Medical Examination," Vietnam Experi-
ence Study, Centers for Disease Control, January 1989, p. 207.
96 Ibid., p. 220.
97 Op. cit., p. 4. See footnote 3.
98 Hearing, 1989. Houk testimony, p. 44.
ties. They were more prone to abnormal clinical findings, such as
possible left ventricular hypertrophy, hearing loss, peripheral neu-
ropathy, evidence of past hepatitis B infection, and abnormal labo-
ratory findings, such as gamma-glutamyl transferase, fasting glu-
cose, thyroid-stimulating hormone, and occult stool blood.
9 9
CDC was also aware of other clinical effects that were assumed
to be related to exposure to the chemical constituents of Agent
Orange. Studies of workers exposed to dioxin found a host of ef-
fects.
TABLE OF CLINICAL EFFECTS ASSOCIATED WITH
EXPOSURE TO THE CONSTITUENTS OF AGENT
ORANGE 1
00
Dermatological: Chloracne, porphyria cutanea tarda, hy-
perpigmentation and hirsutism, contact dermatitis.
Hepatic: Liver damage ( ... mild fibrosis, fatty changes,
hemofuscin deposition, parenchymal-cell degeneration), in-
creased serum hepatic enzyme levels.
Other internal: Disorders of fat metabolism, disorders of
carbohydrate metabolism, cardiovascular disorders ...
elevated serum cholesterol levels, abdominal pains and di-
arrhea, pancreatic disorders.
Respiratory: ... SMR for lung cancer, dyspnea.
Genito-urinary: Hemorrhagic cystitis, hematuria protein-
uria, SMR for malignant neoplasms of the genito-urinary
organs and the bladder.
Neuro-muscular: Asthenia, fatigue, headaches, sleep dis-
turbances, sweating, irritability, confusion, peripheral
neural damage, polyneuropathies, delayed peripheral
nerve conduction, encephalomyelitis sensorial impair-
ments. . . . Hyperesthesia, impotence . . . lassitude and
weakness, vertigo, ataxia, hyporeflexia, rigidity, myotonia,
pain in the extremities, joint pain, severe aches in the
lower extremities, difficulty in muscular and mental co-
ordination, profound muscular weakness, neuroarthro-
pathy.
Other: Loss in body weight, thymic atrophy, decreased
immune competence.
At the time the Agent Orange exposure study had been canceled,
CDC's reputation for objectivity and accuracy was beyond reproach.
Thus, its false interpretation that few ground troops had been ex-
posed to Agent Orange was accepted as gospel by the scientific
community. Other Federal agencies, particularly the VA, were
quick to parrot CDC's work and apply it to other research. For ex-
ample, a 1987 report by the VA on soft tissue sarcoma and military
service in Vietnam found that subgroups of ground troops studied
who appeared to have higher opportunities for exposure to Agent
Orange were at greater risk of contracting the rare cancer, and
99 Ibid. Testimony of Dr. Steven Stellman, pp. 178-179.
100 "Observed Clinical Effects of 'Agent Orange,"' Memorandum from Dennis Smith, Visiting
Scientist, Agent Orange Projects, Centers for Disease Control, to Dan L. McGee, Ph.D., Senior
Statistician, Agent Orange Projects, January 30, 1985, pP. 1-2.
that Army veterans who served combat missions showed a 2.6
times elevated risk of soft tissue sarcoma, as compared to veterans
who did not serve in combat. 1
The study concluded that "the possibility of a modestly increased
risk of STS associated with Agent Orange exposure in Vietnam
among select groups of Vietnam veterans can be neither confirmed
nor ruled out in this study. Additional studies using better charac-
terization of exposure are needed to answer this question." 102 The
researchers had planned to conduct such a study. "Initially, an
elaborate computer matching of troop location to recorded aerial
spray missions . . . was planned. However, an expert government
panel has subsequently determined that military records alone
could not be used to locate troops with enough precision to allow a
scientifically valid estimate of the likelihood of exposure to herbi-
cides." 103 The expert Government panel to which the researchers
referred is the AOWG, which based its conclusions on the CDC ex-
posure study.
The VA also downplayed the findings of a mortality study it con-
ducted of Army and Marine Corps veterans of Vietnam. The study
found that Marines serving in Vietnam had statistically significant
higher elevations of lung cancer and non-Hodgkin's lymphoma, two
diseases associated with exposure to phenoxy herbicides.
The veterans who served in the Marine Corps in Viet-
nam were seen to have a statistically significant ...
excess of non-Hodgkin's lymphoma when compared with
Marines who did not serve in Vietnam.
Non-Hodgkin's lymphoma has been associated with ex-
posure to phenoxy herbicides, arsenicals, dapsone, and cer-
tain viruses. The men who served in Vietnam had the po-
tential for exposure to all of these agents. Agent Blue, a
herbicide used in Vietnam, was an organic arsenical com-
pound and dapsone, a sulfone, was used as an antimalarial
drug by some of the troops in Vietnam. Dapsone has been
shown to cause lymphomas in laboratory animals. Dapsone
was given mainly to troops stationed in I Corps and the
central highland areas of Vietnam where falciparum ma-
laria was prevalent. Most of the Marines in Vietnam
served in I Corps.'
0 4
Dr.-Lawrence Hobson, the Director of the VA's Office of Environ-
mental Medicine, testified that the study did not find significantly
higher levels of non-Hodgkin's lymphoma and lung cancer. He con-
tended that a "greater proportion" of the subjects studied died
from those diseases.'
0 5
101 "Soft Tissue Sarcoma and Military Service in Vietnam: A Case-Control Study," Han Kang,
Dr. P.H., Franz Enziger, M.D., Patricia Breslin, Sc.D., Michael Fel, M.S., Yvonne Lee, M.S., and
Barclay Shepard, M.D., JNCI, Vol. 79, No. 4, October 1987.
102 Ibid.
103 Ibid.
104 "Proportionate Mortality Study of US Army and US Marine Corps Veterans of the Viet-
nam War,' Patricia Breslin, ScD, Han K. Kang, DrPH; Yvonne Lee, MSc; Vicki Burt, ScM; and
Barclay M. Shepard, MD, Journal of Occupational Medicine, Vol. 30, No. 5, May 1988.
o10 Hearing, 1989. Testimony of Lawrence Hobson, M.D., Director, Office of Environmental
Medicine, Veterans Administration, p. 138.
It isn't fair to say that you had more deaths among the
marines or a higher rate of death or a higher incidence of
death from soft tissue sarcoma among those marines.
The only thing you can say is that more of the ones who
died seemed to have died from soft tissue sarcoma than
from say heart disease or lung disease or some other form
of
cancer.
'
0
6
Dr. Hobson's attempt to minimize the VA's finding that Marines
were at greater risk of certain diseases obfuscates the facts uncov-
ered by the Agency's research. The committee believes that Feder-
al scientists have an obligation to use pertinent information re-
garding the health of Vietnam veterans as a tool for further explo-
ration of the risks associated with service in Vietnam. Instead, the
VA has consistently attempted to dismiss the relevance of the in-
formation to allay the concerns of Vietnam veterans.
Admiral Elmo R. Zumwalt, former Chief of Naval Operations,
and more recently a special advisor to the Secretary of Veterans
Affairs on Agent Orange, testified before the subcommittee of a
conspiracy to withhold the truth about the herbicide from the
public.
The sad truth which emerges from my work is not only
that there is credible evidence linking certain cancers and
other illnesses with Agent Orange, but that government
and industry officials credited with examining such link-
age intentionally manipulated or withheld compelling in-
formation of the adverse health effects associated with ex-
posure to the toxic contaminants contained in Agent
Orange.
'
0 7
Dr. Daniel Thau Teitelbaum, a toxicologist who, through litiga-
tion involving herbicide manufacturers, has reviewed their internal
records, testified that the companies had withheld information
about toxic contaminants in their products. He said the contami-
nants, called xanthen-9-ones, are toxic in animal species and are
contained in 2,4-D, a chemical prevalent in herbicides, including
homeowner's weed killers.'
0 8
Dr. Teitelbaum testified that the
presence of these contaminants had not been reported to the Envi-
ronmental Protection Agency or State registration agencies.'
0 9
An epidemiologist
with the National Cancer Institute, Dr. Shelia
Hoar Zahm, testified that a recent study she conducted, scheduled
for publication in September 1990, has proven 2,4-D to be a carcino-
gen in agricultural
workers who use herbicides with that ingredi-
ent.11
2,4-D was a major ingredient
of Agent Orange, but has not been
explored in Federal research regarding the effects of the herbicide
on Vietnam veterans.
06 Ibid.
107 Hearing, 1990. Prepared statement of Adm. Elno R. Zumwalt, Jr., p. 4.
108 Ibid.
109 Ibid.
110 Ibid.
V. RECOMMENDATIONS
A. CONGRESS SHOULD REQUIRE THE DEPARTMENT OF DEFENSE TO
CREATE AN AGENT ORANGE EXPOSURE INDEX
In light of the committee's determination that the CDC Agent
Orange Exposure Study was flawed and canceled based on errone-
ous information, the committee believes that an exposure index,
matching troop movement records to spray data, still needs to be
and can be developed. The committee believes that the Department
of Defense has the expertise and resources to conduct such an expo-
sure index-provided adequate, independent peer review is provid-
ed by expert bodies not associated with the Department-and rec-
ommends that Congress mandate the development of an index.
B. WHEN AN ADEQUATE EXPOSURE INDEX IS DEVELOPED, THE FEDERAL
GOVERNMENT SHOULD CONTRACT THROUGH THE NATIONAL ACADEMY
OF SCIENCES FOR A PRIVATE, INDEPENDENT EPIDEMIOLOGICAL STUDY
MATCHING THE HEALTH OUTCOMES OF VIETNAM VETERANS AGAINST
THE EXPOSURE INDEX
The development of an exposure index would be only the first
part of a study determining the association or non-association of ill-
ness to Agent Orange exposure. Given the bias demonstrated by
Federal agencies in the past, the committee believes veterans
would be better served by an epidemiological study prepared by ob-
jective scientists. Using the index developed by the Department of
Defense, it is suggested that a private organization be awarded a
contract, to be administered by the National Academy of Sciences,
to conduct a proper epidemiological study.
C. ALL SCIENTIFIC RESEARCH CONDUCTED BY FEDERAL AGENCIES
SHOULD BE DONE WITHOUT INTERFERENCE FROM FEDERAL COMPO-
NENTS OUTSIDE THEIR RESPECTIVE AGENCIES
The White House made crucial decisions affecting the course of
CDC's Agent Orange research, and the outcome. This kind of politi-
cal interference is inappropriate and casts doubts on the integrity
and credibility of Federal research. While components of the Feder-
al Government should be free to make policy determinations based
on Federal research, they must not be allowed to have policy deter-
mine the outcome of such research. The committee recommends
that all scientific research conducted in the future by Federal agen-
cies be done independently from the White House or other political
organizations. If regulations do not already require such independ-
ence, then each agency should promulgate rules to prevent political
interference.
DISSENTING VIEWS OF HON. RICHARD K. ARMEY, HON.
FRANK HORTON, HON. HOWARD C. NIELSON, HON. J.
DENNIS HASTERT, HON. JON L. KYL, AND HON. CHUCK
DOUGLAS
Agent Orange is a highly emotional and controversial issue with
strong political pressure for answers. We have deep sympathy for
veterans and their families who believe their suffering is due to
Agent Orange exposure, and we want the record to be clear that
we remain committed to compensating American veterans for serv-
ice-connected disabilities.
Regrettably, however, instead of advancing the debate on Agent
Orange in a positive direction, the Human Resources Subcommit-
tee has abused this issue in order to launch an ideological assault
upon a Republican White House with which it has never agreed.
Consequently, constructive suggestions for further review and de-
tached review of science are given a back seat to unsubstantiated
charges of a political coverup, and the Committee Report seems
more concerned with writing the President out of the Constitution
than it does with evaluating the science surrounding Agent
Orange. Our view is buttressed by the Subcommittee Chairman's
steadfast refusal to rethink his charges against the White House
and work with us to develop an Agent Orange report that all mem-
bers of this Committee could support. In fact, the draft report was
already printed in final galley form by the Government Printing
Office before minority members were ever given a copy.
Our concerns are not limited to procedural grounds, however,
and we have many specific reservations with the report. In addi-
tion to being highly emotional and controversial, studying Agent
Orange is also an extraordinarily complex task, and these enor-
mous complexities cannot be as readily dismissed as they were by
the Committee Report. The fact is that while CDC was charged
with studying the exposure of Vietnam veterans to Agent Orange,
determining exposure is just not as easy as it sounds. For example,
when we refer to exposure, do we mean opportunity for exposure
or actual exposure to Agent Orange. The distinction is crucial.
If we're referring to actual exposure, several additional factors
come into play. What was the quantity of the exposure? How long
was the veteran exposed? Was exposure on the skin or clothes?
How long did exposure occur after'spraying? Was the exposure re-
peated? The answer to these questions were important factors in
the National Cancer Institute studies cited by the Committee
Report to substantiate
the link between herbicides and certain can-
cers. Good science suggests that these questions should not have
been ignored in the Agent Orange study. Furthermore,
the answers
to these questions
are extraordinarily
difficult to ascertain twenty
years after actual exposure. Consequently,
when we recognize
that
the CDC study was premised
upon determining
actual exposure,
and after CDC made several attempts to accurately determine ex-
posure, CDC's conclusion that a scientifically valid exposure study
could not be done is not as dubious or devious as some would sug-
gest.
Conversely, if you're conducting an exposure opportunity study
as the Committee Report implies the CDC should have done, you
invite serious misspecification errors which could render further
study useless. Misspecification means that we identify persons as
exposed when they were not exposed and vice versa. CDC was re-
luctant to use this approach because it believed it would be a dis-
service to veterans to measure the existence or non-existence of
health effects on persons who could not be conclusively determined
to have been exposed to Agent Orange.
While proponents of an exposure opportunity study readily admit
that this approach creates an opportunity for misspecification, they
downplay this concern by asserting that these errors will cancel
each other out; however, this is not an acceptable response. Epide-
miologists who argue that misspecification errors will cancel each
other out are guilty of using the statistician's equivalent of the
punt. There is no scientifically justifiable explanation for asserting
that misspecification errors are random, and the cancelling out ar-
gument is nothing more than a convenient way to dismiss errors
that can't otherwise be explained away.
The bottom line is that any study put together by CDC would be
subject to legitimate attack for its shortcomings. In fact, we suspect
that if CDC conducted an exposure opportunity study concluding
that there is no relationship between exposure to Agent Orange
and adverse health effects, the Human Resources Subcommittee
would be the first in line to attack the very model it now supports
because of its misspecification weaknesses.
The Committee Report's author would probably seek to refute
our arguments by pointing to the independent National Academy
of Sciences' (NAS) Institute of Medicine (IOM) report which raises
the same criticisms as the Committee Report regarding the CDC's
handling of troop data. However, this report is mischaracterized in
the Committee Report and is inappropriately used to support the
Human Resources Committee's political conclusions.
On July 19th, Institute of Medicine President, Samuel Thier, re-
sponded by letter to an article appearing in the July 23rd issue of
Time Magazine (Time reaches the newsstands before its published
date) entitled "A Cover-Up on Agent Orange." This article reported
similar coverup and manipulation charges as those raised in the
Committee Report. However, in his letter, Dr. Thier says that the
Time article,
incorrectly says CDC suppressed a 1986 report from the In-
stitute of Medicine (IOM) of the National Academy of Sci-
ences. The document in question was actually an internal
report . . . [and] the report did not criticize the CDC study;
it was about the Defense Department, not the CDC.
More important, however, Dr. Thier asserted that the report con-
cluded,
[The] Defense Department was capable of handling troop
data, but also expressed concerns that these might not be
accurate measures of exposure to Agent Orange. [Emphasis
ours.]
The 1987 IOM pre-publication review of CDC's Agent
Orange Validation Study was critical of the CDC's work
because the committee was explicitly asked to critique the
agency's draft report. However, the IOM report also states
that the CDC design, methods and analyses were well con-
ceived, appropriately executed, and clearly presented. [Em-
phasis ours.]
Unfortunately, the Committee Report does not end with its criti-
cisms of the CDC study, but chooses to extend its outrageous accu-
sations of coverup to the Reagan White House.
After subpoenaing virtually every imaginable document pre-
pared, reviewed or witnessed by the White House regarding Agent
Orange, the most the Committee Report can tell us is that the
White House opposed legislation which presumed that simple expo-
sure alone to Agent Orange constituted a service-connected disabil-
ity meriting compensation. We are then asked to conclude that this
opposition provided the impetus for its plot to cover up the truth
about Agent Orange. This conclusion is simply not supported by
the facts.
First, we need to acknowledge that compensating veterans for
simple exposure to Agent Orange is a political decision, not a scien-
tific one. Scientific information made available to the White House,
from both inside and outside of government, has failed to establish
any supportable conclusive link between simple exposure to Agent
Orange and adverse long-term health effects. Therefore, the practi-
cal effect of the White House's legislative position was to oppose
spending billions of tax dollars to compensate for an event that vir-
tually no scientist has linked to long-term disabilities.
Consequently, we find it highly ironic and extremely troublesome
that while the White House is repeatedly attacked in the Commit-
tee Report for ignoring science and using political manipulation,
when the White House makes a decision clearly supported by sci-
ence, this decision is seized upon by the majority as the key evi-
dence supporting a political coverup. We wonder who's the real one
playing politics.
Next, we find it totally implausible for the White House to mas-
termind a coverup on its own when Congress was intimately in-
volved in the Agent Orange proceedings since their inception. The
Office of Technology Assessment (OTA) and the House Veterans
Affairs Committee under the able leadership of its Chairman
Sonny Montgomery had either participated in or had overseen the
progress of the CDC study, and both supported the CDC's activities
and conclusions.
In a House Veterans Affairs Subcommittee press release dated
March 29, 1990, Chairman Montgomery is quoted,
Agent Orange has been and remains an emotional and
controversial
issue, but the findings of [the CDC] and all
reputable studies are convincing: There is nothing to show
decisively that there are long-term adverse health effects
as a result of exposure to Agent Orange.
Regarding the report prepared by Admiral Zumwalt and relied
on so extensively by the Human Resources Subcommittee, the OTA
had this to say in a July 23rd, 1990 letter from John Gibbons to
Congressman Montgomery,
The report seems to take the form more of a legal brief
than of a scientific review of evidence.. . . Based on a
review of the areas in which OTA has been involved, we
conclude that many of the assertions made in the report
supporting a conclusion that Agent Orange is responsible
for a wide range of health problems among Vietnam veter-
ans, are incorrect. These are not mainly matters of differ-
ing opinion, but matters of fact-what did or did not
happen. For those aspects about which OTA staff have de-
tailed knowledge, it appears that Admiral Zumwalt's argu-
ments are based, in many instances, on faulty information
or incorrect interpretation of data.
Therefore, the Committee Report cannot accuse the White House
of engineering a coverup without also implicating Congress and
several highly respected members of this body, and we refuse to do
so. Charging Members of Congress for a coverup is simply ludicrous,
and so is charging the White House as well.
We suspect that the real motivation behind the Human Re-
sources Subcommittee Chairman's zealous attack on the White
House is to lay the groundwork for his final recommendation and
real goal: writing Republican Presidents out of the Constitution. By
recommending that all scientific research conducted in the future
be done independently from the White House or other "political"
organizations, this goal is made clear.
However, the last time we looked, Article II, Section I of the Con-
stitution of the United States still vested executive power in the
President of the United States. Notwithstanding the desires of the
Committee Report's authors, we are unaware of any Constitutional
Amendments to change this "political" office.
This final recommendation of the Committee Report is most
troubling because it highlights an ongoing double standard applied
by the Human Resources Subcommittee regarding the work of sci-
entists, both inside and outside of government. The implications of
this double standard are clear: If you're a scientist who agrees with
the political conclusions of the Human Resources Subcommittee
Chairman, your views will be put forth as gospel, and no one has a
right to question them. On the other hand, if you're a scientist who
disagrees with the Chairman's political conclusions, you'll either be
lucky enough to be ignored, or you'll face the unfortunate pros-
pects of having your views attacked and your integrity questioned
before a publicly held Congressional hearing.
We are simply tired of seeing Administration officials constantly
dragged through the Human Resources Subcommittee wringer
under the banner of science over politics, when the Subcommittee
is the one so clearly guilty of playing politics itself.
43
Veterans deserve much better than what this report seeks to
offer.
RICHARD K. ARMEY.
FRANK HORTON.
HOWARD C. NIELSON.
J. DENNIS HASTERT.
JON L. KYL.
CHUCK DOUGLAS.
0