Gulf War Illnesses: Procedural and Reporting Improvements Are Needed in
DOD's Investigative Processes (Chapter Report, 02/26/99,
GAO/NSIAD-99-59).

Pursuant to a congressional request, GAO provided information on the:
(1) Department of Defense's (DOD) progress in establishing an
organization to address Gulf War illnesses issues; and (2) thoroughness
of DOD's Office of the Special Assistant for Gulf War Illnesses'
(OSAGWI) investigations into and reporting on incidents of veterans'
potential exposure to chemical or biological warfare agents during the
Gulf War.

GAO noted that: (1) DOD has made progress in carrying out its mandate to
comprehensively address Gulf-War illnesses-related issues; (2) it has
assisted veterans through its outreach program by clearing large
backlogs of veterans' inquiries, using a toll-free hot line, setting up
a Web site, and publishing a newsletter; (3) in addition, it has
assisted veterans in obtaining medical examinations and other services
at DOD and Department of Veterans Affairs (VA) facilities; (4) through
the course of its investigations and other work, OSAGWI has identified
needed improvements in DOD's equipment, policies, and procedures and has
worked with various DOD agencies to implement changes designed to
provide better protection to U.S. servicemembers on a contaminated
battlefield; (5) OSAGWI generally applied appropriate investigative
procedures and techniques in conducting its work; (6) however, GAO found
that three of the six case narratives it reviewed contained weaknesses
such as failures to follow up with appropriate individuals to confirm
key evidence, to identify or ensure the validity of some evidence, to
include some important information, and to interview some key witnesses;
(7) in the remaining three cases, OSAGWI conducted its investigations
without evidence of these weaknesses; (8) in all six cases, OSAGWI
missed an opportunity to perform more complete investigations because it
did not take advantage of potentially valuable sources of relevant
information in DOD and VA clinical databases; (9) GAO does not know
whether the investigatory and reporting weaknesses it found in its
review of these six cases might also exist in the cases that OSAGWI
later investigated; (10) despite these weaknesses, GAO agreed with
OSAGWI's conclusions about the likelihood of the presence of chemical
warfare agents in five of the six cases it reviewed; (11) the one
exception involved a potential exposure of U.S. Marine Corps personnel
to a chemical warfare agent during a minefield breaching operation; (12)
OSAGWI concluded that exposure in this case was unlikely; (13) however,
GAO found that OSAGWI had overlooked some information it had in its
possession and also did not include all relevant information in its case
narrative; (14) after reviewing the overlooked information and
considering all relevant information OSAGWI had in its files, GAO
believes that OSAGWI should reassess the likelihood of exposure in this
case; and (15) GAO believes that the lack of effective quality assurance
policies and practices in OSAGWI's investigating and reporting processes
contributed to the weaknesses noted.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  NSIAD-99-59
     TITLE:  Gulf War Illnesses: Procedural and Reporting Improvements 
             Are Needed in DOD's Investigative Processes
      DATE:  02/26/99
   SUBJECT:  Hazardous substances
             Safety standards
             Investigations by federal agencies
             Medical information systems
             Chemical warfare
             Biological warfare
             Reporting requirements
             Disease detection or diagnosis
             Medical research
             Veterans
IDENTIFIER:  DOD Comprehensive Clinical Evaluation Program
             Kuwait
             VA Persian Gulf War Health Registry
             Persian Gulf War
             DOD Persian Gulf War Health Surveillance System
             Iraq
             Al Jubail (Saudi Arabia)
             Desert Storm
             
******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved.  Major          **
** divisions and subdivisions of the text, such as Chapters,    **
** Sections, and Appendixes, are identified by double and       **
** single lines.  The numbers on the right end of these lines   **
** indicate the position of each of the subsections in the      **
** document outline.  These numbers do NOT correspond with the  **
** page numbers of the printed product.                         **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************


Cover
================================================================ COVER


Report to the Honorable Lane Evans, Ranking Minority Member,
Committee on Veterans Affairs, House of Representatives

February 1999

GULF WAR ILLNESSES - PROCEDURAL
AND REPORTING IMPROVEMENTS ARE
NEEDED IN DOD'S INVESTIGATIVE
PROCESSES

GAO/NSIAD-99-59

Gulf War Illnesses

(703223)


Abbreviations
=============================================================== ABBREV

  ASP - Ammunition Supply Point
  CCEP - Comprehensive Clinical Evaluation Program
  DOD - Department of Defense
  EPMU - Environmental and Preventive Medicine Unit
  NMCB - Naval Mobile Construction Battalion
  OSAGWI - Office of the Special Assistant for Gulf War Illnesses
  VA - Department of Veterans Affairs

Letter
=============================================================== LETTER


B-279127

February 26, 1999

The Honorable Lane Evans
Ranking Minority Member
Committee on Veterans Affairs
House of Representatives

In response to your request, we have reviewed the operations of the
Department of Defense's Office of the Special Assistant for Gulf War
Illnesses.  This report focuses on the thoroughness of this Office's
investigations into and reporting on veterans' potential exposure to
chemical or biological agents during the Persian Gulf War. 

We are sending copies of this report to the Chairmen and Ranking
Minority Members of the House Committee on Appropriations, the House
Committee on Government Reform and Oversight, the Senate Committee on
Appropriations, and the Senate Committee on Governmental Affairs and
to the Director, Office of Management and Budget.  We will make
copies available to others on request. 

If you or your staff have any questions concerning this report,
please contact me on (202) 512-5140.  Major contributors to this
report are listed in appendix IV. 

Sincerely yours,

Mark E.  Gebicke
Director, Military Operations
 and Capabilities Issues


EXECUTIVE SUMMARY
============================================================ Chapter 0


   PURPOSE
---------------------------------------------------------- Chapter 0:1

Many servicemembers who served in the Persian Gulf War have
subsequently experienced health problems such as fatigue, muscle and
joint pain, gastrointestinal complaints, headaches, memory loss, and
sleep disturbances.  Whether these health problems are related to
these servicemembers' exposures to chemical, biological, or
environmental agents during their Gulf War service has been a topic
of much controversy.  To ensure that all issues related to Gulf War
illnesses were comprehensively addressed, the Department of Defense
(DOD) established the Office of the Special Assistant for Gulf War
Illnesses (OSAGWI) in November 1996. 

To determine whether DOD is diligently addressing issues related to
Gulf War illnesses, the Ranking Minority Member of the House
Committee on Veterans Affairs asked GAO to examine selected OSAGWI
operations.  Specifically, GAO's objectives were to (1) describe
DOD's progress in establishing an organization to address Gulf War
illnesses issues and (2) evaluate the thoroughness of OSAGWI's
investigations into and reporting on incidents of veterans' potential
exposure to chemical or biological warfare agents during the Gulf
War. 


   BACKGROUND
---------------------------------------------------------- Chapter 0:2

More than 100,000 Gulf War veterans have participated in health
examination programs established by DOD and the Department of
Veterans Affairs (VA).  Of those examined, nearly 90 percent have
reported a wide array of health complaints and disabling conditions. 
Some veterans suspect that their health problems may be linked to
chemical or biological warfare agents that Iraq may have used during
the Gulf War.  Other causes, such as stress, smoke from oil well
fires, reactions to pesticides or vaccines, and exposure to depleted
uranium munitions, have also been suggested as causes of these
illnesses.  Research to better identify the causes is ongoing but
will not be completed for years. 

Following the Gulf War, DOD claimed that chemical weapons were not
present in the Gulf War theater.  However, the Central Intelligence
Agency disclosed in 1995 that chemical weapons were found at an
ammunition storage site at Khamisiyah, Iraq.  Following an
investigation, DOD acknowledged in July 1997 that U.S.  troops might
have been exposed to a chemical warfare agent at Khamisiyah when
demolitions were used there to destroy Iraqi rockets.  Other
incidents involving potential chemical warfare agent exposures have
been cited by veterans in testimonies before various congressional
committees.  Consequently, some have called into question DOD's
credibility on Gulf War illnesses issues. 

In November 1996, DOD established OSAGWI to restore public confidence
in DOD's efforts to deal with Gulf War illnesses issues.  OSAGWI has
focused its efforts on (1) establishing effective two-way
communications with veterans and veterans groups, (2) investigating
and reporting on incidents of possible chemical warfare agent
exposures, and (3) applying lessons learned from the Gulf War
experience to better protect U.S.  servicemembers on a contaminated
battlefield. 

Each OSAGWI investigation into possible exposures of servicemembers
to chemical warfare agents results in a summation document called a
case narrative.  The case narrative, a document updated as new
evidence becomes known, contains all important investigative facts
and OSAGWI's assessment--in terms of "definitely," "likely,"
"indeterminate," "unlikely," or "definitely not"--of the likelihood
that servicemembers were exposed to chemical or biological warfare
agents.  The standard OSAGWI used for its assessments was whether all
available facts would lead a reasonable person to conclude that a
chemical or biological warfare agent was or was not present.  At the
time GAO began its evaluation, OSAGWI had issued eight case
narratives.  OSAGWI pursued these cases first because they involved
incidents that were the most prominent and controversial. 

GAO evaluated six of these eight investigations.  GAO did not review
the case narrative about the alleged exposure at Khamisiyah because
it was already being heavily reviewed by other organizations, such as
the Presidential Advisory Committee on Gulf War Veterans' Illnesses
and the Senate Committee on Veterans Affairs' Special Investigation
Unit.  GAO also did not review the Possible Chemical Agent on SCUD
Missile Sample case narrative because it appeared to be less
controversial than the other case narratives.  In conducting its
evaluations, GAO (1) traced each statement in these reports to its
underlying supporting documentation in OSAGWI files, (2) reviewed
OSAGWI documentation associated with the incident to determine if all
relevant information was included in the report, (3) contacted key
sources of information to verify the accuracy and completeness of the
information these sources provided to OSAGWI, (4) independently
sought other sources of information, and (5) contacted key
participants not originally interviewed to determine if relevant
information was available that might affect OSAGWI's assessment of
possible exposures to chemical warfare agents. 



   RESULTS IN BRIEF
---------------------------------------------------------- Chapter 0:3

DOD has made progress in carrying out its mandate to comprehensively
address Gulf War illnesses-related issues.  It has assisted veterans
through its outreach program by clearing large backlogs of veterans'
inquiries, using a toll-free hot line, setting up a Web site, and
publishing a newsletter.  In addition, it has assisted veterans in
obtaining medical examinations and other services at DOD and VA
facilities.  Through the course of its investigations and other work,
OSAGWI has identified needed improvements in DOD's equipment,
policies, and procedures and has worked with various DOD agencies to
implement changes designed to provide better protection to U.S. 
servicemembers on a contaminated battlefield.  OSAGWI generally
applied appropriate investigative procedures and techniques in
conducting its work.  However, GAO found that three of the six case
narratives it reviewed contained weaknesses such as failures to
follow up with appropriate individuals to confirm key evidence, to
identify or ensure the validity of some evidence, to include some
important information, and to interview some key witnesses.  In the
remaining three cases, OSAGWI conducted its investigations without
evidence of these weaknesses.  In all six cases, OSAGWI missed an
opportunity to perform more complete investigations because it did
not take advantage of potentially valuable sources of relevant
information in DOD and VA clinical databases.  GAO does not know
whether the investigatory and reporting weaknesses it found in its
review of these six cases might also exist in the cases that OSAGWI
later investigated. 

Despite these weaknesses, GAO agreed with OSAGWI's conclusions about
the likelihood of the presence of chemical warfare agents in five of
the six cases it reviewed.  The one exception involved a potential
exposure of U.S.  Marine Corps personnel to a chemical warfare agent
during a minefield breaching operation.  OSAGWI concluded that
exposure in this case was "unlikely." However, GAO found that OSAGWI
had overlooked some information it had in its possession and also did
not include all relevant information in its case narrative.  After
reviewing the overlooked information and considering all relevant
information OSAGWI had in its files, GAO believes that OSAGWI should
reassess the likelihood of exposure in this case.  There is potential
that this case could be more appropriately assessed as
"indeterminate."

GAO believes that the lack of effective quality assurance policies
and practices in OSAGWI's investigating and reporting processes
contributed to the weaknesses noted.  Although OSAGWI has taken steps
to improve its quality assurance procedures, certain features should
be incorporated to ensure that all of its investigations are
thoroughly conducted and accurately reported. 


   PRINCIPAL FINDINGS
---------------------------------------------------------- Chapter 0:4


      DOD HAS MADE PROGRESS IN
      ESTABLISHING AN ORGANIZATION
      TO ADDRESS GULF WAR
      ILLNESSES ISSUES
-------------------------------------------------------- Chapter 0:4.1

DOD established OSAGWI to repair the credibility problems it faced
regarding its past efforts to address Gulf War illnesses issues.  It
provided OSAGWI with an operating authority much broader than its
predecessor, the Persian Gulf Illnesses Investigation Team, namely,
to coordinate all aspects of DOD's programs concerning Gulf War
illnesses.  Compared with its predecessor, OSAGWI represents a
significant increase in resources directed toward investigations and
outreach efforts.  For example, in 1996, the Persian Gulf Illnesses
Investigation Team operated with a staff of 12 persons and a budget
of $4.1 million.  In contrast, OSAGWI had a staff of 200 persons as
of October 9, 1998, and a fiscal year 1998 budget of $29.4 million. 
In addition, while the Persian Gulf Illnesses Investigation Team
reported to the Assistant Secretary of Defense for Health Affairs,
OSAGWI reports directly to the Deputy Secretary of Defense. 

DOD has made progress in addressing Gulf War illnesses issues.  To
improve communications with veterans, OSAGWI has established the
means to receive input from and provide information to veterans. 
Within its first year of operation, OSAGWI successfully cleared a
backlog of 1,200 veterans' inquiries through personal telephone
calls, and received an additional 1,200 letters and 2,700 E-mail
messages.  By January 1, 1999, OSAGWI had received 2,850 letters and
4,906 E-mail messages.  OSAGWI officials met with the public and
veterans at 18 town hall meetings and appeared at 41 national
veterans conventions.  Its Internet site reportedly receives over
60,000 inquiries each week, and over 12,000 individuals receive
OSAGWI's bimonthly newsletter.  OSAGWI also refers veterans to
various sources of medical services.  Finally, OSAGWI communicates
directly with veterans that are affected by its investigations. 
After OSAGWI completes an investigation and publishes the
corresponding case narrative, it sends to each directly affected
veteran a letter that contains a synopsis of the investigation's
results. 

OSAGWI's mission requires that it advise the Secretary of Defense on
changes needed in military equipment, policies, and procedures in
order to better protect servicemembers during operations on a
contaminated battlefield.  OSAGWI has identified several areas
needing improvement on the basis of its experience in investigating
and reporting on possible chemical, biological, or environmental
exposures.  OSAGWI is working with DOD and other executive branch
agencies to implement these lessons learned.  For example, OSAGWI was
instrumental in prompting the Deputy Secretary of Defense to issue a
requirement that the military services review their depleted uranium
training programs.  These programs are important in addressing
potential health problems related to the use of depleted uranium in
armor and ammunition.  We did not review the impact this activity has
had on making changes within DOD.  However, in October 1998, OSAGWI
established a directorate to focus on ensuring that lessons learned
are implemented. 


      INVESTIGATIVE AND REPORTING
      PROCEDURES HAVE VARIOUS
      WEAKNESSES
-------------------------------------------------------- Chapter 0:4.2

GAO found procedural, investigative, or reporting problems in three
of the six cases it reviewed.  These weaknesses were not evident in
the other three cases.  Specifically, it found that OSAGWI
investigators sometimes failed to follow up with appropriate
individuals to confirm key evidence, identify or ensure the validity
of key evidence, include important information, and interview key
witnesses.  Despite these weaknesses, the preponderance of evidence
led GAO to agree with the conclusions in OSAGWI case narratives
concerning the presence of chemical warfare agents in all but one of
the six cases GAO reviewed.  This one exception involved a potential
exposure of U.S.  Marine Corps personnel during a minefield breaching
operation.  OSAGWI concluded that an exposure in this case was
"unlikely." However, GAO found that this case narrative did not
include some key information contained in OSAGWI files. 
Specifically, OSAGWI had information regarding the presence of
artillery fire that contradicted one of its primary
determinations--that no artillery fire or chemical mines were present
and therefore no means of chemical warfare agent delivery existed. 
Also, OSAGWI did not include information that chemical detection
paper attached to a vehicle used in the operation changed color,
indicating the potential presence of a chemical warfare agent.  After
reviewing the overlooked information and considering all relevant
information OSAGWI had in its files, GAO concluded that reassessment
is needed and that the probability of exposure might more
appropriately be assessed as "indeterminate."

The other two of the three cases in which GAO found investigative or
reporting weaknesses involved (1) a possible exposure of a
servicemember to a mustard agent and (2) a possible exposure of
servicemembers to chemical agents in Al Jubayl, Saudi Arabia.  In the
case involving the potential exposure of a servicemember to a mustard
agent during an inspection of an Iraqi bunker complex, OSAGWI did not
follow up adequately to confirm whether an in-theater urinalysis test
was administered.  GAO found insufficient evidence to support the
existence of such a test.  Moreover, OSAGWI did not establish whether
clothing tested for chemical warfare agent in this case actually
belonged to the individual allegedly exposed.  Finally, OSAGWI
reached its conclusion without interviewing some key witnesses. 
Despite these weaknesses, the evidence in this case supported
OSAGWI's conclusion that exposure to a chemical warfare agent was
"likely." In the case involving potential exposure to chemical agents
in Al Jubayl, Saudi Arabia, GAO found that the available evidence
generally supported OSAGWI's conclusions.  However, OSAGWI did not
include important information that would have made the case narrative
more complete.  Had OSAGWI included this information, it would have
avoided any appearance that it had not completely reported what was
known from the investigation.  Specifically, OSAGWI did not report
that many of the individuals associated with this case had reported
unusually high levels of health problems since their service during
the Persian Gulf War.  Without this information, a reader could
conclude that there was little basis for concern about exposure to
hazardous substances in this case.  The case report also failed to
mention that health problems affecting many individuals associated
with this incident were among the first Gulf War illnesses-related
incidents reported and the subject of several major DOD
investigations and studies. 

For all six cases, GAO found that OSAGWI had not taken advantage of
DOD and VA clinical databases that contain information on the health
of thousands of Gulf War veterans who may have symptoms of the types
commonly associated with Gulf War illnesses.  Use of these databases
is identified in OSAGWI's methodology for conducting investigations,
and they were used by OSAGWI in some other investigations.  Their use
might have provided leads regarding whether more investigative effort
was needed in cases where exposure to chemical warfare agents or
other environmental hazards might have occurred. 

During its review of the case narratives, GAO noted weaknesses in
OSAGWI's internal quality assurance practices that contributed to
some of the problems it found.  In responding to GAO's findings,
OSAGWI officials said that subsequent to the publication of these
cases, they implemented internal review and quality assurance
procedures that should prevent such shortcomings in future reports. 
This internal review mechanism has been evolving since July 1997.  It
remains to be seen whether these procedures will effectively provide
the quality assurances necessary for OSAGWI to thoroughly investigate
potential chemical, biological, and environmental exposures and to
maintain credibility with veterans. 


   RECOMMENDATIONS
---------------------------------------------------------- Chapter 0:5

To ensure that OSAGWI's case narratives contain all the facts that
have surfaced to date, GAO recommends that the Secretary of Defense
direct the Special Assistant for Gulf War Illnesses to

  -- revise the Marine Minefield Breaching, the Exposure to Mustard
     agent, and the Al Jubayl case narratives to reflect the new
     and/or unreported information identified by GAO and

  -- determine whether OSAGWI's conclusion in the Marine Minefield
     Breaching case that exposure to chemical warfare agent was
     "unlikely" should be changed to "indeterminate" in light of the
     additional information known about this case. 

To enhance the thoroughness of OSAGWI's investigative and reporting
practices, GAO recommends that the Secretary of Defense direct the
Special Assistant for Gulf War Illnesses to

  -- use the DOD and VA Gulf War clinical databases to assist in
     designing the nature and scope of all OSAGWI investigations and

  -- ensure that OSAGWI's internal review procedures provide that (1)
     those reviewing an investigation and related report are
     independent of the team investigating the incident and (2) steps
     are in place that will lead the reviewers to thoroughly check
     that all relevant information obtained by the investigation
     teams has been included in the case narrative reports, all
     conclusions have been fully substantiated by the facts, and all
     logical leads have been pursued. 

More detailed recommendations are found on pages 44 and 45. 


   AGENCY COMMENTS AND GAO'S
   EVALUATION
---------------------------------------------------------- Chapter 0:6

GAO asked DOD and VA to comment on a draft of this report. 

DOD generally concurred with the report.  In response to GAO's
findings and recommendations, DOD agreed to revise OSAGWI's reports
to include new or unreported data identified by GAO's efforts and to
use this information in reassessing case narrative findings.  DOD
also stated that follow-up investigations were either planned or
under way regarding the Marine Minefield Breaching, Reported Mustard
Agent Exposure, and Al Jubayl case narratives.  While DOD agreed to
update the Marine Minefield Breaching narrative, it also noted that
there were still inconsistencies regarding the presence of artillery
fire.  DOD said that as part of its follow-up investigation, it would
objectively consider all information and detail more completely the
artillery issue and its relevance to whatever final assessment is
made. 

DOD and VA both disagreed with GAO's recommendation that OSAGWI
incorporate the use of DOD and VA clinical databases into its
evaluations.  Their disagreement was based on their concern that
these databases might be inappropriately used to establish a causal
relationship between an event and the medical findings of the
registries.  However, DOD agreed that the databases needed to be
examined and analyzed for what they can contribute to understanding
the illnesses of Gulf War veterans. 

GAO continues to believe that the VA and DOD databases could provide
relevant information to investigators about whether individuals that
were at or near a site under investigation are reporting health
problems.  This information could then be combined with other
information to help guide the nature and scope of OSAGWI
investigations.  GAO agrees that information for these databases
cannot be used to establish a causal association as described by DOD
and VA and did not intend that this information should be used for
such purposes. 

DOD agreed that independent reviewers are critical to a thorough and
acceptable report on its investigations.  DOD commented that this was
the reason it established its current multilevel review process. 
This is now being supplemented by the President's Special Oversight
Board, which is examining OSAGWI cases in detail. 

DOD and VA general comments are addressed in more detail in chapter
3.  DOD and VA comments in their entirety and our evaluation of them
are included in appendixes I and II, respectively. 


INTRODUCTION
============================================================ Chapter 1

Many Persian Gulf War veterans have complained of illnesses since the
war's end in 1991.  Over 100,000 of the approximately 700,000 Gulf
War veterans have participated in health examination programs
established by the Department of Defense (DOD) and the Department of
Veterans Affairs (VA).  Many of those examined reported health
complaints, including fatigue, muscle and joint pain,
gastrointestinal problems, headaches, depression, neurologic and
neurocognitive impairments, memory loss, shortness of breath, and
sleep disturbances.  Many veterans claim that their medical symptoms,
some of them debilitating in nature, were not present before their
service in the Persian Gulf War.  Some veterans suspect that their
health problems may be linked to chemical or biological warfare
agents that Iraq may have used during the Gulf War. 

Various organizations have researched the causes of Gulf War
illnesses--the source of much controversy over the past 7 years.  By
the end of 1996, DOD and the VA together had funded 82 research
projects related to Gulf War illnesses.  Despite these efforts, it
remains unclear why some Gulf War veterans became ill following their
service in the Persian Gulf War.  It also remains unclear whether the
rates of reported illnesses for veterans that deployed to the Gulf
are higher overall than the rates for those that did not deploy or
than the rates for the civilian or military population as a whole. 
Also unexplained are differences in the frequency of symptoms
reported by reserve units and active duty units and any correlations
between the location of units and the occurrence of particular
illnesses.  Research designed to answer these and many other Gulf War
illnesses-related questions will not be completed for years.  Of the
151 current federally sponsored research projects, less than 25
percent have been completed, and many are not scheduled for
completion until after 2000. 


   ESTABLISHMENT OF THE PERSIAN
   GULF ILLNESSES INVESTIGATION
   TEAM
---------------------------------------------------------- Chapter 1:1

Prompted by the continuing controversy over Gulf War illnesses,
President Clinton, in 1995, ordered DOD and other federal agencies to
reexamine whether possible exposure to chemical or biological agents
occurred during the Gulf War.  In March 1995, the Deputy Secretary of
Defense established the Persian Gulf Illnesses Investigation Team
within the Office of the Assistant Secretary of Defense for Health
Affairs to explore this question.  The Investigation Team was
established as DOD began to lose credibility among veterans and
veterans' groups in its efforts to determine the causes of Gulf War
illnesses and to support the problems experienced by veterans.  The
12-member team included intelligence officers, an Army Chemical Corps
officer, a pilot, a chemist, a physician, and a criminal
investigator.  Beginning in 1991, senior Defense officials had taken
the position, in testimony before the Congress and in press
interviews, that Iraq did not use chemical or biological weapons
during the Persian Gulf War and that no U.S.  forces were exposed to
chemical or biological agents.  DOD officials maintained this
position as late as 1994.  This position came under attack because
both U.S.  and foreign detection teams had reported that chemical
warfare agents were present on the battlefield.  In 1995 and 1996,
Central Intelligence Agency and U.N.  reports established that during
the Gulf War, Iraq had stored rockets filled with sarin, a deadly
chemical warfare agent, at an ammunition storage site located at
Khamisiyah, Iraq, about 60 miles from Kuwait's border.  In June 1996,
DOD announced that U.S.  troops at Khamisiyah in March 1991 were
likely to have destroyed a bunker of rockets containing chemical
agents.  By July 1997, DOD acknowledged that U.S.  troops near
Khamisiyah may have unknowingly been exposed to low levels of sarin
when they used demolitions to destroy these rockets. 

In the midst of this controversy, DOD became dissatisfied with the
results of the Investigation Team's efforts.  The Investigation Team
did not have the resources needed to accomplish its mission.  For
example, it was unable to follow up on more than 1,200 toll-free
calls received on DOD's hot line with Gulf War veterans.  In
addition, its operation was criticized in the December 1996 report by
the Presidential Advisory Committee on Gulf War Veterans'
Illnesses.\1 The report cited, for example, the Investigation Team's
failure to take advantage of its unique access to classified and
routine military records to fully investigate and help answer the
public's questions about veterans' possible exposure to chemical and
biological warfare agents. 

A DOD team asked by the Deputy Secretary of Defense to evaluate DOD's
responses to Gulf War illnesses concluded that DOD's work in this
area needed a broader focus, a strategy for systematically examining
the various theories concerning the nature and causes of Gulf War
illnesses, and a method of effectively communicating DOD's findings
to U.S.  veterans and the public.  On November 12, 1996, the Deputy
Secretary of Defense established the Office of the Special Assistant
for Gulf War Illnesses (OSAGWI). 


--------------------
\1 Final Report, Presidential Advisory Committee on Gulf War
Veterans' Illnesses, December 31, 1996, pp.  44-45. 


   OSAGWI'S MISSION AND
   IMPLEMENTATION STRATEGY
---------------------------------------------------------- Chapter 1:2

The goal of restoring public confidence in DOD shaped the mission and
organizational focus of OSAGWI.  OSAGWI's mission was broadly defined
as ensuring that (1) veterans of the Gulf War are appropriately cared
for, (2) DOD is doing everything possible to understand and explain
Gulf War illnesses, and (3) DOD puts into place all required military
doctrine and personnel and medical policies and procedures to
minimize any future problems from exposure to chemical and biological
warfare agents and other environmental hazards. 

Although OSAGWI's mission statement charges it with ensuring that
veterans are appropriately cared for, specific responsibility for
providing health care to servicemembers still on active duty and for
conducting the health research program continues to reside with the
Office of the Assistant Secretary of Defense for Health Affairs. 
Similarly, VA remains the primary health care provider for those who
have left military service.  OSAGWI officials told us, however, that
they assist servicemembers and veterans with health care matters
related to Gulf War illnesses by providing them with referrals to
sources of health care or helping them with the registration and
examination processes associated with DOD's Comprehensive Clinical
Evaluation Program or the VA's Persian Gulf Registry.  OSAGWI also
works with the Assistant Secretary of Defense for Reserve Affairs to
(1) help ensure that reservists receive all entitled benefits and (2)
recommend changes to legislation or rules where needed. 

At the time of our review, OSAGWI believed that its core activity
involved investigating and reporting on incidents of possible
exposure to chemical and biological warfare agents and investigating
related military operations during the Gulf War.  After OSAGWI has
completed its investigation of an incident, the investigator writes a
summation document called a case narrative.  The purpose of OSAGWI's
case narratives is essentially to get all of the facts before the
American people about what OSAGWI has learned from its investigation
of an incident.  The case narrative, a document updated as new
evidence becomes known, is to contain all important investigative
facts and OSAGWI's assessment--in terms of "definitely," "likely,"
"indeterminate," "unlikely," or "definitely not"--of the likelihood
that servicemembers were exposed to chemical or biological warfare
agents.  The standard OSAGWI used for its assessments was whether all
available facts would lead a reasonable person to conclude that a
chemical or biological warfare agent was or was not present. 

As of January 1, 1999, OSAGWI had published a total of 19 reports--13
case narratives, 2 environmental exposure reports, and 4 information
papers.  At that time OSAGWI also had 27 active investigations under
way.  Appendix III lists OSAGWI reports and their dates of
publication as well as OSAGWI's active investigations. 


   OBJECTIVE, SCOPE, AND
   METHODOLOGY
---------------------------------------------------------- Chapter 1:3

On July 8, 1997, the Ranking Minority Member of the House Committee
on Veterans Affairs asked us to examine OSAGWI operations. 
Specifically, we were asked to (1) describe DOD's progress in
establishing an organization to address Gulf War illnesses issues and
(2) evaluate the thoroughness of OSAGWI's investigations into and
reporting on veterans' potential exposure to chemical or biological
agents during the Gulf War.  We did not review OSAGWI activities to
coordinate and monitor research on the causes of Gulf War illnesses
because this subject is addressed by other reviews.  To determine
DOD's progress in establishing an organization to address Gulf War
illnesses issues, we obtained briefings from OSAGWI officials
covering the range of activities performed to fulfill their mission
objectives and reviewed associated documentation. 

OSAGWI had issued eight case narratives at the time we began our
review.  It pursued these eight cases first because they involved
incidents that were the most prominent and controversial at the time. 
To evaluate the thoroughness of OSAGWI's investigations and reporting
on veterans' possible exposures to chemical or biological warfare
agents, we reviewed six of these eight case narratives.  The case
narratives we selected for review were (1) "Reported Mustard Agent
Exposure"; (2) "U.S.  Marine Corps Minefield Breaching"; (3) "Fox
Detections in an Ammunition Supply Point (ASP) Orchard"; (4) "Al
Jubayl, Saudi Arabia"; (5) "Al Jaber Air Base"; and (6) "Reported
Detection of Chemical Agent, Camp Monterey, Kuwait." We did not
review the case narrative about the alleged exposure to chemical
warfare agents at Khamisiyah, Iraq, because it was already being
heavily reviewed by other organizations, such as the Presidential
Advisory Committee on Gulf War Veterans' Illnesses and the Senate
Committee on Veterans Affairs' Special Investigation Unit.  We also
did not review the "Possible Chemical Agent on SCUD Missile Sample"
case narrative because it appeared to be less controversial than the
other case narratives. 

In reviewing each case narrative, we generally used as criteria
OSAGWI's methodology, which had itself been derived from the United
Nations and other international community protocols for investigating
chemical warfare incidents.  This methodology included (1)
substantiating the incident by searching for documentation from
operational, intelligence, and environmental logs; (2) documenting
the medical reports related to the incident; (3) interviewing
appropriate people; (4) obtaining information available to external
organizations; and (5) assessing the results.  We also used the
criterion that the case narrative should accurately and fully
disclose all materially significant information relevant to the
investigation of the incident in order to avoid any appearance that
OSAGWI was selectively reporting what had actually happened. 

We initially traced each statement in the published case narrative to
its underlying supporting document to identify the accuracy and
completeness of the text in the narrative.  For those statements
missing adequate supporting documentation, we requested that OSAGWI
provide us with the appropriate documentation.  We also reviewed
additional documentation collected by the OSAGWI investigators in
performing the investigation, even though some of this documentation
might not have been cited in the published narrative.  We looked for
any inconsistencies in information that was not addressed in the
published narrative.  In addition, for the selected case narratives,
we contacted 71 individuals interviewed by OSAGWI that were key
sources of information and requested that they verify the accuracy
and completeness of both the OSAGWI case narrative and the OSAGWI
write-up of the investigator's discussions.  We also contacted some
key participants not originally interviewed by OSAGWI to determine
whether other relevant information was available that might affect
OSAGWI's assessment of possible exposures to chemical warfare agents. 
Finally, we contacted several Gulf War veterans organizations,
including the following:  the American Legion; the Disabled American
Veterans; the Veterans of Foreign Wars; the National Gulf War
Resource Center; GulfWatch; the Desert Storm Justice Foundation; the
Operation Desert Storm/Shield Association; the Gulf War Veterans of
Long Island, New York; and the Chronic Illnesses Net for Persian Gulf
Veterans.  We asked them to provide us with any information they had
that refuted or added to the OSAGWI information.  We did not
systematically approach veterans' groups to obtain their assessments
of overall OSAGWI effectiveness because this was beyond the scope of
our review. 

To further verify the case narratives, we interviewed officials and
obtained pertinent documentary evidence from officials at the
following locations:  OSAGWI, located in Falls Church, Virginia; the
U.S.  Army Chemical and Biological Defense Command at Aberdeen,
Maryland; the U.S.  Army Chemical Center and School at Ft. 
McClellan, Alabama; the Office of the Surgeon General of the Navy,
Washington, D.C.; the Naval Health Research Center, San Diego,
California; the Department of Veterans Affairs, Washington, D.C.; the
Deployment Surveillance Team, which operates the Comprehensive
Clinical Evaluation Program, Falls Church, Virginia; and the U.S. 
Army Gulf War Declassification Project, Falls Church, Virginia. 

We conducted our review from September 1997 to January 1999 in
accordance with generally accepted government auditing standards. 


OSAGWI HAS MADE PROGRESS IN
ADDRESSING ISSUES RELATED TO GULF
WAR ILLNESSES
============================================================ Chapter 2

In the face of severe criticism by veterans, veterans groups, and
others of its handling of Gulf War illnesses issues, DOD committed
additional resources to its efforts to determine the cause of
veterans' health problems.  With greater resources and a much broader
mandate than its predecessor, OSAGWI has made significant progress in
reestablishing communications between DOD and veterans.  In addition,
OSAGWI is actively engaged in identifying improvements DOD needs to
make to protect servicemembers on contaminated battlefields. 


   DOD INCREASES EMPHASIS ON
   DETERMINING CAUSE OF GULF WAR
   VETERANS' HEALTH PROBLEMS
---------------------------------------------------------- Chapter 2:1

DOD is investing significantly more resources for OSAGWI's
investigations and outreach efforts than it did for the Persian Gulf
Illnesses Investigation Team.  In 1996, the Investigation Team
operated with a staff of 12 persons and a budget of $4.1 million.  In
contrast, as of October 9, 1998, OSAGWI had a staff of about 200
persons and a fiscal year 1998 budget of $29.4 million.  In addition,
OSAGWI was given much broader authority than the Investigation Team. 
Finally, OSAGWI reports directly to the Deputy Secretary of Defense;
the Investigation Team reported to the Assistant Secretary of Defense
for Health Affairs. 

OSAGWI officials said that with an adequate budget and sufficient
operating authority within DOD, they were generally unconstrained in
their efforts to pursue OSAGWI's mandate.  According to these
officials, OSAGWI's operations have been fully funded, and OSAGWI has
had largely unrestricted access to personnel, files, and other data
necessary for its work.  For example, OSAGWI has had full access to
classified information from the military services and intelligence
agency sources.  To date, OSAGWI has over 12 million pages of
classified information in its computerized database and approximately
500,000 additional pages of classified data in hard-copy format. 

The Special Assistant (the head of OSAGWI) has been free to staff
OSAGWI according to his needs.  This authority has made it possible
for him to obtain the expertise needed for OSAGWI's investigations. 
From the start, OSAGWI management decided to make extensive use of
contractors to quickly obtain personnel with specific expertise and
maintain the flexibility to change the mix of staffing as needed.  By
October 9, 1998, 173 (87 percent) of OSAGWI's personnel were
contractor employees.  As needed, OSAGWI has obtained specialized
expertise from individuals in various governmental agencies, such as
the Central Intelligence Agency, the Defense Intelligence Agency, and
the Army's Chemical and Biological Defense Command.\1 OSAGWI also has
the authority to contract with private organizations to perform
specialized functions. 


--------------------
\1 The Chemical and Biological Defense Command was later renamed the
Soldier and Biological Chemical Command. 


   OSAGWI HAS IMPROVED
   COMMUNICATIONS WITH VETERANS
---------------------------------------------------------- Chapter 2:2

A key element of OSAGWI's attempt to regain credibility with
veterans, veterans' organizations, and the public was to improve
communications with them.  OSAGWI recognized that major improvements
were needed from earlier DOD efforts to listen to veterans' concerns
and incorporate the information they provided into DOD's
investigations and help provide health referral services to veterans. 
Our review confirmed that OSAGWI has made significant progress in
establishing communications with veterans and others. 

OSAGWI established an E-mail address and encouraged veterans and
others to use both this and the DOD toll-free hotline to communicate
with OSAGWI regarding Gulf War illnesses issues.  Within the first
year of operation, it received almost 1,200 letters and 2,700 E-mail
messages.  OSAGWI staff contacted over 3,900 veterans through
personal telephone calls, which included the vast majority of the
Investigation Team backlog of unanswered calls from 1,200 veterans. 
According to OSAGWI, as of January 1, 1999, it had received 2,850
letters and 4,906 E-mail messages and answered 2,803 and 4,866,
respectively.  OSAGWI used a staff specifically trained to deal with
Gulf War veterans' concerns, obtain information from veterans,
provide information about OSAGWI activities, and make referrals for
those needing medical support from DOD or VA. 

OSAGWI uses a variety of methods to disseminate information on its
operations.  For example, it uses a Web site called GulfLINK on which
it publishes its case narrative reports, information papers, and much
of the supporting documentation used in its investigations.  OSAGWI
reports that this site typically receives over 60,000 inquiries each
week.  OSAGWI also publishes a bimonthly newsletter called GulfNEWS. 
Over 12,000 individuals receive the newsletter.  OSAGWI's leadership
and staff have met with veterans at 18 town hall meetings and made
appearances at 41 national veterans conventions.  In addition, OSAGWI
officials frequently meet with veterans and military service
organizations to discuss Gulf War illnesses topics of interest to
them. 

Finally, OSAGWI communicates directly with veterans that are affected
by its investigations.  After OSAGWI completes an investigation and
publishes the corresponding case narrative, it sends to each affected
veteran a letter that contains a synopsis of the investigation's
results.  For example, following its investigation of the potential
chemical warfare agent exposure in Khamisiyah, Iraq, OSAGWI sent
letters to 97,837 veterans concerning the possibility that they might
have been exposed to low levels of sarin, a chemical warfare agent. 


   OSAGWI HAS IDENTIFIED CHEMICAL
   AND BIOLOGICAL WARFARE FORCE
   PROTECTION ISSUES REQUIRING
   ATTENTION
---------------------------------------------------------- Chapter 2:3

According to OSAGWI officials, OSAGWI must go beyond investigating
and reporting on possible veterans' exposures to chemical or
biological warfare agents and identify ways to better protect
servicemembers from nontraditional battlefield threats.  From its
investigations and reports on possible veterans' exposures to
chemical, biological, or environmental agents, OSAGWI has identified
force protection issues that need improvement.  These lessons learned
generally fall into the following three categories:  how to build
trust and confidence in DOD, how to better account for what happened
on the battlefield, and how to better protect servicemembers on the
battlefield.  Specific examples of the lessons learned include the
need for

  -- institutionalizing a veterans' outreach capability after OSAGWI
     is disestablished;

  -- improving systems for tracking troop movements during a conflict
     so that accurate data is available to show where individuals or
     units were located on the battlefield at any point in time;

  -- improving wartime records development and post-war records
     management systems and addressing issues such as the lack of a
     uniform records management program for joint commands;

  -- improving chemical and biological warfare agent detection
     equipment to make it less prone to false alarms and requiring
     doctrinal changes to collect and retain detector-produced
     printouts of detections;

  -- implementing techniques to better safeguard the health of
     deployed troops, such as deploying forward field laboratories
     early and taking samples to determine whether contamination may
     have occurred subsequent to the use of depleted uranium
     ammunition; and

  -- improving and implementing depleted uranium training programs. 

OSAGWI is presently working with DOD agencies to implement the
lessons learned.  Discussions by the Special Assistant with the
Director of the Joint Staff and the military service Chiefs of Staff
resulted in revised Joint Staff policy concerning record-keeping by
joint commands.  OSAGWI was also instrumental in developing a
DOD-initiated requirement for the military services to review their
depleted uranium training programs.  We did not review what impact
OSAGWI's lessons learned have had toward making changes within DOD. 
Until recently, OSAGWI had no office for monitoring and measuring the
extent to which OSAGWI lessons learned were being acted upon.  In
October 1998, the Special Assistant created a new OSAGWI directorate
to focus attention on ensuring that lessons learned are effectively
communicated to and implemented by the responsible DOD agencies. 


SOME CASE NARRATIVES HAVE
INVESTIGATIVE AND REPORTING
WEAKNESSES
============================================================ Chapter 3

We reviewed six of the eight case narratives OSAGWI had published at
the time we began our review to evaluate the thoroughness and
accuracy of OSAGWI's investigations.  OSAGWI generally followed its
investigation methodology and used appropriate investigative
procedures and techniques.  However, we found significant weaknesses
in the scope and quality of OSAGWI's investigations for three of the
six cases:  the Reported Exposure to Mustard Agent, the Marine
Minefield Breaching, and the Al Jubayl, Saudi Arabia, case
narratives.  Also, OSAGWI did not use DOD or Department of Veterans
Affairs medical databases on Gulf War illnesses in conducting any of
the six investigations.  Despite the weaknesses we noted, in all but
one case--the Marine Minefield Breaching case--we found no basis to
question OSAGWI's determinations of the likelihood that chemical
warfare agents were present. 

Except for failing to take advantage of the VA and DOD medical
databases, we did not find significant weaknesses in the remaining
three cases:  the Camp Monterey, the Al Jaber Airfield, and the ASP
Orchard case narratives.  In investigating these cases, OSAGWI
followed its methodology, identified and interviewed important
witnesses, appropriately used information from other key sources,
included all important information in the case narratives, and
accurately presented the information found.  These investigations
were performed in a generally thorough manner, and the evidence
collected by OSAGWI supported its assessments. 

OSAGWI officials told us that they have revised their internal review
processes for conducting and reporting investigations.  They said
that (1) improvements to these processes have evolved since the
publication of the six case narratives we reviewed, (2) some of the
process revisions were influenced by the findings we reported as our
review progressed, and (3) enhancements to their processes would
considerably minimize the recurrence of similar weaknesses in future
case narratives. 


   OSAGWI'S INVESTIGATIONS AND
   REPORTING PROCEDURES HAVE
   VARIOUS WEAKNESSES
---------------------------------------------------------- Chapter 3:1

Our review of the six selected case narratives disclosed some
weaknesses in the investigations and in the accuracy and completeness
of OSAGWI's reporting.  OSAGWI's investigations were usually
conducted in accordance with the established methodology.  OSAGWI
also generally identified and interviewed the appropriate witnesses,
obtained relevant evidence and information, accurately documented
witness testimonies, and otherwise generally used appropriate
investigative techniques and procedures.  However, we found that
three of the six selected case narratives still contained significant
investigative and reporting problems.  The types of problems varied. 
In three of the six case narratives, we found investigative problems
such as failures to (1) follow up with appropriate individuals to
confirm key evidence, (2) identify or ensure the validity of key
physical evidence, (3) include important information, and (4)
interview key witnesses.  Following is a more detailed description of
the three case narratives containing most of these weaknesses. 


      CASE NARRATIVE ON REPORTED
      EXPOSURE TO MUSTARD AGENT
-------------------------------------------------------- Chapter 3:1.1

This case narrative addresses the reported exposure of an individual
soldier to mustard agent while he was exploring an Iraqi bunker. 
OSAGWI assessed this incident as a "likely" exposure.  OSAGWI's
assessment of this case has been highly controversial.  Some veterans
organizations and others believe that the evidence presented in
OSAGWI's case narrative and the Army's presentation of the Purple
Heart medal to this soldier for his injuries warranted an assessment
of "definite" exposure.  However, we found that this case was
affected by many investigative and evidentiary problems.  Some of
these are more closely associated with shortcomings in DOD procedural
practices during the Gulf War than with how OSAGWI did its
investigation.  Despite the problems identified, we believe that
OSAGWI's original assessment of "likely" exposure remains appropriate
for this case. 


         INCIDENT SYNOPSIS
------------------------------------------------------ Chapter 3:1.1.1

According to OSAGWI's case narrative, the soldier (an Army armored
cavalry scout) was exploring enemy bunkers in southeastern Iraq near
Kuwait's border on March 1, 1991.  He entered one bunker through a
tight passageway and twice brushed against the bunker's doorway and
wall.  About 8 hours later, he began to experience a stinging pain on
the skin of his left upper arm.  Three hours later, blisters had
formed there.  About 15 hours after the exposure, the company medic
checked the soldier's blisters and suspected a heater burn.  Eight
hours later, after more blisters had formed on the soldier's arm, aid
station medical personnel suspected he might be a casualty of blister
agent, treated him, and evacuated him to the company support
battalion.  There, an Army physician photographed the blisters and
confirmed the diagnosis of exposure to a blister agent. 

An Army chemical officer also observed the soldier's blisters and
examined his clothing.  He observed a wet spot on the soldier's
coveralls.  The officer took the coveralls to a Fox vehicle for
testing.\1 From its tests on March 2, 1991, the Fox vehicle
reportedly confirmed the presence of a mustard chemical warfare
agent.  After this positive test, the soldier's coveralls were buried
at the scene in Iraq as contaminated waste. 

On March 3, 1991, a senior medical officer (a physician and an expert
in chemical warfare agents who was also at the time the Commander of
the U.S.  Army Medical Research Institute of Chemical Defense)
examined the soldier's blisters and concluded that they had been
caused by exposure to a liquid mustard agent.  This officer based his
diagnosis largely on (1) the latent period of 8 hours between
exposure and the first symptoms, which is characteristic of mustard
exposure and (2) the absence of any other known chemical compounds
present on the battlefield that have this characteristic. 

On March 4, 1991, following an order from chemical officers at the
division level to confirm the positive results from the first day of
Fox vehicle testing, tests on the soldier's flak vest were performed
by two Fox vehicles--apparently because the vest had not been buried
along with the coveralls.  Initially, both Fox vehicles registered
the potential presence of chemical warfare agents, but only one was
apparently able to confirm the presence of mustard agent.  At the
bunker complex where the soldier was injured, a Fox vehicle also
initially detected a chemical warfare agent but was unable to confirm
the presence of mustard or any other chemical warfare agent. 

The case narrative reported that an in-theater analysis of the
soldier's urine tested positive for thiodiglycol, a breakdown product
of mustard agent.  It also reported that a second urinalysis was
performed by the U.S.  Army Medical Research Institute of Chemical
Defense at Aberdeen Proving Ground, Maryland.  This analysis found no
evidence of thiodiglycol.  Clothing samples were also sent to the
U.S.  Army Chemical Research, Development and Engineering Center for
analysis.  Tests of these items also revealed no evidence of any
chemical warfare agent.  However, the negative test results from one
of the urinalyses were not considered unusual due to the low level of
the exposure. 

OSAGWI based its assessment of "likely" exposure primarily on the
following factors:  (1) the medical assessments of two physicians who
examined the soldier--a senior medical officer and a physician who
had recently been trained to identify chemical warfare agent
injuries; (2) the latent period of 8 hours between the soldier's
exposure and his first symptoms, which is consistent with exposure to
mustard agent; and (3) the positive detections of mustard agent made
in-theater from analyses of the soldier's clothing and urine. 


--------------------
\1 The Fox Nuclear Biological and Chemical Reconnaissance Vehicle was
the most sophisticated and technically complex piece of chemical
detection equipment that the United States used during the Persian
Gulf War.  It was designed to provide an initial alert to warn
personnel of the possible presence of dangerous chemicals and
subsequently to provide detailed confirmation by means of an on-board
mass spectrometer. 


         OUR REVIEW OF OSAGWI'S
         INVESTIGATION
------------------------------------------------------ Chapter 3:1.1.2

We agree with OSAGWI's assessment that exposure to a chemical agent
was "likely." However, we found several investigative procedural
problems with this case, primarily concerning insufficient follow-up
with witnesses, failure to interview key officials about tests
conducted on the soldier's clothing, and uncertainties about the
identity and validity of key physical evidence sent to the United
States for testing. 

First, information we discovered causes us to question the existence
of the soldier's positive in-theater urinalysis for mustard agent. 
OSAGWI based the existence of this test on an Army Central Command
message reporting a positive in-theater test for thiodiglycol. 
However, OSAGWI was unable to find any documented test results from
this urinalysis, and OSAGWI investigators did not perform sufficient
follow-up with the involved individuals to verify that this test had
actually taken place. 

In discussing what OSAGWI knew about the positive in-theater
urinalysis, we learned that OSAGWI had not interviewed either the
senior medical officer or the officer who wrote the message
describing the positive in-theater analysis during its investigation. 
Instead OSAGWI relied upon the senior medical officer's testimony to
the Presidential Advisory Committee, his medical journal article, and
his review of OSAGWI's draft case narrative.  However, this procedure
failed to identify important information.  In early 1998, our
subsequent interviews with the senior medical officer and OSAGWI's
interviews with him revealed that he was unaware of the existence of
any in-theater urinalysis involving the soldier.  He also stated
that, because of his position in the theater as the head of a team of
scientists responsible for assessing any chemical casualties, he
would have known about the existence of any positive urinalysis
performed there.  We then contacted the officer who had written the
Army Central Command message and asked him about his basis for
reporting the positive urinalysis.  He told us that his message was
based on 3rd Armored Division reports that the senior medical officer
had found thiodiglycol in the soldier's urine specimen.  The
available evidence is thus contradictory and insufficient to
establish that this test actually occurred. 

Second, the results of the tests conducted on March 2, 1991 (the
first day of testing), for mustard agent on the soldier's clothing
cannot be confirmed with the available documentation, and OSAGWI did
not interview some key officials involved in the case about the
tests.  According to the Commander of the Fox vehicle involved, the
Fox tests on the soldier's clothing conducted on March 2, 1991,
indicated the presence of blister agent on the soldier's coveralls. 
However, the Fox printout of the test results was apparently lost. 
We located and interviewed the Fox test operator involved, who told
us that several tests were conducted on the soldier's clothing that
day and that there was one positive confirmation for mustard agent. 
During our review, OSAGWI found a printout from one of these tests in
its files, but it was negative for chemical agent.  We noted that
this printout had not been logged into OSAGWI's document receipt
system.  We also noted that OSAGWI had never interviewed the Fox
vehicle Commander in person or the operator who conducted the tests. 
OSAGWI relied upon information provided by E-mail from the Commander
of the Fox vehicles involved because he was then stationed in Germany
and could not easily be interviewed in person.  OSAGWI said it did
not interview the test operator because it could not locate him. 

On the second day of Fox testing, the Fox Commander returned with
both the original and a second Fox vehicle to confirm the positive
test results from the first day of Fox testing of the soldier's
clothing.  One of the Fox vehicles was unable to confirm the presence
of mustard agent on the soldier's flak vest because of a high
concentration of oil products on the vest.\2 The other Fox vehicle,
whose detailed confirmatory procedure was videotaped by a crewmember
but for which the printout is unavailable, did show the presence of
mustard agent.  DOD sent the printout from the original Fox vehicle
and the videotape from the second one to the U.S.  Army Chemical and
Biological Defense Command at Aberdeen Proving Ground, Maryland, for
analysis.  A Command expert found that the surviving printout did not
confirm the presence of chemical warfare agent when the detailed
confirmatory procedure was performed.  However, after examining the
printout and viewing the videotape, this official concluded that the
incident had involved an actual mustard agent detection. 

We found other procedural discrepancies that raise questions
regarding this case.  First, DOD did not adequately identify or
ensure the validity of important physical evidence.  We noticed a
difference between the inventory of items that the Commander of the
Fox vehicles had reportedly packaged for shipment back to the United
States for analysis and the items that were received at the U.S. 
Army Chemical Research, Development and Engineering Center.  The
Commander reported on his inventory list that he did not include
samples from the soldier's coveralls since they were unavailable;
however, the Center's inventory showed receipt of such samples.  When
we interviewed the Commander, he told us that he believed the sample
material was in fact from the Commander's own protective suit that he
wore during the Fox vehicle testing.  These discrepancies raise the
possibility that either someone recovered the soldier's coveralls and
then repackaged the contents for shipment to the United States or
that at least some of the clothing sent back to the United States for
testing was not the soldier's.  The circumstances surrounding the
testing of the soldier's clothing in-theater thus remain unclear.  It
is impossible to determine whether the samples are actually from this
soldier. 

In discussing the investigative weaknesses we found, the OSAGWI lead
investigator told us that this investigation had begun under the
Investigation Team before OSAGWI was established and that the case
was carried over to OSAGWI.  She said that the case's outcome
appeared to be obvious on the surface--particularly since the soldier
had received a medical diagnosis indicating exposure to mustard
agent.  She said that the investigation process at OSAGWI has matured
since this case narrative was published.  She also said that OSAGWI
would do more cross-checking of the facts if this investigation were
being done today. 

Despite the investigation's shortcomings, we believe that OSAGWI's
assessment of "likely" exposure to a chemical warfare agent in this
case is reasonable.  The senior medical officer's clinical diagnosis
that the soldier's injuries were caused by exposure to mustard agent
is significant in that this expert in chemical warfare agents made
his assessment contemporaneously at the time of the injury and
continues to believe that the latent period of 8 hours from exposure
to the first symptoms supports his diagnosis.  In addition, an expert
at the U.S.  Army Chemical and Biological Defense Command, after
reviewing the Fox vehicle printout and viewing a videotape of another
Fox vehicle conducting tests, concluded that this incident involved a
valid detection of mustard agent.  However, we believe the lack of
confirmation of exposure through urinalysis or retained confirmatory
printouts from the Fox vehicles involved prevents OSAGWI's exposure
assessment in this case from being classified as "definitely."


--------------------
\2 This was confirmed by analysis of the available printout from
these Fox vehicle tests. 


      MARINE MINEFIELD BREACHING
      CASE NARRATIVE
-------------------------------------------------------- Chapter 3:1.2

This case narrative addresses reports that U.S.  Marines might have
been exposed to chemical warfare agents while breaching minefield
barriers on the first day of Operation Desert Storm's ground war. 
OSAGWI concluded that the presence of chemical warfare agents was
"unlikely" during this incident, in part because it found that no
mechanism was present for delivering such agents.  However, we found
that OSAGWI overlooked information indicating that a means for
delivering chemical warfare agents might have been present, and that
the case narrative does not include other relevant information
indicating that chemical warfare agents might have been present.  We
believe that these shortcomings are sufficient to cause a reasonable
person to question OSAGWI's assessment. 


         INCIDENT SYNOPSIS
------------------------------------------------------ Chapter 3:1.2.1

On February 24, 1991, the first day of Operation Desert Storm's
ground war, Marine Corps forces breached two rows of minefields that
stretched for miles near the border between Saudi Arabia and Kuwait. 
As they passed through the first row of minefields, two Fox vehicles
(one assigned to units of the 1st Marine Division and another
assigned to the 2nd Marine Division) indicated potential detections
of chemical agents.  The detection by the 1st Division's Fox vehicle
was described as a trace detection of such a small magnitude that no
official report of the detection was made and no Fox printout was
kept to document the detection.  OSAGWI concluded that the presence
of chemical warfare agents in the 1st Division area was "unlikely."

The detection by the 2nd Division's Fox vehicle, however, indicated
the potential presence of mustard, sarin, and lewisite--all chemical
warfare agents.  In this instance, the Fox vehicle printouts were
kept, but because of the hostile environment, the Fox vehicle was not
stopped to perform a more detailed confirmation procedure to
conclusively determine whether chemical warfare agents were present. 

One possible chemical warfare agent injury was reported during the
breaching:  a 2nd Division Marine riding in an amphibious assault
vehicle at the time of the detection claimed his hands were burned,
presumably by a chemical warfare agent, as he closed the vehicle
hatch after hearing the Fox vehicle alert by radio.  However, the
validity of this reported injury was controversial.  Some witnesses
supported the Marine's claim that his hands were blistered, but the
examining physician stated that the Marine had no injury of any kind. 

In investigating the breaching incident, OSAGWI interviewed key
participants in the breaching operations, including members of the
Fox vehicle crews, chemical warfare specialists, some unit
commanders, the Marine who claimed to have been injured, other
Marines from the injured man's unit, and the medical personnel who
examined him.  The investigators also reviewed unit logs and other
pertinent documentation, including classified data, and consulted
with Fox vehicle and chemical weapons technical experts. 

On the basis of reviews of the 2nd Division Fox vehicles' printouts
by three different laboratories, OSAGWI concluded that the Fox
vehicle detections were false alarms, probably caused by the high
concentrations of smoke from oil well fires and petroleum particles
in the atmosphere.  OSAGWI further indicated that except for the
possible injury to one Marine, no other troops reported claimed
chemical warfare agent injuries.  In its overall assessment of the
incident, OSAGWI stated that the presence of chemical warfare agent
was "unlikely." In supporting its assessment, OSAGWI stated that
since no chemical land mines were ever found in Kuwait and since no
artillery fire was encountered by the Marines who breached the first
row of mines, there was no delivery mechanism for chemical warfare
agents. 


         OUR REVIEW OF OSAGWI'S
         INVESTIGATION
------------------------------------------------------ Chapter 3:1.2.2

OSAGWI overlooked a key piece of evidence and did not report other
significant information in its case narrative.  OSAGWI concluded that
the Marines had encountered no Iraqi artillery fire as they moved
through the first row of Iraqi minefields.  This conclusion was based
on comments made by the commanding officer and others of the Marine
company that carried out the minefield breach where the 2nd Division
Fox vehicle reported the presence of a chemical warfare agent. 
However, our review of OSAGWI files disclosed a Marine Corps unit log
entry indicating that Iraqi artillery and mortar fire was present
during the first minefield breach.  The OSAGWI investigator told us
that he had inadvertently overlooked this information during his
investigation.  We also interviewed Marines who told us that Iraqi
artillery and mortar fire was present as they passed through the
first minefield.  Consequently, we believe a delivery mechanism for
chemical warfare agent may have been present. 

Also, the timing of events was significant.  For example, the log
entry indicating that enemy artillery was encountered was made around
6:15 a.m.  on February 24, 1991.  The Fox vehicle detection was made
at
6:22 a.m.  of that same day.  The Marine who claimed to be injured
was riding in an amphibious assault vehicle that was following the
Fox vehicle.  He said his injury occurred just after he heard the Fox
vehicle's report of the chemical warfare agent detection over the
radio. 

We also learned that the Commander of the 2nd Division's Fox vehicle
told OSAGWI investigators that chemical detection paper taped to the
outside of the Fox vehicle was noted to have changed colors after
passing through the first minefield (indicating possible contact with
a chemical agent).  However, this information was not reported in
OSAGWI's narrative.  The OSAGWI investigator said that this
information was omitted because technical experts had told him that
the detection paper could change colors because of the heavy
concentrations of petroleum products in the air coming from the oil
well fires the Iraqis had set.  Furthermore, as mentioned in the case
narrative, three different laboratories had reviewed the Fox vehicle
printout and concluded that the detections were probably false
alarms.  The narrative did not point out, however, that one of the
three laboratories had also said that it could not rule out the
possibility of the presence of a chemical warfare agent. 

Finally, a classified document in OSAGWI's files contained
intelligence evidence not included in the narrative that could
support the possibility of an Iraqi chemical attack.  This
information, some of which has since been declassified, refers to a
report indicating the end of a chemical attack on February 24, 1991,
the same date as this incident.  OSAGWI was aware of this
information, but because of its vagueness, unknown origin,
fragmentary nature, and time of report (about 4 hours after the
breaching event), it was not given much weight during OSAGWI's
analysis.  We agree that the potential impact of this evidence is
unclear.  However, when combined with the other information we have
cited, it provides additional cause for further investigation by
OSAGWI, regardless of its potential for association with this case. 

We believe that OSAGWI's assessment of "unlikely" in this case is
subject to question.  While the information we found does not
conclusively prove that chemical warfare agents were present, it does
increase the potential that some might have been present.  In our
opinion, the weaknesses we found in this case narrative are
sufficient to warrant OSAGWI's reconsideration of its assessment.  We
discussed our findings with OSAGWI investigators and officials, and
they agreed that this information needs to be evaluated.  OSAGWI
officials told us they would include this information in their
follow-up investigation of the minefield breaching incident and would
address the questions we raised. 


      AL JUBAYL, SAUDI ARABIA,
      CASE NARRATIVE
-------------------------------------------------------- Chapter 3:1.3

Regarding this case narrative about three significant events
occurring in the Al Jubayl area during the Persian Gulf War, OSAGWI
concluded that the presence of chemical warfare agents was "unlikely"
for one of the events and "definitely did not occur" in the remaining
two.  We believe that the available evidence generally supports
OSAGWI's assessment, but OSAGWI is still performing work regarding
alternate explanations for some events affecting this case.  However,
we also found that OSAGWI did not include important information in
this case narrative regarding the unusually high levels of post-war
veterans' complaints of medical symptoms they associated with the
incidents involved in this case.  Furthermore, OSAGWI did not
adequately identify and coordinate some of this information that
could potentially provide evidence to help resolve research questions
concerning whether there is a correlation between high levels of
reported Gulf War illnesses symptoms and duty during the Gulf War at
Al Jubayl. 


         INCIDENT SYNOPSIS
------------------------------------------------------ Chapter 3:1.3.1

Al Jubayl is the largest of eight planned industrial cities in Saudi
Arabia.  It consists of an industrial zone and port facilities, as
well as residential and other noncommercial areas.  The Al Jubayl
area was developed during the early 1980s along what was then
essentially undeveloped coast line and was designed to take advantage
of Saudi Arabia's vast oil resources.  Al Jubayl played a crucial
role during the Gulf War--many U.S.  and coalition military units
either passed through or were stationed there. 

OSAGWI's case narrative addresses three separate events that
allegedly involved exposure to chemical agents in the Al Jubayl area: 
the "loud noise" event and alerts on January 19 through 21, 1991; an
Iraqi SCUD missile attack on February 16, 1991; and a noxious fumes
event on March 19, 1991, which some U.S.  military personnel claim
caused them to experience medical problems and turned portions of the
T-shirts they were wearing from brown to purple. 

The need for OSAGWI to investigate these events was underscored by
concerns about Gulf War illnesses expressed in a May 1994 report of
the U.S.  Senate's Banking, Housing, and Urban Affairs Committee
(known as the Riegle Committee) by veterans of Naval Mobile
Construction Battalion 24 (NMCB-24).  NMCB-24 was a reserve "Seabee"
or military construction battalion of 724 enlisted persons and 24
officers.  During Operation Desert Shield/Desert Storm, NMCB-24 was
stationed alongside NMCB-40, an active duty "Seabee" battalion.  Both
units occupied Camp 13, a housing and billeting area located in the
Al Jubayl industrial zone that was commanded by the senior officer of
NMCB-40. 


         THE "LOUD NOISE" EVENT
------------------------------------------------------ Chapter 3:1.3.2

OSAGWI found that the "loud noise" event actually referred to several
loud explosive-like noises and related events occurring between
January 19 and 21, 1991.  As stated in the OSAGWI narrative and
confirmed by our review, early on January 19, a very loud noise like
an explosion was heard throughout the Al Jubayl area.  Units in the
area subsequently reported additional explosions, went on alert, and
conducted tests for the presence of a chemical warfare agent.  A
variety of confusing and contradictory actions subsequently occurred. 
All NMCB-24 tests for chemical warfare agent were officially reported
as negative, but one member of this unit alleged that he had obtained
positive test results for a chemical warfare agent in two of three
attempts.  British units in the vicinity initially reported positive
tests for a chemical warfare agent, but detection teams sent to
investigate these reports were unable to confirm any such agents. 
Some eyewitnesses from NMCB-24 reported a large fireball that
illuminated the sky and medical symptoms such as runny noses, burning
sensations, blisters, and numbness.  They stated that those
experiencing symptoms reported for medical attention within the next
few days.  However, other NMCB-24 personnel said that although they
were unprotected during these events, they experienced no such
symptoms.  After reviewing NMCB-24's medical logs, neither OSAGWI nor
we found any records indicating that medical attention for these
symptoms was sought on or shortly after January 19, 1991.  OSAGWI and
our interviews with the NMCB-24 Commander, medical personnel, and
senior noncommissioned officers similarly revealed no evidence that
any medical attention was sought. 

OSAGWI found, and we confirmed, that many coalition aircraft were
engaged in the air war on the day in question, and Air Force records
show that two coalition aircraft flew over the Al Jubayl area at
supersonic speed during the early hours of January 19, 1991.  OSAGWI
concluded that the loud noise and related events were due to sonic
booms from these aircraft.  It also concluded that the presence of
chemical or biological warfare agents was "unlikely" because (1) DOD
records show that no SCUD missiles were launched toward Saudi Arabia
by Iraq on January 19, (2) no verifiable tests in the Al Jubayl area
were positive for chemical warfare agents, and (3) no records were
found of any individual receiving treatment for symptoms associated
with exposure to chemical or biological warfare agents. 

On January 20-21, 1991, air raid sirens and explosions were heard
again in the Al Jubayl area, but available records reviewed by
OSAGWI, and checked by us, indicated that chemical detection tests
were again negative.  OSAGWI again concluded that the presence of
chemical or biological warfare agents was "unlikely" because (1)
records show a SCUD missile aimed at Dhahran was intercepted and
destroyed at high altitude by a Patriot air defense missile at
approximately the same time as this incident, (2) there is no record
of an impact site in the Al Jubayl area, and (3) no records were
found of anyone receiving medical treatment for symptoms associated
with exposure to chemical or biological warfare agents. 


         THE SCUD MISSILE ATTACK
------------------------------------------------------ Chapter 3:1.3.3

A second possible exposure of veterans to chemical and biological
warfare agents in the Al Jubayl area occurred as the result of an
Iraqi SCUD missile attack early in the morning of February 16, 1991. 
The OSAGWI narrative explains that U.S.  national sensors detected
this missile early in flight and provided warning of the launch.  The
missile landed in the waters of Al Jubayl harbor, and the site of
impact was quickly found and marked by Coast Guard and Navy boat
crews.  Later that day, a Navy explosive ordnance disposal team
surveyed the marked area with an underwater television system and
located missile debris on the harbor's bottom.  Divers confirmed that
the missile had broken apart and that the site contained an intact
SCUD warhead, guidance section, rocket motor, and miscellaneous
components.  Recovery of the smaller SCUD components began on
February 19 and concluded with the warhead on March 2.  During the
recovery operation, tests were conducted, but no evidence was found
indicating the presence of chemical or biological agents.  The Joint
Captured Material Exploitation Center then took custody of the SCUD
components, which were subsequently shipped to the Army Missile
Command in Huntsville, Alabama.  The Command's evaluation of the
recovered SCUD missile components confirmed that the warhead did not
contain chemical or biological warfare agent. 

Some eyewitnesses to this event reported that the SCUD missile was
intercepted and shot down by a Patriot missile and during this
process could have dispersed chemical or biological warfare agents
over Al Jubayl.  A Patriot battery was defending Al Jubayl at the
time.  However, OSAGWI found and we confirmed that this battery was
not operational for maintenance reasons at the time of the attack and
therefore was not able to engage the SCUD.  OSAGWI concluded in its
case narrative that while an Iraqi SCUD missile had hit the waters of
Al Jubayl harbor, it had not detonated, had caused no damage or
injuries, had tested negative for chemical warfare agents, and
therefore was definitely not armed with chemical warfare agents. 


         THE PURPLE T-SHIRT EVENT
------------------------------------------------------ Chapter 3:1.3.4

The third known possibility of exposure to chemical agents at Al
Jubayl occurred on March 19, 1991, when personnel from NMCB-24 were
exposed to unidentified airborne noxious fumes.  These fumes affected
nine persons working in three separate groups.  They experienced
acute symptoms such as burning throats, eyes, and noses and
difficulty in breathing.  In addition, portions of the brown T-shirts
being worn by these individuals turned purple, as did some of the
individuals' combat boots.  Seven persons composing two of the groups
immediately sought medical attention and returned to work with no
further symptoms after showering and changing clothes.  The two
persons in the third group did not seek medical assistance and
continued to work.  The nine persons involved stated that they had
experienced a choking sensation when a noxious cloud enveloped them. 
None saw the origin of the cloud, but all believed it had come from
one of the industrial plants located nearby. 

Evidence collected by OSAGWI regarding the source of the noxious
fumes was inconclusive.  One eyewitness of the event said that he had
seen purple dust falling in the area that was coming from a
smokestack at a nearby fertilizer plant.  The Navy's Environmental
and Preventive Medicine Unit No.  2 (EPMU-2) conducted an
environmental/occupational hazard investigation and site visit to Al
Jubayl in 1994.  The resulting EPMU-2 study did not determine the
source of the irritant.  It noted, however, that the camp was located
in a heavily industrialized area and that emissions from a
petrochemical plant or from a spill within the camp's motor park
could have been the source of the irritant.  The T-shirts and the
boots that changed color were given to unnamed U.S.  military and
Saudi officials.  However, the chain of custody cannot be identified,
and no reports have been found other than an informal telephone call
to NMCB-24 shortly after the incident indicating that "there was
nothing to worry about." The U.S.  Army Material Test Directorate and
the Natick Research Development and Engineering Center later
conducted tests on the type of military T-shirts involved.  The
Natick tests showed that these T-shirts do turn purple when exposed
to acids such as sulfuric (battery) acid or oxides from nitric acid. 

OSAGWI concluded that chemical warfare agents were definitely not
involved in the purple T-shirt event.  OSAGWI reached this conclusion
because (1) the event occurred after the cessation of Gulf War
hostilities, (2) there was no record of hostile attack during the
time period of the event, and (3) the types of medical problems
affecting the individuals involved and their rapid recovery are not
consistent with exposure to chemical warfare agents. 


         OUR REVIEW OF OSAGWI'S
         INVESTIGATION
------------------------------------------------------ Chapter 3:1.3.5

As a result of our review of evidence, procedures, and other
information obtained from OSAGWI and other sources regarding the Al
Jubayl case narrative, we generally concur that OSAGWI's assessments
of whether chemical warfare agents were present are reasonable.  The
evidence generally supports OSAGWI's assessment that chemical warfare
agents were "definitely not" involved in the SCUD missile and purple
T-shirt events.  The loud noise incident involved some contradictions
in evidence or testimony that we could not resolve, but our work
confirmed the credibility of the vast majority of the evidence used
by OSAGWI.  We noted the existence of another potential explanation
of some of the events involved in the loud noise incident.  Some
documents and other evidence we acquired from a veterans'
organization indicate that an Iraqi aircraft or a patrol boat might
have been involved in an attempted chemical attack on Al Jubayl at
the time of this incident.  OSAGWI is currently investigating this
version of events.  However, pending the outcome of this continuing
investigation, we believe that the currently available evidence still
provides a reasonable level of support for OSAGWI's conclusion that
exposure to chemical warfare agents was "unlikely" in this incident. 

Although we concur with OSAGWI's assessments in the Al Jubayl case,
we believe that the case narrative is not complete and could be
misleading because it does not mention the fact that many members of
NMCB-24 have reported unusually high levels of health problems since
their service in the Persian Gulf War.  We also found that OSAGWI had
not coordinated some information developed during this investigation
with the Naval Health Research Center for inclusion in its Gulf War
illnesses research on Seabees. 

OSAGWI's Al Jubayl case narrative states that the methodology it used
was designed to investigate reports of exposure to chemical warfare
agents and to determine whether chemical weapons were used.  OSAGWI
officials told us that in this case they had expanded their
methodology to include a considerable amount of information in the
narrative regarding environmental cleanliness factors affecting the
Al Jubayl area.  They said they had done this in an effort to better
explain the circumstances of the case because some veterans had
expressed concern over the hazardous materials they could have been
exposed to while they were in Al Jubayl.  The narrative thus
contained much information explaining that (1) Saudi environmental
protection standards were equivalent to those of the U.S. 
Environmental Protection Agency, (2) these standards were monitored
and maintained by the Saudis throughout Operation Desert Storm/Desert
Shield, and (3) Saudi monitoring records indicate no detections that
normal standards were exceeded on the date of the purple T-shirt
incident.  The environmental data included in the narrative, much of
which was obtained by EPMU-2, thus indicated that Al Jubayl was no
worse or better than comparable industrialized sites in the United
States. 

We concur that OSAGWI's decision to expand its stated methodology in
order to include this information was appropriate.  As indicated at
the beginning of the narrative, OSAGWI's charge is to investigate all
possible causes of Gulf War illnesses.  However, most of the
information presented in this case narrative leads the reader to
conclude that exposure to either chemical warfare agents or other
chemical agents at Al Jubayl was "unlikely" and probably did not
involve a health threat in the limited incident involving the purple
T-shirts.  The narrative mentions that some NMCB-24 veterans
testified before the Congress (the Riegle Commission) but does not
state why.  The narrative text also contains no information regarding
significant DOD actions taken to address the high incidence of
post-war health problems reported by members of NMCB-24. 

DOD has long been aware of health problems reported by NMCB-24.  In
1992, DOD began to identify clusters of military personnel who were
complaining of medical symptoms they attributed to their Gulf War
service.  As a result, DOD initiated two field investigations.  One
of these, performed at the request of the Navy Surgeon General, was a
study of illnesses reported by members and former members of NMCB-24
conducted during 1993-94 by the same unit (EPMU-2) that conducted the
Al Jubayl environmental study.  EPMU-2 personnel visited 6 of
NMCB-24's 12 detachments during this period, conducted a
questionnaire study, performed medical examinations, reviewed
military and other medical records, interviewed veterans and family
members, and otherwise attempted to identify prevalent symptoms
experienced by the members of NMCB-24 and diagnoses of their
illnesses.  Much of the information they collected was computerized
and used to produce a series of tables and other statistical data
relevant to Gulf War illnesses issues and included in EPMU-2's final
report.  This report contained the following conclusions: 

  -- A significant number of NMCB-24 veterans of the Gulf War have
     experienced an array of nonspecific symptoms since returning
     from the Persian Gulf.  More than 41 percent of the veterans
     from three of the six detachments experienced 10 or more
     symptoms. 

  -- No common syndrome or diagnosis was identified in these
     veterans. 

  -- The diagnoses identified were the same as those that might be
     expected in a group of the same age that had not served in the
     Persian Gulf War. 

  -- More research was needed. 

Our review of OSAGWI's files, our visit to EPMU-2, our interviews of
current and former EPMU-2 officials, and our review of all remaining
EPMU-2 documentation related to this study revealed additional
information.  For example, 44 of the 67 witnesses OSAGWI interviewed
regarding the facts of the loud noise incident are now reporting
health problems they attribute to their service during the Persian
Gulf War.  A former EPMU-2 physician directly involved in the EPMU-2
study told us that while he had no factual baseline for comparison,
it appeared to him that the frequency of symptoms found in NMCB-24
veterans was greater than the frequency to be expected in the general
population.  This observation, along with the high symptom rates, was
one of the reasons the EPMU-2 report recommended more research. 
NMCB-24 veterans have been involved in testimony before the Congress
regarding health problems they attribute to their service in the
Persian Gulf War, and the Naval Health Research Center in San Diego,
California, is currently performing a major, multiyear, Gulf War
illnesses-related epidemiological study involving the vast majority
of the Navy's Seabees.  NMCB-24 veterans have also been the subject
of several additional research studies related to Gulf War illnesses. 

OSAGWI was aware of the existence of the EPMU-2 medical study and had
a copy on file that was originally obtained by its predecessor, the
Persian Gulf Illnesses Investigation Team, in 1996.  However, no
OSAGWI investigators visited EPMU-2 to review files regarding this
study.  No information regarding this study, the Naval Health
Research Center research project, or other epidemiological studies or
research on Gulf War illnesses was included in the case narrative.  A
high-ranking OSAGWI official told us that OSAGWI investigators had
been instructed to consider such medical information as outside their
charter for inclusion in the case narratives.  This official said
that they had been so instructed because this line of inquiry was
more appropriately the responsibility of the Office of the Assistant
Secretary of Defense for Health Affairs and because OSAGWI did not
have the expertise to conduct or evaluate epidemiological studies
such as the one performed by EPMU-2. 

We believe that much more information regarding the health complaints
of NMCB-24 veterans should have been included in the case narrative. 
OSAGWI was aware of this information and could have included it
without conducting or evaluating epidemiological studies.  Including
information developed by EPMU-2 regarding the environmental
cleanliness of Al Jubayl but excluding EPMU-2's report and other
information specifically related to post-war health complaints by
NMCB-24 veterans makes OSAGWI vulnerable to an appearance of bias. 
Such omissions tend to reinforce the beliefs of some that DOD is
inappropriately withholding information. 

We also found that some information developed by OSAGWI might have
significantly added to what is known about Gulf War illnesses issues
involving NMCB-24 had OSAGWI coordinated the information with the
Naval Health Research Center for use in its currently ongoing Seabee
epidemiological study.  For example, as determined by OSAGWI and
reported in the Al Jubayl case narrative, both NMCB-24 and NMCB-40
were located at Camp 13 during Operations Desert Shield and Desert
Storm.  Complaints by NMCB veterans regarding post-war medical
problems they attribute to Persian Gulf service are well known,
having been the subject of several congressional hearings, various
research efforts, and other activities addressing Gulf War illnesses
issues.  An OSAGWI official told us that interviews with selected
NMCB-40 personnel indicated that personnel from this unit were not
experiencing health problems of the same nature and extent as those
reported by NMCB-24 veterans. 

Since NMCB-24 and NMCB-40 occupied the same camp at Al Jubayl, we
believe that a determination of whether NMCB-40 veterans are
encountering medical problems similar to those being reported by
NMCB-24 veterans would be of considerable interest to those concerned
with resolving Gulf War illnesses issues.  The Naval Health Research
Center study is obtaining for analysis a wide range of Gulf War
illnesses-related information from current and former Seabees and
plans to perform a multifaceted analysis of the information
collected. 

In August 1998, Naval Health Research Center officials told us they
had coordinated with OSAGWI officials regarding the Seabee study on
several occasions but that OSAGWI officials had not informed them of
the relationship between NMCB-24 and NMCB-40.  The study's
methodology therefore did not include plans to specifically compare
Gulf War illnesses information obtained from veterans of these two
units.  They acknowledged, however, that such comparisons could be
conducted and that they might provide useful information.  They said
they would be willing to discuss adding such comparisons if OSAGWI
officials requested that they do so.  We believe such comparisons,
especially regarding the extent and nature of post-war medical
symptoms, might provide information important to OSAGWI's
investigation and reporting of Gulf War illnesses issues involving
the Al Jubayl and other case narratives. 

OSAGWI officials agreed that the Al Jubayl case narrative needed to
be modified to acknowledge the high rate of symptoms reported by
members of NMCB-24 and that they would modify the case narrative
accordingly.  They also told us they would coordinate with the Naval
Health Research Center regarding new information that might be
developed through comparisons of NMCB-24 and NMCB-40 data in the
Naval Health Research Center Seabee study. 


   OSAGWI DID NOT USE DOD AND VA
   MEDICAL DATABASES IN CONDUCTING
   ITS INVESTIGATIONS FOR CASES WE
   REVIEWED
---------------------------------------------------------- Chapter 3:2

DOD and the VA maintain databases that contain self-reported health
information and clinical information on thousands of Gulf War
veterans.  Some of these veterans may have symptoms associated with
Gulf War illnesses.  Although OSAGWI's methodology calls for the use
of the DOD and VA databases in its investigations, we found it did
not access them for the six case narratives selected for our review. 
Therefore, OSAGWI missed an opportunity to determine whether
individuals involved in possible exposure incidents were also
reporting symptoms in the databases.  Information thus obtained could
provide leads to help scope and guide the nature of the investigation
and potentially could be combined with other evidence and research
efforts conducted by DOD and others to help evaluate whether chemical
warfare agents might have been present. 


      GULF WAR ILLNESSES DATABASES
      MAINTAINED BY DOD AND VA
-------------------------------------------------------- Chapter 3:2.1

In response to the complaints of many military personnel that
returned from the Gulf War with health problems they believed were
related to their deployment, DOD and VA created programs to track the
health of Gulf War veterans.  Information collected in these programs
is stored in databases that describe the health status of a large
group of Gulf War veterans who have undergone a standardized
examination process to document their health. 


         DOD'S COMPREHENSIVE
         CLINICAL EVALUATION
         PROGRAM
------------------------------------------------------ Chapter 3:2.1.1

The multiphase Comprehensive Clinical Evaluation Program (CCEP) was
implemented by DOD in June 1994 to provide a systematic clinical
evaluation for the diagnosis and treatment of active duty military
personnel who have medical complaints they believe could be related
to their service in the Persian Gulf.  Phase I of the CCEP consists
of a medical history, physical examinations, and laboratory tests
that are comparable to an evaluation conducted during an inpatient
internal medicine hospital admission.  CCEP participants are
evaluated by a primary care physician at their local medical
treatment facility and receive specialty consultations if deemed
appropriate. 

The primary care physician may refer patients to phase II for further
specialty consultations depending on the clinical findings of phase
I. 
Phase II evaluations consist of targeted, symptom-specific
examinations; laboratory tests; and consultations.  During this
phase, potential causes of unexplained illnesses are assessed,
including infectious agents, environmental exposures, psychological
factors, and vaccines.  DOD maintains a database that summarizes the
clinical evaluations of CCEP participants.  The database shows
self-reported complaints and symptoms from everyone and physician
diagnoses for examined participants.  In addition, the database shows
unit assignments, medical complaints, diagnoses, and possible
exposures of individuals who were part of units during the Gulf War
that may have come in contact with chemical warfare agents or other
environmental hazards.  As of October 31, 1998, the CCEP database
contained health information on 34,963 service members who had
received clinical evaluations as a part of the program. 


         VA'S PERSIAN GULF
         REGISTRY
------------------------------------------------------ Chapter 3:2.1.2

The VA's Persian Gulf Registry (VA Registry) was established in 1992. 
Any Gulf War veteran may participate in the registry, even if that
person has no current health complaints.  Like the CCEP, the registry
consists of a two-phase examination process.  During phase I, the
veteran completes a standardized questionnaire on exposures during
the Gulf War and health complaints and undergoes a physical
examination with laboratory testing.  Veterans who have health
problems that remain undiagnosed after phase I are referred to more
extensive phase II medical evaluations. 

VA maintains a database that summarizes the results of clinical
evaluations of registry participants.  It contains information on
symptoms and complaints self-reported by veterans and diagnosed by
physicians.  It also contains information on exposures, birth
defects, and undiagnosed illnesses.  Like the DOD database, the
registry database also contains information on which units the
participants were assigned to during the Gulf War.  As of July
31,1998, the VA Registry contained information on the health
conditions of 70,051 Gulf War veterans who had physical examinations
under the VA program. 


      IDENTIFYING PROGRAM
      PARTICIPANTS COULD HELP
      OSAGWI BETTER FOCUS ITS
      INVESTIGATIVE EFFORTS
-------------------------------------------------------- Chapter 3:2.2

Each of the case narratives selected for our review describes
possible chemical exposure incidents that involve individuals acting
alone or as a part of larger units.  Many of these individuals may
have enrolled in either the CCEP or the VA Registry.  OSAGWI could
use this data to identify whether individuals involved in the
incidents described in the case narratives might be experiencing
health problems. 

Several of the case narratives included in our review describe events
that could have been the subject of further analysis using the CCEP
and VA Registry.  For example, OSAGWI's ASP Orchard case narrative
describes chemical warfare agent alarms at an ammunition storage
facility near an orchard outside Kuwait City, Kuwait.  OSAGWI
collected information from many of the personnel that inspected this
facility and from a variety of other sources, such as the Central
Intelligence Agency and the Defense Intelligence Agency.  OSAGWI
concluded that the alarms were false and that chemical warfare agents
probably had not been stored at this facility. 

However, for the six case narratives we reviewed, OSAGWI
investigators did not query the CCEP or the VA Registry in an attempt
to determine whether any of the several personnel that inspected the
site or any of the hundreds of other personnel encamped nearby had
enrolled and had reported or been diagnosed with health problems. 
Although it would not be definitive, unusually high levels of
participation accompanied by the reporting of certain health problems
and possible exposures might have led OSAGWI to investigate further. 
Performing this investigative step would serve to enhance the
credibility of OSAGWI's case narratives and would confirm OSAGWI's
intention to investigate these events leaving no stone unturned. 

We noted that OSAGWI's investigative methodology includes the use of
the CCEP and the VA registry and that OSAGWI had used such an
analysis in investigating the Khamisiyah incident and in developing
its Depleted Uranium environmental exposure report issued on August
4, 1998.  For example, in performing the investigation on depleted
uranium, OSAGWI investigators queried the CCEP to determine whether
an unusually high proportion of the participants involved in the case
had experienced kidney damage--a possible medical effect of being
exposed to depleted uranium.  According to OSAGWI, the analysis
showed that these CCEP participants did not suffer unusually high
rates of kidney damage compared to the general U.S.  population. 


   THREE CASE NARRATIVES APPEAR TO
   HAVE BEEN APPROPRIATELY
   INVESTIGATED
---------------------------------------------------------- Chapter 3:3

Except for not using the DOD and VA medical databases, the Al Jaber
Air Base, ASP Orchard, and Camp Monterey case narratives generally
did not have the weaknesses we found in the other three cases.  In
investigating these cases, OSAGWI followed its methodology,
identified and interviewed important witnesses, appropriately used
information from other key sources, included all important
information, and accurately presented the information found.  These
investigations were performed in a thorough manner, and the evidence
collected by OSAGWI convincingly supported its assessments. 

The Camp Monterey case is a good example.  In this case, soldiers of
the 8th U.S.  Army Infantry Division were moving wooden Iraqi crates
containing metal canisters out of a building in a bivouac area north
of Kuwait City, Kuwait, so that it could be used to house troops. 
One of the canisters broke open, spilling a white powder-like
substance and causing several soldiers to become ill.  At the request
of the local commander, two Fox vehicles tested the spilled
substance.  Both Fox vehicles initially reported detections of sarin,
a deadly nerve agent, and this apparently led to some initial reports
that soldiers had been exposed to a nerve agent.  Later, mass
spectrometer tests by these Fox vehicles confirmed that the substance
was actually a relatively harmless riot control agent rather than
sarin.  OSAGWI found, and we confirmed, that after interviewing the
personnel present (including the Fox crews) and after reviewing Fox
crew and laboratory analyses of the Fox printouts, the initial alarm
for sarin was an error.  Similarly, in both the Al Jaber and ASP
Orchard cases, initial Fox alarms for persistent chemical warfare
agent could not be confirmed in some instances even by repeated
attempts by the same Fox vehicles.  OSAGWI concluded, and we agreed,
that had the chemical warfare agents been present, they would have
been detected in the repeated tests. 


   OSAGWI HAS MADE CHANGES TO
   IMPROVE ITS INVESTIGATIVE AND
   REPORTING PROCESSES
---------------------------------------------------------- Chapter 3:4

We believe that inadequate quality control procedures within OSAGWI
contributed to the investigative and reporting problems discussed in
this report.  During our review of OSAGWI operations, we periodically
briefed OSAGWI officials on the nature and types of weaknesses we had
found and on our preliminary observations.  OSAGWI officials agreed
that they needed to improve their investigations and their reporting
of the investigation results.  They said that they have instituted
several changes to their internal quality assurance practices that
they believe will considerably strengthen their investigative and
reporting processes. 

According to OSAGWI officials, their current investigative and
reporting process has evolved over the 2 years since OSAGWI was
established.  Consequently, certain enhancements are now in place
that were not present when the six case narratives we reviewed were
published.  More specifically, OSAGWI now requires its investigators
to prepare a written investigation plan.  The investigation plan must
specify the information that will be obtained, the direction the
investigation will take, and the schedule.  The plan is expected to
mirror the overall methodology adopted by the division within the
Investigation and Analysis Directorate for its investigations.  The
division chief is to review the investigation plan and provide
feedback to the investigator on the scope and direction of the
investigation and the proposed schedule.  Following approval of the
plan by the division chief, the investigator can begin the
investigation. 

Also, the process now includes a requirement for a team directional
guidance meeting when the investigation is 50- to 75-percent
complete.  At this meeting, the investigator briefs a small group of
analysts from within the investigator's division on the
investigation's scope, direction, and findings to that point.  The
purpose of the meeting is to identify at an early stage any problems
in the direction of the investigation and to identify any major
information sources that are not being used. 

According to OSAGWI, each case investigation is now periodically
reviewed by the Director of the Investigations and Analysis
Directorate to allow the Director to adjust, as necessary, the scope
of the investigation and the case narrative development. 
Furthermore, the peer review process for case narratives is now more
robust because the peer review team, comprising experienced
individuals, reviews the completed case narrative along with the
source materials.  The peer reviewers are responsible for ensuring
that the text in the case narrative is supported by the source
material and also for identifying portions of the text needing
footnotes to source materials.  In addition, an OSAGWI official said
the internal review of case narratives by key individuals within the
OSAGWI organization is more rigorous than it used to be.  OSAGWI
officials believe that these enhancements to their review processes
will preclude the recurrence of the types of investigative and
reporting weaknesses we found. 


   CONCLUSIONS
---------------------------------------------------------- Chapter 3:5

The weaknesses in the scope and quality of OSAGWI's investigations
and in reporting the results of these investigations in the Reported
Exposure to Mustard Agent, Marine Minefield Breaching, and Al Jubayl
case narratives are significant; however, we agree with OSAGWI's
assessments of the likelihood of the presence of chemical warfare
agents in all but the Marine Minefield Breaching case narrative.  In
our opinion, the lack of effective quality assurance policies and
practices within OSAGWI contributed to the weaknesses we noted.  A
stronger quality control mechanism for its investigations would
provide greater assurance that all relevant facts are included and
that the information presented is accurately and properly sourced. 
More consistent use of some types of medical information would also
strengthen the rigor of OSAGWI's investigations.  By querying
available medical databases for all cases, OSAGWI investigators might
have been able to better determine whether personnel at or near the
sites of incidents had reported or been diagnosed with unusual health
problems, thus helping indicate whether increased investigative
efforts regarding the potential presence of chemical warfare agents
or other environmental hazards in these incidents might be
appropriate. 

OSAGWI's changes to its internal review process appear to be positive
steps in ensuring the quality of investigations and the related case
narrative reports.  Because OSAGWI initiated these changes after the
case narratives we reviewed were published, we could not determine
their effectiveness in ensuring the quality of OSAGWI investigations
and reports.  However, the procedures should incorporate two features
to enhance the credibility of the review process.  First, it is
critical that those named to review OSAGWI's investigations are
independent of the team investigating the incidents to avoid the
appearance of a conflict of interest.  Second, it is important that
the procedures in place lead reviewers to thoroughly check to ensure
that all relevant information obtained by the investigation teams has
been included in the case narrative reports, that all important leads
have been pursued, and that the investigation team has reached
conclusions that are fully substantiated by the facts. 

Information about the potential for differences in the occurrence of
Gulf War illnesses symptoms between NMCB-24 and NMCB-40 developed
during the Al Jubayl case investigation was not shared with the Naval
Health Research Center for consideration for inclusion in its ongoing
Gulf War illnesses research.  We believe this information has
potential for use in helping DOD evaluate issues related to the high
levels of health problems reported by many of the Seabees stationed
at Al Jubayl during the Gulf War. 


   RECOMMENDATIONS
---------------------------------------------------------- Chapter 3:6

To ensure that OSAGWI's case narratives contain all relevant facts,
we recommend that the Secretary of Defense direct the Special
Assistant for Gulf War Illnesses to

  -- revise the Marine Minefield Breaching, Exposure to Mustard
     Agent, and Al Jubayl, Saudi Arabia, case narratives to reflect
     the new and/or unreported information noted in our report and

  -- examine whether it should change its conclusion about the
     likelihood of the presence of chemical warfare agents in the
     Marine Minefield Breaching case from "unlikely" to
     "indeterminate" in light of the additional information now known
     about this case. 

To enhance the thoroughness of OSAGWI's investigative and reporting
practices, we recommend that the Secretary of Defense direct the
Special Assistant for Gulf War Illnesses to

  -- use the DOD and VA Gulf War clinical databases to assist in
     designing the nature and scope of all OSAGWI investigations;

  -- include relevant medical information in its case narratives
     where it is needed to fully explain incidents of possible
     exposure to chemical agents or other potential causes of Gulf
     War illnesses; and

  -- ensure that its internal review procedures provide that (1)
     those reviewing an investigation and related report are
     independent of the team investigating the incident and (2) steps
     are in place that will lead the reviewers to thoroughly check
     that all relevant information obtained by the investigation
     teams has been included in the case narrative reports, all
     conclusions have been fully substantiated by the facts, and that
     all logical leads have been pursued. 

Because of the potential research value of information developed
through OSAGWI investigations, we further recommend that OSAGWI
contact the Naval Health Research Center regarding the usefulness and
desirability of comparing data between the veterans of NMCB-24 and
NMCB-40 for purposes such as helping to determine whether veterans of
these two units are reporting the same types and numbers of symptoms. 


   AGENCY COMMENTS AND OUR
   EVALUATION
---------------------------------------------------------- Chapter 3:7

DOD generally concurred with a draft of this report, agreeing to
revise the case narratives we reviewed to include new or unreported
data, and to reassess case narrative findings based upon any new
evidence.  In particular, DOD agreed to update the Marine Minefield
Breaching case to reflect new information, conduct additional
analysis on the issue of artillery fire during the breaching
operation, and reassess its conclusions as appropriate. 

DOD disagreed with our proposed use of the CCEP and the VA Gulf War
Health Examination Registry in OSAGWI investigations.  In commenting
on this report, DOD stated it was concerned that these databases
might be inappropriately used to establish a causal relationship
between an event and the medical findings of the registries.  DOD
therefore maintains it would be inappropriate for case
investigations, which were designed to report simply on what happened
on the battlefield, to make assumptions about the significance or
validity of the data in these databases without the establishment of
a causal association by scientific research.  DOD also stated
concerns about preempting scientific research in this area and
drawing premature conclusions that would be fallacious.  However, DOD
agreed that these databases need to be examined and analyzed for what
they can contribute to understanding the illnesses of Gulf War
veterans, and noted that the Department has been involved in a number
of research and other analyses of these databases. 

We agree that information from these databases should not be used by
investigators to establish a causal association and/or conclusions as
described by DOD, and did not intend that it should be used for this
purpose.  We also agree that the establishment of Gulf War illnesses
causal relationships is most appropriately a research activity. 
However, we also believe that the VA and DOD databases could
potentially provide relevant information to the investigator about
whether individuals who were at or near a site under investigation
are reporting health problems, and that this information could be
appropriately used, when combined with other information, to help
guide the nature and scope of OSAGWI investigations.  For example,
case investigators could use VA Registry and CCEP data, particularly
where it shows that large numbers of individuals at or near a given
site are reporting health problems, as an indicator for providing
investigative leads and for use in establishing the nature and scope
of an investigation.  This does not mean, as implied in DOD's
comments, that such use of these databases would entail routine
inclusion of the reviewed data in the published case narratives,
their use as a replacement for research activities, or that its use
would result in interpretations of non-scientifically based cause and
effect relationships.  We believe that these databases can be used by
investigators to help guide and scope their efforts without entailing
the types of misuse described by DOD.  We modified the final report
text and recommendatons to clarify our position regarding this
finding. 

DOD agreed that the Al Jubayl case narrative needed to be modified to
place the events of this incident in fuller context, and that this
would include that some servicemembers stationed at Al Jubayl,
especially members of NMCB-24, have reported high levels of health
problems.  DOD also agreed to request that the Naval Health Research
Center undertake an analytical comparison regarding NMCB-24 and
NMCB-40, and that independent reviewers are critical to a thorough
and acceptable report on OSAGWI investigations. 

VA also disagreed with our proposed use of the CCEP and the VA Gulf
War Health Examination Registry in OSAGWI investigations in its
written comments on a draft of this report.  VA's comments were
similar to DOD's regarding this matter.  VA also expressed doubts
regarding the usefulness to research of data comparisons involving
NMCB-24 and NMCB-40. 

Additional discussion of DOD's and VA's comments and our evaluation
is included in appendixes I and II. 




(See figure in printed edition.)Appendix I
COMMENTS FROM THE DEPARTMENT OF
DEFENSE
============================================================ Chapter 3



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)



(See figure in printed edition.)


The following are our comments on the Department of Defense's (DOD)
letter dated February 4, 1999. 


   GAO COMMENTS
---------------------------------------------------------- Chapter 3:8

1.  Our report states that the Office of the Special Assistant for
Gulf War Illnesses (OSAGWI) case investigators did not attempt to use
Comprehensive Clinical Evaluation Programs (CCEP) or Department of
Veterans Affairs (VA) registry information in the six cases we
reviewed.  The report acknowledges OSAGWI's use of the CCEP and VA
registry regarding to the Khamisiyah incident and the Depleted
Uranium Environmental Exposure Report. 

2.  Our report recommends that OSAGWI examine whether to change its
conclusion about the likelihood of exposure to a chemical agent in
light of the additional information now known about this case.  We
agree that additional assessment is needed by OSAGWI to make this
determination, and that some of the evidence regarding this incident
is contradictory and otherwise in need of additional analysis.  Until
additional analysis is performed, it is not clear whether the
likelihood of the presence of a chemical warfare agent in this case
should be assessed as "unlikely" or "indeterminate." However, we
believe the new evidence tends to increase the possibility that an
"indeterminate" assessment might be more appropriate. 

3.  OSAGWI should have identified the potential research value of
information it had in its files regarding the relationship between
Naval Mobile Construction Battalion 24 (NMCB-24) and NMCB-40, and
shared this information with researchers at the Naval Health Research
Center for use in the Seabee study--a major Gulf War illnesses
research project.  We agree that this finding cannot be used by
itself as sufficient evidence to show an overall lack of diligence by
OSAGWI in sharing information from case investigations with
researchers.  However, the fact that neither we nor OSAGWI could find
any evidence that an attempt was made to identify or coordinate this
information in the Al Jubayl case does raise questions about the
adequacy and effectiveness of OSAGWI procedures for identifying and
referring this kind of information.  The word "diligence" was removed
from the final report. 

4.  The sentence referring to how the OSAGWI investigator should have
been alerted to the need for further investigation based on the
absence of thiodiglycol in the urine sample has been deleted from the
final report.  We agree that other inferences could also be drawn
from the absence of thiodiglycol in this analysis. 

5.  Information has been added to the final report regarding the
senior medical officer's testimony, his medical journal article, and
his review of the narrative draft.  However, OSAGWI was remiss in
failing to interview the senior medical officer, especially in view
of the importance of this witness' involvement in the case.  This
officer was still on active duty and stationed in the Washington,
D.C., area at the time of our review.  OSAGWI could have avoided some
of the accuracy problems associated with this case narrative had it
interviewed this officer prior to publication of the narrative. 

6.  We agree that this message was fragmentary, incomplete, and
leaves many unanswered questions about its meaningfulness and
reliability.  However, this message deserves further investigation
because of its date and reference to a chemical attack.  In our
opinion, the fact that the message was received hours after the
incident does not rule out the possibility there could have been a
delay between the time of the event and the time the message was
transmitted.  Even if the message is shown to be unassociated with
the incident in question, its very nature justifies further
investigation by OSAGWI.  OSAGWI officials agreed that they would
attempt to investigate further. 

7.  We agree with OSAGWI regarding the need for caution when
interpreting the relationship between an event, medical findings of
the CCEP and VA registries, and other medical information. 
Accordingly, the final report text was modified regarding the term
"direct linkage with post-war veterans' complaints."

We do not agree that the issue in point necessarily implies such
connotations.  Our concern is simply that while the Al Jubayl
narrative contains much information to the effect that chemical
warfare agents were either "definitely not" or "unlikely" to have
been present at Al Jubayl and that the Al Jubayl area appeared
environmentally clean during the Gulf War, it fails to point out that
(1) many servicemembers stationed there are now reporting unusually
high levels of health problems and (2) DOD has conducted or is
conducting several investigations and major research projects
addressing this issue.  These important facts need to be mentioned in
the case narrative.  If DOD is concerned about the possible misuse of
information regarding reported veterans' illnesses, then the need for
caution regarding its use and research implications could also be
included in the case narrative.  We trust that OSAGWI's planned
modifications to the Al Jubayl case narrative will resolve this
issue. 

8.  In response to this comment, we have changed the report in
several places to refrain from using the word "disclose." However, in
the case narrative involving the Seabees, as well as in one other
case, OSAGWI for various reasons originally chose not to include
information that we believe should have been included. 




(See figure in printed edition.)Appendix II
COMMENTS FROM THE DEPARTMENT OF
VETERANS AFFAIRS
============================================================ Chapter 3



(See figure in printed edition.)


The following are our comments on VA's letter dated February 10,
1999. 


   GAO COMMENTS
---------------------------------------------------------- Chapter 3:9

1.  We are not suggesting that OSAGWI should use data from the DOD
and VA registries to reach conclusions about causal relationships
between participants' health outcomes and the likelihood of their
exposure to chemical warfare agents.  We recognize that these
databases contain the results of medical examinations for voluntary,
self-selected individuals that if used for research purposes could be
affected by participation bias.  However, the databases contain
information about whether the participants believe that they were
exposed to various chemical or environmental hazards, their general
health status, and the results of medical examinations performed by
DOD or VA.  It is also possible that some or many of these
participants may have been at or near a site under investigation by
OSAGWI.  Consequently, the databases may contain potentially relevant
information about individuals that were at a site under investigation
by OSAGWI--information which OSAGWI did not access for the cases we
reviewed.  We are not suggesting that this information would
necessarily change the course of the OSAGWI investigation; however,
review of this information could possibly suggest additional
investigative steps that should be undertaken. 

2.  In our report, we recommend that OSAGWI contact the Naval Health
Research Center regarding the usefulness and desirability of
comparing data about veterans of NMCB-24 and NMCB-40.  Center
researchers told us that such a comparison might be useful.  The
point of our recommendation is that information developed in OSAGWI
investigations that might have research usefulness should be
forwarded to organizations performing the research.  DOD agreed with
this recommendation.  Furthermore, it should be noted that the Seabee
study is one of the research projects being performed under the
management of the Research Working Group of Persian Gulf Veterans
Coordinating Board (Project DOD-1E) and as such is one of the
federally sponsored research projects addressing Gulf War illnesses. 
This project is using scientific methods for collecting data from
both former and current Seabees and plans a multifaceted comparison
of this data.  We made no judgments regarding what the outcome of
this work might be or how it might be reviewed by the epidemiology
research community.  However, we believe that all data or ideas for
comparisons that might have applicability to Gulf War illnesses
research should be forwarded for consideration by the appropriate
research organization.  Otherwise, an opportunity for learning more
about Gulf War illnesses could be missed. 

3.  OSAGWI's own methodology for chemical incident investigations,
which was derived from the United Nations and the international
community, calls for obtaining information from the DOD and VA
registries about the medical condition of personnel involved in an
incident under investigation.  We are not suggesting that OSAGWI
establish a hypothesis from which it could derive undisputed
conclusions.  We are suggesting that the DOD and VA databases may
contain potentially relevant information that could assist OSAGWI in
determining the scope and nature of its investigations. 


OSAGWI REPORTS AND ACTIVE
INVESTIGATIONS
========================================================= Appendix III

Table I.1 lists reports published by OSAGWI.  It is followed by a
listing of active OSAGWI investigations. 



                         Table I.1
          
             OSAGWI Published Case Narratives,
           Information Papers, and Environmental
           Exposure Reports (as of Jan. 1, 1999)

Case name                     Publication date
----------------------------  ----------------------------
Khamisiyah                    April 15, 1997

Camp Monterey                 May 22, 1997

Fox Information Paper         July 29, 1997

Marine Minefield Breaching    July 29, 1997

Al Jubayl, Saudi Arabia       August 13, 1997

SCUD Piece                    August 13, 1997

Exposure to Mustard Agent     August 28, 1997

Al Jaber Air Base             September 25, 1997

ASP/Orchard                   September 25, 1997

M8A1 Information Paper        October 30, 1997

MOPP Information Paper        October 30, 1997

Medical Surveillance During   November 6, 1997
ODS/DS Information Paper

Tallil Air Base               November 13, 1997

Kuwaiti Girls' School         March 19, 1998

An Nasiriyah SW               August 4, 1998

Czech/French Detections       August 4, 1998

Depleted Uranium              August 4, 1998
Environmental Exposure
Report

11th Marines                  November 5, 1998

Oil Well Fire Environmental   November 5, 1998
Exposure Report
----------------------------------------------------------

      ACTIVE OSAGWI INVESTIGATIONS
      (AS OF JAN.  1,1999)
----------------------------------------------------- Appendix III:0.1

Air Campaign Information Paper
Al Muthanna
Biological Warfare
CARC Paint Environmental Exposure Report
Cement Factory
Chemical Munitions Markings Information Paper
Chemical Weapons Sites
Edgewood Tapes
Injured Marine
Khamisiyah - Update
M256 Information Paper
Marine Breaching Followup
Medical Record Keeping Information Paper
Medical Surveillance Information Paper
Muhammadiyat
Pesticides/Insecticides Environmental Exposure Report
Possible Terrorist Attack at Al Jubayl
Possible Post-War Chemical Warfare Use on Iraqis
Rafha M256 Detections
Inhibited Red Fuming Nitric Acid Information Paper
Retrograde Equipment Environmental Exposure Report
Sand Environmental Exposure Report
SCUD Information Paper
Ukhaydir
Vaccine Administration Information Paper
XM21 RSCAAL Detection
JCMEC-TEU Sampling Process Information Paper


MAJOR CONTRIBUTORS TO THIS REPORT
========================================================== Appendix IV


   NATIONAL SECURITY AND
   INTERNATIONAL AFFAIRS DIVISION,
   WASHINGTON, D.C. 
-------------------------------------------------------- Appendix IV:1

Donald L.  Patton
William W.  Cawood
Raymond G.  Bickert
William J.  Rigazio


   NORFOLK FIELD OFFICE
-------------------------------------------------------- Appendix IV:2

Steve J.  Fox
Lynn C.  Johnson
William L.  Mathers

*** End of document. ***