Medical Readiness: DOD Faces Challenges in Implementing Its Anthrax
Vaccine Immunization Program (Letter Report, 10/22/1999,

Pursuant to a congressional request, GAO reviewed the Department of
Defense's (DOD) implementation of the anthrax vaccination program as it
is currently structured, focusing on DOD's: (1) ability to maintain an
adequate supply of anthrax vaccine for its immunization schedule; (2)
system for recording and tracking servicemembers' vaccinations; and (3)
steps to educate servicemembers about the program.

GAO noted that: (1) as of July 1999, DOD had given about 1 million
anthrax vaccinations to more than 315,000 servicemembers, but supply
problems jeopardize its schedule for vaccinating all 2.4 million
servicemembers, and DOD lacks a contingency plan in the event these
problems are not resolved in time; (2) test failures and problems in the
testing itself have slowed or precluded release of 26 of the 40 vaccine
lots since testing began in January 1998; (3) only 14 lots have been
released to DOD since January 1998, and most of these have already been
used; (4) moreover, the manufacturer has yet to receive Food and Drug
Administration (FDA) permission to release lots produced after
restarting operations in May 1999 following a 17-month shutdown for
renovations; (5) DOD has fallen behind its original schedule by 5
months, and it risks further disruption if more vaccine does not become
available by August 2000; (6) DOD has a new recording and tracking
system for vaccinations that is better than the one used during the Gulf
War and in Bosnia, but DOD is not meeting its requirement to record
vaccination data consistently in paper records and in its central
database; (7) GAO's comparison of records from DOD's central database
and files at three military installations showed that 85 to 97 percent
of paper and electronic records agreed on the number of anthrax
vaccinations given to servicemembers, but agreement was lower at two of
those sites for dates and lot numbers; (8) agreement in all categories
was much lower at a fourth installation, with match rates of 8 to 22
percent, in part because individuals' duty stations had not been
updated; (9) this data is vital for: (a) scheduling the FDA-licensed
regimen of 6 vaccinations and boosters; and (b) tracking who receives
vaccinations from a specific lot, should health concerns about a lot
later emerge; (10) delays in updating data on individuals' duty stations
have impeded DOD's ability to use its central database to manage
vaccination schedules and assess unit readiness; (11) DOD has used data
from the Vaccine Adverse Event Reporting system to monitor adverse
reactions to anthrax vaccinations; (12) however, DOD has not
systematically informed its personnel on how to provide needed data into
the system; (13) as a result, DOD may not have data on adverse reactions
that is important for monitoring vaccine safety; and (14) DOD has
employed a high-visibility campaign to educate servicemembers about the
program and has taken steps to address the controversy surrounding the

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

     TITLE:  Medical Readiness: DOD Faces Challenges in Implementing
	     Its Anthrax Vaccine Immunization Program
      DATE:  10/22/1999
   SUBJECT:  Immunization services
	     Safety standards
	     Data integrity
	     Reporting requirements
	     Biological warfare
	     Defense procurement
	     Military personnel
	     Product safety
	     Medical information systems
	     Health hazards
IDENTIFIER:  DOD Anthrax Immunization Program
	     DOD Enrollment Eligibility Reporting System
	     CDC/FDA Vaccine Adverse Events Reporting System
	     DOD Composite Health Care System

** 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:                                   **
**                                                              **
**                    <[email protected]>                        **
**                                                              **
** with the message 'info' in the body.                         **

Report to the Chairman and Ranking Minority Member, Committee on Veterans'
Affairs, U.S. Senate

October 1999

medical readiness

DOD Faces Challenges in Implementing Its Anthrax Vaccine Immunization
<Graphic -- Download the PDF file to


Letter                                                                     3


Appendix I:Scope and Methodology


Appendix II:Packaging and Shipping Protocol


Appendix III:Survey of Servicemember Views of the Anthrax Vaccine
Immunization Program


Appendix IV:Comments From the Department of Defense


Appendix V:GAO Contacts and Staff Acknowledgments


Related GAO Products


Table 1:  Status of 18 Stockpiled Lots Subject to Supplemental
Testing                                         14

Table 2:  Survey on Sources of Anthrax Vaccine Program
Information                                     30

Table 3:  Collection and Review of Electronic and Paper Records39

Figure 1:  Status of Testing for 40 Lots Produced Prior to Shutdown
for Renovations                                 13

Figure 2:  Status of Doses Remaining in Stockpile15

Figure 3:  Comparison of Paper and DEERS Records21

Figure 4:  Respondents' Assessment of Helpfulness of Information31

Figure 5:  Percentage of Participants Wanting More Information44

Figure 6:  Percentage of Respondents Reporting Short-term Adverse Effects,
by Service45

Figure 7:  Percentage of Respondents Reporting Adverse Effects
and Wanting Information, by Age Group           46

Figure 8:  Percentage of Respondents Reporting Adverse Effects
and Wanting Information, by Race                47

CDC     Centers for Disease Control and Prevention

CHCS    Composite Health Care System

DEERS   Defense Enrollment Eligibility Reporting System

DOD     Department of Defense

DMDC    Defense Manpower Data Center

FDA     Food and Drug Administration

GAO     General Accounting Office

USAMMA  U.S. Army Medical Materiel Agency

VAERS   Vaccine Adverse Event Reporting System

Medical Readiness: Safety and Efficacy of the Anthrax Vaccine (GAO/T-NSIAD-
This excludes the first shipment of 20,000 vials (464 vials destroyed of
197,487 shipped as of July 2, 1999). Including that first shipment, the
program's total success rate is still
Other reactions cited by the 111 respondents included redness at the
injection site
(12 respondents, or 11 percent), nausea (4 respondents, or 4 percent),
loss of appetite
(2 respondents, or 2 percent), headaches (6 respondents, or 5 percent),and
                                                      National Security and
                                             International Affairs Division


October 22, 1999

The Honorable Arlen Specter
The Honorable John D. Rockefeller IV
Ranking Minority Member
Committee on Veterans' Affairs
United States Senate

The Department of Defense (DOD) regards the biological agent anthrax, an
infectious disease that is 99-percent lethal if inhaled by unprotected
humans, as the single greatest biological weapon threat to U.S. military
forces. To counter this threat, the Secretary of Defense announced in
December 1997 a plan to immunize all active and reserve military personnel
with a licensed anthrax vaccine. The Secretary stipulated that
immunizations would not begin until DOD (1) established a means of testing
the vaccine over and above tests required by the Food and Drug
Administration (FDA), (2) developed a system for tracking vaccinations,
(3) approved operational and communication plans for the vaccination
program, and (4) had an outside expert review the health and medical
aspects of the program. In May 1998, the Secretary announced that his
conditions had been met, and in August 1998, DOD began immunizations,
giving first priority to personnel deployable to southwest and northeast
Asia, areas where U.S. forces are considered at high risk of exposure to

The vaccination program has been the subject of increasing controversy.
Public debate has centered on whether the vaccine is safe and effective,
and whether it is prudent to rely on only one vaccine manufacturer. Since
the Secretary's announcement, we have reviewed various aspects of the
program. In April 1999, we testified on research on the vaccine's safety
and efficacy, noting the lack of studies on long-term safety and on human
efficacy testing against inhaled anthrax./Footnote1/ In June 1999, we
reported on DOD's financial relationship with the sole-source vaccine

and attributed the manufacturer's serious cash-flow problems to an overly
optimistic business plan./Footnote2/ The following month, we reported that
DOD's data on adverse reactions resulting from vaccinations indicated that
female servicemembers reported such events in greater numbers than male
servicemembers and that no studies had been done to determine the optimum
number of doses of the vaccine./Footnote3/ We also noted that DOD had
conducted some research on a second-generation anthrax vaccine but
considered such research an unfunded requirement and that the Department
of Health and Human Services had recently funded several research grants
to develop a second-generation vaccine.

Although the policy to vaccinate the entire force has been questioned, our
review focussed on the implementation of the vaccination program as
established by DOD. Given the program's scope, DOD's poor medical record-
keeping during the Gulf War, and serious previous shortcomings at the
vaccine manufacturing facility, you asked us to review DOD's
implementation of the vaccination program as it is currently structured.
Specifically, as you requested, we assessed DOD's

o   ability to maintain an adequate supply of anthrax vaccine for its
  immunization schedule,

o   system for recording and tracking servicemembers' vaccinations,

o   efforts to monitor possible adverse reactions to anthrax
  vaccinations, and

o   steps to educate servicemembers about the program. 

To assess the vaccine supply, we reviewed the quantity of vaccine in
stockpile, the status of efforts to test the stockpiled vaccine, and
schedules for producing new vaccine. To assess DOD's tracking of
servicemembers' vaccinations, we compared electronic and paper records of
vaccinations at four locations (one per service). To assess tracking of
adverse reactions, we evaluated DOD's data on adverse reactions and
interviewed medical personnel and vaccine recipients. Finally, to assess
DOD's education efforts, we surveyed vaccine recipients during our four
site visits and discussed education efforts with commanders and program
officials. A detailed discussion of our scope and methodology is in
appendix I.

Results in Brief

As of July 1999, DOD had given about 1 million anthrax vaccinations to
more than 315,000 servicemembers, but supply problems jeopardize its
schedule for vaccinating all 2.4 million servicemembers, and DOD lacks a
contingency plan in the event these problems are not resolved in time.
Test failures/Footnote4/ and problems in the testing itself have slowed or
precluded release of 26 of the 40 vaccine lots since testing began in
January 1998. In all, only 14 lots/Footnote5/ have been released to DOD
since January 1998, and most of these have already been used. Moreover,
the manufacturer has yet to receive FDA permission to release lots
produced after restarting operations in
May 1999 following a 17-month shutdown for renovations. As a result, DOD
has fallen behind its original schedule by 5 months, and it risks further
disruption if more vaccine does not become available by August 2000. DOD's
plans for maintaining an adequate supply of vaccine are optimistic, given
testing problems, and assume that FDA will grant approval of tested lots
in less time than in the past. Consequently, DOD may not be able to
augment its stock of usable vaccine as currently planned. The
manufacturer's financial problems, which had threatened vaccine supply,
have been recently mitigated by a renegotiated contract, but financial
concerns could re-emerge if there are further delays in releasing vaccine.
Although DOD has considered options, should the vaccine manufacturer have
further delays in or lose its ability to produce FDA-approved vaccine, DOD
does not have a formal contingency plan to deal with such possibilities.

DOD has a new recording and tracking system for vaccinations that is
better than the one used during the Gulf War and in Bosnia, but DOD is not
meeting its requirement to record vaccination data consistently in paper
records and in its central database. Our comparison of records from DOD's
central database and files at three military installations showed that 85
to 97 percent of paper and electronic records agreed on the number of
anthrax vaccinations given to servicemembers, but agreement was lower at
two of those sites--ranging from 17 to 69 percent--for dates and lot
numbers. Agreement in all categories was much lower at a fourth
installation, with match rates of 8 to 22 percent, in part because
individuals' duty stations had not been updated. This data is vital for
(1) scheduling the FDA-licensed regimen of six vaccinations and boosters
and (2) tracking who receives vaccinations from a specific lot, should
health concerns about a lot later emerge. Delays in updating data on
individuals' duty stations have impeded DOD's ability to use its central
database to manage vaccination schedules and assess unit readiness.
Commanders need updated duty station information to ensure their personnel
receive vaccinations on time so that they may be ready for deployment. In
addition, DOD does not collect data on those refusing vaccination or
leaving the service to avoid vaccination. This leaves DOD without an
important tool to gauge the extent of resistance to the program and target
training resources to provide servicemembers with the information they want.

DOD has used data from the Vaccine Adverse Event Reporting System to
monitor adverse reactions (or events) to anthrax vaccinations. The system
relies on medical personnel or servicemembers to provide needed data.
However, DOD has not systematically informed these personnel on how to
provide needed data into the system. As a result, DOD may not have data on
adverse reactions (or events) that is important for monitoring vaccine
safety. DOD uses the number of data entries into the system to determine
an adverse reaction rate. However, this data does not provide sufficient
basis for reporting a reaction rate because the information is inadequate
to directly link the health condition of a servicemember to the anthrax
vaccination. Moreover, such events may be underreported. Further,
preliminary data from DOD surveys of vaccine recipients indicates a
greater rate of reaction than is indicated by the reporting system, which
reported 215 adverse events after over 978,000 vaccinations as of July
1999. The reaction rates reported by DOD surveys varied (between 21 and 70
percent), in part due to methodological limitations such as lack of
control groups or adjustments for factors such as physical activity and
age. DOD has reported that there is no evidence of a pattern of serious,
long-lasting adverse reactions.

DOD has employed a high-visibility campaign to educate servicemembers
about the program and has taken steps to address the controversy
surrounding the program. In addition, it recently expanded its
communications efforts by updating the program's Internet site, opening a
toll-free anthrax information line, and forming a speakers' bureau of
anthrax experts. However, a survey we performed at four military
installations, though not projectible beyond the 249 respondents,
indicated that servicemembers want more information about some issues
related to the program. More than two-thirds of survey respondents
reported that the information they received on reasons for the program,
shot requirements and schedules, and consequences of refusals was at least
moderately helpful. However, over half said they either received no
information on possible long-term side effects and procedures for
reporting side effects or found the information less than moderately
helpful. Although many respondents wanted more information on long-term
side effects, data on this topic is limited because no long-term studies
have been carried out. DOD officials recently discussed conducting
additional studies to increase their understanding of possible long-term
health effects. 

This report includes recommendations to the Secretary of Defense to
develop plans in the event that the vaccine does not become available as
currently anticipated, to provide guidance for the consistent reporting of
adverse events, and to establish data collection measures that allow the
program to monitor performance and target training and research resources.


According to the Chairman of the Joint Chiefs of Staff, anthrax is the
greatest biological weapon threat. DOD considers vaccination one of the
measures critical to protecting U.S. forces against such weapons. As a
result, it has begun immunizing all U.S. military personnel--about
2.4 million servicemembers, including all active and reserve--against
anthrax. The Secretary of the Army is the Executive Agent of the program,
which is being implemented in three phases to vaccinate the entire force
by 2004.

o   Phase 1--begun in 1998 and ongoing: 423,000 members assigned or
  rotating to high-threat areas have begun or will begin

o   Phase 2--slated to begin in January 2000: early deploying units--about
  1 million personnel--begin vaccinations.

o   Phase 3--the remaining approximately 1 million personnel begin

The regimen for this vaccine is an initial series of three vaccinations at
2-week intervals, followed by a series of three vaccinations at 6-month
intervals, with annual boosters thereafter.

Production and Testing of Anthrax Vaccine

The anthrax vaccine was licensed in 1970 to protect occupational groups
such as veterinarians, meat packers, wool workers, and health officials
who might come into contact with the disease primarily through the skin.
Its effectiveness against inhalation anthrax in humans has not been
proven, as it would be unethical to conduct such studies on humans.
However, as we reported in our April 1999 testimony,/Footnote7/ studies on
the efficacy of the vaccine in guinea pigs, rabbits, and monkeys support
the view that the vaccine can protect against exposure to inhaled anthrax
in these animals, but the correlation of that protection to humans has not
been established. DOD recently sought to develop an animal model to
establish such a correlation.

DOD currently procures the anthrax vaccine solely from one private
manufacturer, BioPort Corporation. Formerly, the facility was known as the
Biologic Products Division of the Michigan Department of Public Health,
then the Michigan Biologic Products Institute. The manufacturer is the
only FDA-licensed anthrax vaccine manufacturer in the United States.
BioPort produces the vaccine in lots individually numbered for tracking
purposes. Each lot generally consists of about 20,000 vials containing 10
doses each. The lots must be tested according to standard FDA protocols
for purity, potency, sterility, and safety./Footnote8/ Successful results
are then submitted to the FDA for review. If the test results satisfy FDA,
it assigns each approved lot an expiration date and notifies the
manufacturer that the lot can be released for use.

This vaccine has a 3-year shelf life, measured by FDA from the date it
passed the FDA's potency test. The manufacturer can request a 3-year
extension of the shelf life by retesting for potency and submitting
passing results to FDA for approval. FDA also allows retesting of lots
that initially fail potency tests, provided the reason for the failure is
investigated and explained and the retested vaccine meets appropriate
standards. Once a

vial of vaccine is labeled for shipment, its expiration date is changed to
a maximum of 1 year (not to exceed its 3-year shelf life)./Footnote9/

In March 1997, the FDA cited the manufacturer for repeated deviations from
applicable standards. According to DOD, in January 1998 the manufacturer
stopped production as part of a previously scheduled renovation plan to
support the production, testing, and stockpiling of the anthrax vaccine.
These renovations were largely funded by DOD. When the manufacturer
suspended production, it still had 40 lots of anthrax vaccine stored at
its plant. Of these, 31 had already passed all the tests and had received
FDA approval for release./Footnote10/ Nine had not yet been tested. DOD
decided to subject the 31 approved lots to a series of supplemental tests
for purity, potency, sterility, and safety as a prudent
safeguard./Footnote11/ DOD contracted with an independent firm to oversee
the supplemental tests, which were conducted by BioPort. DOD also decided
that the remaining nine lots would not need to undergo supplemental
testing, as these had never been released and would be undergoing FDA-
mandated testing for the first time. 

BioPort resumed production of vaccine in the renovated facility in May
1999. As part of its effort to receive FDA approval of its renovations and
operational changes, BioPort must submit test data to demonstrate that the
lots produced are consistent with each other and with anthrax vaccine
previously produced in the old facility. Once these new lots, called
consistency lots, pass the FDA tests, and once FDA, upon inspecting the
facility and operations and reviewing the test results, approves the
renovations and consistency lots, BioPort will be permitted to resume full
commercial operations--i.e., sell its newly produced vaccine. Without
FDA's approval of its renovations and successful completion of tests on
consistency lots, however, Bioport can produce vaccine but cannot release
it for use.

Packing and Shipping the Vaccine

DOD manages the transport of anthrax vaccine from BioPort to initial
military recipients. To obtain its goal of zero defects and to maintain
vaccine accountability, DOD and BioPort designed a packing and shipping
protocol that maintains the temperature-sensitive vaccine within a
constant temperature range during transport./Footnote12/ Most anthrax
vaccine is shipped via commercial carriers. It is packaged in temperature-
monitored boxes for domestic shipments and in refrigerated containers for
international shipments. Appendix II describes the packing and shipping

Recording, Tracking, and Reporting Immunizations 

As of July 1999, DOD had given about 1 million anthrax vaccinations to
over 315,000 servicemembers. To meet the requirement for a system to track
servicemembers receiving anthrax vaccinations, DOD's Defense Manpower Data
Center added anthrax data fields to an existing DOD-wide database of
personal, service-related, benefits, and residence information. This
database, the Defense Enrollment Eligibility Reporting System (DEERS), now
includes fields to record, among other things, the date and lot number of
each anthrax vaccination given to each servicemember. Also, each service
developed its own interim database to fully document vaccination
information at locations where vaccinations are performed and to
electronically send the information to DEERS, the central repository for
such information./Footnote13/ DOD planned to use an upgrade of its
Composite Health Care System to replace the interim service-specific
tracking systems. Both the service interim systems and DEERS were designed
to be used by unit commanders to ensure that their personnel receive their
vaccinations according to schedule and by the services to report
vaccination rates in their joint monthly readiness reports.

According to the services' implementation guidelines, vaccination
information is to be recorded on two paper forms--the servicemember's
medical record and form PHS-731, commonly known as the yellow shot record.
The medical record is the property of the government, and the yellow shot
record belongs to the individual. Procedures for yellow shot records
varied at the installations we visited. For example, at the Air Force
location, servicemembers were not given their vaccination unless they had
their yellow shot record, while other locations did not have this
requirement. Planning guidance issued by the Joint Staff also required the
Joint Staff Inspector General to review compliance with requirements to
document anthrax vaccinations. The review includes a random sample of
medical records for personnel who received vaccinations between March and
August 1998. The Inspector General's review was assigned in May 1998, and
a report is scheduled to be issued later this year, but preliminary
results were not yet available at the end of our review.

Tracking Adverse Reactions to the Vaccine

DOD submits data on adverse events temporally associated with the anthrax
vaccine to the Vaccine Adverse Event Reporting System (VAERS). VAERS is a
passive surveillance system, meaning that it alerts FDA and the Centers
for Disease Control and Prevention of adverse events that may be
associated with licensed vaccines through information voluntarily reported
by health care providers, patients, or families. VAERS also serves as a
warning signal for detection of previously unreported, unusual adverse
events and/or unexpected increases in reported events. A panel of experts
commissioned by the program reviews all VAERS reports after they have been
submitted to FDA to identify any signaling event that would identify
problems stemming from the anthrax vaccine. As of July 1999, the panel had
found no pattern of causality stemming from the use of the anthrax vaccine.

Supply Problems Jeopardize DOD's Vaccination Schedule

The most critical component of the program, an adequate supply of vaccine,
is threatened by testing delays and possible loss of production
capability. Testing problems have already delayed release of stockpiled
vaccine,/Footnote14/ many lots of which are still unavailable for use.
BioPort has also fallen behind schedule in submitting to FDA test results
on the lots produced after it resumed operations in May 1999. If testing
problems are not resolved soon, or if FDA withholds approval of BioPort's
renovations or newly produced lots, DOD will have difficulty in (1)
providing phase 1 vaccinations beyond August 2000 and (2) beginning phase
2, which has already been delayed 5 months. BioPort also faces financial
problems and some security weaknesses that put the supply of vaccine at
risk. On the positive side, the program has nearly eliminated loss of
vaccine in transit to the field thanks to a highly successful shipping and
packing system. However, despite the risks to the vaccine supply, DOD has
not prepared a formal, written contingency plan for vaccinating
servicemembers should a steady supply be further delayed or disrupted.

Testing Problems Have Delayed Release of Vaccine

As of June 23, 1999, 26 of the 40 stockpiled vaccine lots were still not
available for use (see fig. 1). Most of these--18 lots--had undergone but
not passed all the supplemental tests or had to be retested. An additional
lot needed to pass FDA-mandated tests. Seven other lots passed
supplemental or FDA tests but had not yet received FDA approval. In all,
of the original 40 lots, only 14 had been released for use since the
program began, and 10 of these had been depleted.

Figure****Helvetica:x11****1:    Status of Testing for 40 Lots Produced
                                 Prior to Shutdown for Renovations

<Graphic -- Download the PDF file to

When supplemental testing began in January 1998, program officials
expected to receive the first positive results by April of that year.
However, problems with testing processes, failure of vaccines to pass
tests, and limited testing resources delayed or precluded the release of
18 lots. All 18 lots have passed safety tests but have at least one
unresolved issue with purity, potency, or sterility.

o   Nine lots failed purity tests because the amount of preservative used
  in the vaccine did not meet FDA standards./Footnote15/ DOD is
  considering permanently removing these lots from the stockpile, given
  the time and resources it would take to resolve the issue.

o   Three lots initially failed sterility tests, then passed them, but
  FDA cited serious concerns about the lots. According to program
  officials, the lots will probably not be retested and will likely be
  withdrawn from the stockpile.

o   Fourteen lots still need to pass potency tests. For two of these,
  test results were invalid due to problems in the test procedures,
  causing BioPort to suspend all further potency tests until the problems
  were resolved. At DOD's request, an outside scientific team reviewed
  the test procedures and recommended several corrective
  measures./Footnote16/ BioPort adopted the team's recommendations, which
  took several months to implement. In all, most potency testing was
  delayed 6 to 9 months. The remaining 12 lots have undergone valid
  testing but have not passed it.

Table 1 summarizes the tests needed for the 18 lots that have not yet
passed supplemental testing.

Table****Helvetica:x11****1:    Status of 18 Stockpiled Lots Subject to
                                Supplemental Testing

| Supplemental tests needed               :            Number of lots  |
| Potency                                 :                         6  |
| Potency and sterility                   :                         3  |
| Potency and purity                      :                         5  |
| Purity                                  :                         4  |
| Total                                   :                        18  |

Source: DOD.

Although testing is performed by lots, vaccination schedules are
predicated on the number of doses available. To understand the
implications of these testing problems for DOD's vaccination program,
therefore, it is necessary to assess available doses--especially because
the number of doses in a lot varies. As of June 23, 1999, 5.6 million
doses remained in the stockpile at BioPort, but 4.9 million (88 percent)
of these were unavailable for use (see fig. 2).

Figure****Helvetica:x11****2:    Status of Doses Remaining in Stockpile 

<Graphic -- Download the PDF file to

Note: Does not include almost 2 million doses that have been released and
shipped to installations.

o   More than 3 million doses cannot be released unless BioPort retests
  its lots, achieves successful results, and receives FDA approval to
  release them. According to program officials, lots containing a total
  of over
  2.2 million of these doses are not likely to be ever retested due to
  the aforementioned purity and sterility test results.

o   More than 1.4 million doses unavailable to DOD are awaiting FDA
  approval of successful testing, and program officials expected to
  successfully test and request FDA approval for an additional almost
  206,000 doses needing FDA tests before October 1999. 

In summary, as of June 23, 1999, only 713,000 doses in the stockpile were
available for use, and more than half of them--about 416,000 doses--will
expire in February and April 2000. On the basis of DOD's estimates of
doses required per month, the 713,000 doses would sustain phase 1 of the
program through December 1999. This estimate does not include doses
already delivered to the field and not yet administered. However,
typically, no more than a 3-month supply of vaccine is delivered to a
location, which means that the program could be sustained at best through
March 2000, on the basis of both delivered and available stockpiled
vaccine doses.

Program officials are not concerned about the status of the stockpiled
vaccine. At the time of our review, they expected FDA to grant release of
stockpiled lots containing a total of 1.6 million doses before October
1999, and they projected this would sustain the program through August
2000. This expectation assumes a quick and positive response by FDA.
Program officials also expected to retest and submit some other lots in
early 2000. However, this expectation seems optimistic. According to these
same officials, BioPort's limited testing resources, overburdened by
competing demands, are now being concentrated on obtaining FDA approval of
renovations. Consequently, performing more supplemental tests is a far
lower priority for both BioPort and DOD.

BioPort Renovations Are Behind Schedule and Have Delayed the Program's
Second Phase

A 5-month delay in completing renovations caused BioPort to delay
production startup from January 1999 to May 1999. This delay, coupled with
testing problems and workload, have in turn delayed production and
approval of vaccine consistency lots. Indeed, BioPort has not yet
performed FDA-mandated testing on any of the consistency lots, and as a
consequence, no test results have been submitted to FDA for approval.

In late July, program officials expected BioPort to submit successful
results for the first consistency lots by September 1999 and expected FDA
to approve renovations, which involves an inspection of the facility, and
permit release of these lots by January 2000. This would allow the program
to begin its second phase 5 months after its scheduled August 1999
starting date. Although BioPort officials say they are coordinating more
closely with FDA now, this expectation seems optimistic. FDA is required
to review and provide a response to the manufacturer regarding test
results within 4 to 6 months, but approval is not automatic. Our analysis
of past test approval periods for potency tests of stockpiled
lots,/Footnote17/ showed that the time from successful test completion to
FDA approval has averaged 10 months. This period, which includes any
delays between test completion and the manufacturer's submission to FDA,
ranged from 2 to 29 months and lasted more than 8 months for almost half
of the lots analyzed. Should FDA question the test results or raise other
production issues, release of new production could be delayed beyond
January 2000. Indeed, FDA concurs that this date for approval of
renovations and release of lots is optimistic. 

BioPort's Finances and Physical Security Could Threaten Vaccine Supply

Although somewhat mitigated by recent contract renegotiations, BioPort's
financial problems have reduced the program's vaccine supply in the short
term and may threaten future supplies altogether if production does not
resume. BioPort must improve its financial health if DOD is to retain this
sole source of anthrax vaccine. In June 1999/Footnote18/ we testified
about several problems at BioPort: (1) renovation delays reduced expected
revenues, causing a serious cash-flow problem; (2) the company lacked the
cash reserves and the ability to obtain commercial financing at reasonable
rates to cover operating expenses; (3) its accounting system was
inadequate; and (4) the company projected a significant operating loss for
the year ending December 1999. As a short-term measure to generate
revenues to improve its financial health, BioPort received authorization
from DOD to sell 70,000 doses of anthrax vaccine to other
customers,/Footnote19/ even though it was not fully meeting its
contractual delivery requirements at the time. This action diminished the
potential supply available to U.S. forces. Moreover, on the basis of
renegotiation of its contract with DOD, BioPort (1) will provide DOD with
fewer doses of the vaccine than its original contract stipulated to better
reflect its production capabilities and (2) will be permitted to increase
its private sales to increase revenues. DOD officials stated that this
reduced availability will still meet the program's needs. 

Although not as pressing as its financial problems, the physical security
of BioPort's facility presents some risk to the vaccine supply. In 1998,
the Defense Special Weapons Agency reviewed security at what was then the
Michigan Biologics Products Institute and recommended numerous physical
and operational measures to correct weaknesses. BioPort implemented many
of these, including improvements of doors, locks, and fences, but rejected
other measures it considered "beyond the scope of a biotechnology
business." These included such measures as increasing surveillance and
modifying existing structures. According to BioPort, if DOD considers
further security measures important, it must also consider funding them.
In the opinion of DOD's program officials, most of the remaining security
recommendations are relatively minor in nature and of less concern than
BioPort's production problems. DOD is determining the most effective means
of addressing and funding any high-cost security measures at BioPort. At
the time of our review, however, DOD did not have plans to implement these
measures. Absent a specific implementation plan, it is unclear when or if
these security weaknesses would be addressed.

Well Designed and Administered Packing and Shipping Eliminate Vaccine
Losses in Transit

DOD and BioPort have worked closely together to solve the challenges of
shipping the temperature-sensitive anthrax vaccine to all sorts of
climates in all types of weather. Although a transport problem in the
first shipment of vaccine (to a U.S. base in Germany) led DOD to destroy
20,000 vials rather than risk distributing vaccine that had been subjected
to below-standard temperatures, the program has had extremely few losses
since. Learning from this incident, program officials and the manufacturer
developed a packaging protocol that maintains a safe temperature range
that is continuously monitored from within the container. They also
devised a shipping system that uses commercial carriers and constantly
tracks packages in transit. Shipments are kept small to limit loss from
misplacement or deliberate destruction. According to the program's data,
99.8 percent of all shipped vials arrived safely after the new procedures
were implemented./Footnote20/ Given this excellent record, DOD is adapting
the program's shipping protocol for other environmentally sensitive
pharmaceuticals that it manages.

DOD Lacks Contingency Plans for Disruption or Loss of Production

Program officials acknowledge that BioPort has had testing, production,
financial, and security problems, but they have developed no formal
contingency plans to ensure that vaccinations continue if the supply of
vaccine is disrupted or lost. These officials believe that enough
stockpiled lots have been released to maintain phase 1 through August
2000. However, implementation of phase 2, which depends on new production
and release of vaccine, has already been postponed by 5 months to January
2000, and even this new date may be unrealistic. If the testing and other
problems continue to delay vaccine production and release, DOD will find
it difficult to provide vaccinations in the latter part of 2000 and beyond.

Program officials have considered how to adjust for limited delays in
releases of the current supply, but they have no formal back-up plans in
case of major delays in release of new lots. Several alternatives to the
current phase 1 schedule may be possible, should BioPort be seriously
delayed in obtaining FDA approval of its renovations. These alternatives
range from redistributing vials already sent to the field to suspending
all further vaccinations except for forces in the highest-risk theaters.
However, program officials could not provide formal criteria for
implementing various alternatives, nor could they cite measures of
potential advantages such as how long a specific alternative might extend
the program or how many personnel it might maintain.

The program also has no contingency plan should BioPort lose its
production capability outright, either through FDA rejection of its
renovations, financial failure, or destruction by natural catastrophe or
hostile agent. Program officials did consider construction of new and
completely redundant production facilities, but this alternative was seen
as too costly and time-consuming. As we noted in an earlier report,
development of a second-generation vaccine that may provide other
manufacturing alternatives has begun, but DOD research in the area remains
unfunded./Footnote21/ The Department of Health and Human Services recently
funded several research grants in the area. However, licensing a new
facility or developing a second-generation vaccine would take several
years--too long to offset any major loss of production by BioPort during
the program's timeline. At present, DOD has no means of continuing
immunizations with anything other than what is available from the BioPort
stockpile, most of which still needs to pass tests before it can be used.

Recording and Tracking Vaccinations Has Improved, but Further Improvements

DOD is more capable of recording and tracking vaccinations today than it
was during the Gulf War in 1991 or the Bosnia operations in 1995. However,
DOD is not meeting its requirement to consistently record vaccination data
in its centralized database and paper records. Such inconsistencies could
cause vaccinations to be given off schedule or hinder subsequent
investigations should questions arise about a specific vaccine lot. Also,
delays in updating data on servicemembers' duty stations, as well as
shortcomings in how the services update the DEERS database, have limited
the utility of the database for determining individual vaccination
schedules and assessing unit readiness. While DOD tracks vaccination
exemptions (including waivers and deferrals) for medical reasons such as
pregnancy or administrative leave, it does not monitor refusals or
voluntary departures from the service that may be due to vaccine-related
concerns. As a result, DOD is not able to use the information to monitor
all aspects of the program's implementation.

Vaccinations Recorded, but Some Data Is Incomplete

The Gulf War and the concerns it subsequently generated about Gulf War
illnesses highlighted shortcomings in DOD's systems for recording and
tracking medical data, including vaccination records. In 1997, we reported
that DOD had improved its medical surveillance during operations in Bosnia
but that documentation of vaccinations was one area still needing

In following up on this deficiency, we found that DOD has improved its
ability to record and centrally collect vaccination information. Our
comparison of DEERS data and paper medical records at four military
installations/Footnote23/ (one per service) indicated that, except at the
Marine Corps installation, the numbers of vaccinations were recorded
consistently. However, agreement between the two systems was not as high
when matching specific dates of vaccinations and vaccination lot numbers.
Inconsistency in dates could lead to vaccinations being given off-schedule
and to inaccurate readiness reports. Inconsistent or missing lot
information could hinder investigations, should concerns arise about a
specific lot. Also, information that is not recorded in paper records
makes it difficult to address adverse reactions needing immediate care or
determine the validity of subsequent claims for disability compensation.
Figure 3 summarizes the agreement between electronic and paper information
on vaccinations by service.

Figure****Helvetica:x11****3:    Comparison of Paper and DEERS Records

<Graphic -- Download the PDF file to

Source: GAO.

We made the following observations:

o   The Army base's low match rate for lot numbers was due to the fact
  that lot numbers were not recorded in the medical records for about
  60 percent of vaccinations. Despite this omission, the base did
  record lot numbers in DEERS, and only 1 percent of vaccinations
  recorded in DEERS were without lot numbers.

o   The fact that almost all ship personnel received vaccinations on the
  same days while deployed at sea contributed to the high match rate
  between DEERS and medical records on the Navy vessel.

o   As shown in figure 3, unlike the other installations we visited, the
  Air Force base relied primarily on the yellow shot record, not the
  medical record, for recording vaccinations on paper. Less than 5
  percent of vaccinations, dates, or lot numbers in the medical records
  matched information in DEERS. Officials at the site said the yellow
  shot records were smaller and therefore easier to carry on deployment.
  However, unlike the yellow shot record, the medical record is
  government property and should be complete because it serves as
  evidence for determining veterans' disability compensation. The
  commander of the medical group at the base told us he planned to have
  the information in the electronic records printed and entered in the
  medical records, but this had not been done at the time of our review.

o   Marine Corps officials were unable to provide specific reasons for
  the low match rate with DEERS but noted that (1) neither DEERS nor the
  Navy database are optimized to handle the frequent changes in units of
  the Marine Corps-as a result, DEERS did not list all the Marines
  deployed at Camp Lejeune; (2) lack of training on the Navy database-
  introduced to the Marine Corps in March 1998, the same month that
  anthrax vaccinations began-could have contributed to inconsistencies;
  and (3) the Navy system uses the date the vaccinations are entered into
  the system as the default, causing inaccuracies if vaccinations are not
  entered into the system the same day they are given.

Services' Use of DEERS Limits Its Utility

DEERS was envisioned as a major source of reports on program
implementation. However, concerns about the timeliness and accuracy of
data in DEERS have caused service representatives to rely on interim,
service-specific tracking systems, and other systems to track and report
vaccination information. For example, Army and Navy officials said they
had concerns about DEERS data because duty station information was not
updated, in some cases for as long as 6 to 9 months, in DEERS.

Problems we encountered obtaining medical records for our review also
demonstrated some of the weaknesses in duty station information. For
example, we found that DEERS did not list all servicemembers assigned to a
particular duty station. We obtained personnel rosters for Fort Stewart
and Camp Lejeune from Army and Marine Corps personnel databases. We
compared a sample 300 records from these two lists with the DEERS roster
of servicemembers assigned to the two duty stations and found that the
DEERS database only listed 210 (70 percent) of Fort Stewart personnel and
111 (37 percent) of Camp Lejeune personnel. 

Army and Air Force officials told us they rely on service-specific
tracking systems rather than DEERS to obtain more timely information for
both day-to-day management of vaccinations and quarterly servicewide
readiness reports. Navy and Marine Corps officials told us that because of
shortcomings in the Navy tracking system, they rely on reports from
individual commanders to manage and obtain servicewide data. Officials
from all four services and the program noted that since the start of the
program, service-level systems have improved and are more responsive to
commanders' reporting needs. 

According to Defense Manpower Data Center (DMDC) officials, delays in
updating DEERS are caused partly by service personnel systems not
providing timely data to DEERS. In May 1999, the officials told us they
and the services had taken steps to update duty station information more
promptly. We were unable to test the effectiveness of these changes
because they were instituted after our analysis. DMDC officials also noted
that some data inconsistencies and delays in resolving errors could have
been avoided if the services had followed the original design of the
tracking system, which allows medical providers to be linked directly to
DEERS through their service-level systems. Such direct linkage (1) ensures
that servicemembers' vaccination records are updated regardless of whether
they are vaccinated by their own or another service and (2) minimizes the
impact of mistakes (such as entering the wrong social security number or
recording the same vaccination twice) by providing immediate feedback to
the user in case of error. However, the Army and Navy have adopted systems
that do not directly link to DEERS. Instead, Army, Navy, and Marine Corps
data are transmitted to central servers in their
service-specific systems, which then upload the data to DEERS. This can
cause delays in correcting errors. DMDC officials reported that the Air
Force, thanks to its direct linkage to DEERS, receives far fewer error
messages and has to do fewer follow-ups than the other services. DMDC
produces lists of errors each day but has not analyzed how frequently
different errors occur.

DOD plans to eventually transition the service-specific databases to a
common system. It has begun testing and in 2000 will install the Composite
Health Care System II (CHCS-II), which, among other things, is designed to
interface with DEERS for updating vaccination data. According to DMDC
officials, the system will ensure consistent data quality across services.
However, it is unclear when the services will abandon their interim,
service-specific databases in favor of CHCS-II. Service officials said
they were reluctant to move to the new system because it will rely on
DEERS for vaccination and duty station data and will not be under the
control of the individual services for program upgrades. Moreover, CHCS-II
is not intended for use by deployed units, so it cannot be used on
locations such as Navy ships. DOD has established a team with
representatives from all services that meets regularly to address problems
associated with vaccine tracking systems.

Goal Performance Measures Do Not Include Exemptions and Refusals

DOD set a timeliness goal of vaccinating 90 percent of all servicemembers
no more than 30 days after their vaccinations are due according to the
licensed regimen./Footnote24/ As of July 1999, all services (except the
Army) had met or exceeded that goal. The Army had a 78-percent compliance
rate at that time. The data used to calculate the percentage of "on-time
shots," however, does not include exemptions or refusals.

Servicemembers can receive exemptions from vaccinations for medical
reasons (e.g., pregnancy) or administrative reasons (e.g., extended leave
to change duty stations). Exemptions accounted for about 5 percent or less
of those who received at least one injection, according to service
officials. As for refusals, the program collected anecdotal data on
refusals until January 1999, but the effort was labor-intensive because it
entailed surveying individual commanders. Due to the small number of
refusals-82 after almost 172,000 servicemembers had received one or more
injections-senior Army officials decided the effort was not productive and
halted data collection. Moreover, reports of refusals did not list
personnel who voluntarily left the services due to concerns about the
vaccine. Although the refusal number at the time may have been low, lack
of data limits the program's ability to gauge the effectiveness of its
education efforts and to effectively respond to any increase in opposition
to the vaccine. 

According to written guidance from the Army and Navy and our discussions
with Air Force and Marine Corps officials, servicemembers who refuse
vaccination are initially provided additional education. Servicemembers
who continue to refuse are given a direct order, which, if disobeyed, can
lead to disciplinary action-including discharge-at the commander's
discretion. The Air Force, the only service with a database to track such
information, plans to collect data on disciplinary actions taken against
those who refuse vaccination, but it has not yet begun to do so. A
provision in the National Defense Authorization Act for Fiscal Year 2000
requires an exit survey of all servicemembers leaving military service to
collect data on, among other things, their reasons for
leaving./Footnote25/ This is also a potential source of anthrax refusal

Possible Adverse Events Are Monitored, but DOD's Use of Data May Be

DOD monitors possible reactions (or adverse events)/Footnote26/ to anthrax
vaccinations primarily by using VAERS. However, reports of such events may
be incomplete because servicemembers have not been fully informed about
reporting procedures. Moreover, DOD has used the VAERS data to report a
rate of reaction to the vaccine. This is misleading because of potential
underreporting of events to VAERS, and the potential for overstating the
reaction rate because reports sent to VAERS are not confirmed to be
causally linked to the vaccination. Preliminary data from DOD studies of
adverse events indicates a higher rate of possible reactions than is
reported by VAERS, but the reporting rates in these studies varied and the
studies have methodological limitations. Thus, DOD does not have reliable
information on the extent of adverse reactions. DOD reported that adverse
events have been few in relation to the number of vaccinations and that
there is no evidence of a pattern of serious, long-lasting adverse
reactions. DOD medical personnel have drafted additional clarifying
guidance on treating and reporting adverse reactions to the vaccine.

Medical Staff and Servicemembers Are Not Well Informed AboutReporting
Adverse Events

According to testimony by DOD officials, as of July 1999, 215 adverse
events/Footnote27/ had been reported to VAERS after about 978,000
vaccinations. VAERS is a so-called passive surveillance system, meaning
that it relies on medical personnel or individuals to report adverse
events they think resulted from a vaccination. DOD medical personnel are
required to file a VAERS report for reactions that cause a servicemember
either to lose more than 24 hours of duty time or to need
hospitalization./Footnote28/ DOD reported, and FDA officials commented,
that this requirement exceeds FDA requirements, which only require vaccine
manufacturers, not physicians, to report to VAERS, though reporting by
physicians is encouraged. 

Nonetheless, VAERS data may be incomplete because DOD medical staff and
servicemembers have not received the guidance needed to submit VAERS
reports. Medical officials at a May 1999 conference convened by the
program to discuss clinical issues expressed concern that they had not
received clear guidance on how and when to complete VAERS forms. According
to DOD officials, medical personnel may also report any other reaction
they think might be caused by the vaccine, but because this is not stated
explicitly in DOD's guidance on vaccinations, some medical personnel may
be unsure about which reactions to report.

Servicemembers and their relatives may also report directly to VAERS any
adverse events they suspect are related to a vaccine. DOD, however,
prefers that VAERS reports be filed through its medical providers to
ensure that data is sufficiently detailed to identify and understand
trends. A program official acknowledged that anthrax vaccine educational
materials initially did not explain how to self-report adverse events.
Moreover, of the 249 servicemembers we surveyed,/Footnote29/ 44 percent
(110) told us they had received no information on how to report adverse

In April 1999, DOD updated its briefings to include information on
reporting adverse events. It is also revising regulations to (1) make
reporting requirements more inclusive, (2) clarify patient and provider
roles and responsibilities, and (3) explain how to obtain and process
VAERS forms. In addition, in July 1999, DOD disseminated draft clinical
guidelines for the management of anthrax vaccine adverse events that
outlines clinical protocols, pre-treatments, specialty referral processes,
contraindications, categorization of local and systemic reactions and
associated treatment algorithms, and directions for reporting to VAERS.

DOD Has Used Adverse Event Data Incorrectly 

In presenting reaction rate data, program and DOD officials have shown
reactions reported to VAERS as a percentage of all vaccinations. They did
so in several briefings to GAO and congressional staff, in prepared
testimony, and on the program's Internet site. However, according to FDA
guidance, incidents in the VAERS database reflect a temporal, not
necessarily a causal, relationship with vaccination and thus should not be
used to calculate the incidence of reactions. DOD's use of such a
percentage is an inaccurate representation of the true reaction rate
because (1) not all adverse events prove to be adverse reactions and
(2) studies have shown that reactions are often underreported in passive
surveillance systems such as VAERS, though the extent of possible
underreporting is unknown. As of July 1999, DOD updated its briefing
information to more accurately describe adverse events reported to VAERS
simply as a VAERS report rate.

Other Data on Adverse Events Varies 

In studies where vaccine recipients were surveyed about their reactions to
the vaccine, adverse reactions were reported at a much higher rate than
adverse events reported to VAERS, though these studies have methodological
limitations. A 1962 study of the vaccine indicated that mild local
reactions (swelling of up to 5 centimeters) were reported in
30 percent of recipients and moderate local reactions (swelling of greater
than 5 centimeters) were reported in 4 percent of vaccine
recipients./Footnote30/ DOD has conducted several subsequent studies of
adverse reactions using active

monitoring, and preliminary results vary. /Footnote31/ For example,
according to DOD testimony, 70 percent of respondents in a 1998 survey of
603 medical personnel who had received the vaccine reported a local
reaction to the anthrax vaccine. In another 1997 study, 16 percent (81
respondents) of
508 servicemembers receiving the vaccine reported mild local reactions,
while 5 percent (25 respondents) had moderate to severe local reactions.
As we testified in July 1999, data from other DOD studies also indicated
that women reported a higher rate of adverse reactions than men. These
studies relied on self-reported data, did not use control groups, and were
not adjusted for factors such as occupation, physical activity level, and

According to our survey, when asked if they had had any side effects due
to the anthrax vaccine, 45 percent of recipients (111 respondents)
reported they had,/Footnote32/ and 30 percent (74 respondents) reported
swelling at the injection site, the most frequently cited symptom. Of
those who reported reactions, less than 5 percent (5
respondents)/Footnote33/ said they had missed work or a planned activity
due to the symptoms, and 13 percent (14 respondents) sought medical
treatment. Further, the percentage of female servicemembers who reported
side effects was considerably higher than that of male servicemembers (64
percent of the 36 women surveyed against 42 percent of the 210 men

On August 24, 1999, the program convened a team of civilian and military
experts to design a set of studies to assess the long-term safety of the
anthrax vaccine. Another long-term study is underway to determine whether
individuals who received multiple vaccines, including anthrax vaccine,
during their past employment at Fort Detrick, Maryland, have had any long-
term health effects. A total of 570 study and control volunteers have been
enrolled in this case-control study that began in 1996.

DOD Has an Extensive Education Campaign but Has Not Systematically
Monitored the Results of Its Efforts

DOD and the services have used a variety of measures to educate
servicemembers about the program and have taken steps to address
controversy surrounding the program. However, many respondents to our
survey indicated that they had not received information on some topics
related to the program and desired additional information. The program
recently established a communications division to implement plans to
address the expressed desire for more information. More effective
monitoring of servicemembers' understanding of the program, including the
number of refusals to take the vaccine, would help DOD redirect
educational efforts to those areas where additional information is needed.

Many Servicemembers Have Received Some Information but Want More on
Long-term Side Effects

DOD and the services have made the vaccination program a high priority. At
the four military installations we visited, the commanders established
procedures for administering vaccinations and providing information. In
addition to giving briefings and distributing pamphlets, the commanders
expected health care professionals and staff to play key roles in
providing expert advice to servicemembers. Further, after having briefed
servicemembers about the threat of anthrax, the safety of the vaccine, and
the requirement for the vaccine, commanders often highlighted the
importance and safety of the vaccine by being among the first to receive
it, often in the servicemembers' presence. As shown in table 2, according
to our survey of 249 servicemembers (not projectible beyond those
surveyed), respondents reported that command briefings and medical staff
were their primary sources of information.

Table****Helvetica:x11****2:    Survey on Sources of Anthrax Vaccine
                                Program Information

|                  :  Percentage of respondents reporting  :              |
|                  :    the following as their primary     :              |
|                  :         source of information         :              |
|                  : Command : Medical :   Radio, :  Other :  Percentage  |
|                  :         :   staff : televisi : sources:   reporting  |
|                  : briefin :         :   on. or :        :        they  |
|                  :       g :         :    print :        :    received  |
|                  :         :         :    media :        :          no  |
|                  :         :         :          :        : information  |
|                  :         :         :          :        :      on the  |
|                  :         :         :          :        :       topic  |
| Reasons for the  :      41 :      19 :       11 :     15 :          14  |
| anthrax vaccine  :         :         :          :        :              |
| program          :         :         :          :        :              |
| Requirement for  :      51 :      10 :       14 :     11 :          13  |
| all              :         :         :          :        :              |
| servicemembers   :         :         :          :        :              |
| to get the       :         :         :          :        :              |
| anthrax vaccine  :         :         :          :        :              |
| Vaccination      :      30 :      47 :        2 :      9 :          11  |
| schedule         :         :         :          :        :              |
| Safety of the    :      20 :      29 :        9 :     21 :          21  |
| vaccine and the  :         :         :          :        :              |
| extent it        :         :         :          :        :              |
| offers           :         :         :          :        :              |
| protection       :         :         :          :        :              |
| against anthrax  :         :         :          :        :              |
| Short-term side  :      13 :      38 :        6 :     18 :          25  |
| effects that     :         :         :          :        :              |
| may occur        :         :         :          :        :              |
| Remote           :       9 :      23 :        8 :     16 :          44  |
| possibility of   :         :         :          :        :              |
| long-term side   :         :         :          :        :              |
| effects          :         :         :          :        :              |
| Procedures for   :      16 :      35 :        1 :      4 :          44  |
| reporting side   :         :         :          :        :              |
| effects          :         :         :          :        :              |
| Consequences of  :      54 :       3 :       14 :     12 :          16  |
| refusing the     :         :         :          :        :              |
| vaccine          :         :         :          :        :              |

Our survey also showed that for many topics, servicemembers found
information they received at least moderately helpful, but information
related to long-term side effects and procedures for reporting side
effects was not as helpful to many respondents. Figure 4 shows how helpful
respondents found information they received about each topic.

Figure****Helvetica:x11****4:    Respondents' Assessment of Helpfulness of

<Graphic -- Download the PDF file to

Source: GAO.

According to our survey, at least 57 percent of respondents reported that
the information they received on the reasons for the program, the
requirement for the vaccine, the consequences of refusing the vaccine, the
vaccination schedule, the protection the vaccine offers against anthrax,
and the short-term effects the vaccine may have was moderately or very
helpful. There were some areas, however, where many servicemembers either
received no information or desired additional information. Our survey
showed that only 35 and 47 percent of respondents, respectively, said the
information they received on the possibility of long-term adverse effects
and on reporting adverse reactions was at least moderately helpful, and 44
percent said they had not received information on the remote possibility
of long-term side effects. Further, when asked what additional information
they wanted, 43 percent (106 respondents) reported a desire for
information on long-term side effects. 

Many of the respondents who said they wanted information on possible long-
term adverse reactions also reported experiencing some side effects. Fifty-
nine of the 111 respondents (53 percent) who reported experiencing short-
term reactions said they wanted information on the possibility of long-
term adverse effects. Air Force servicemembers represented almost 70
percent of this group.

The wish for information on possible long-term adverse reactions was also
highlighted in May 1999, when a commander temporarily halted anthrax
vaccinations at Dover Air Force Base, Delaware, until he determined that
servicemembers' questions on the vaccine's safety and its possible health
risks had been satisfactorily addressed. The questions were spurred by a
magazine article about an unauthorized additive, squalene, alleged to have
been used in some vaccine lots and about the alleged relationship between
the vaccine and Gulf War illnesses. Following an initial meeting at which
servicemembers raised these questions but were unsatisfied with the
responses, several DOD, Air Force, and Army personnel knowledgeable of the
program, including the Air Force Surgeon General, provided responses in a
second set of meetings. These experts reported that independent laboratory
tests performed on the specific lots cited by the media had failed to find
squalene. Subsequently, Dover officials resumed anthrax vaccinations.
Further analysis of all of 13 additional lots also found no evidence of

Concerns similar to those expressed at Dover have been reportedly voiced
at other installations. A primary reason for dissatisfaction with
information about long-term side effects appears to be that research has
not been done to address the topic. According to program officials, such
studies have recently been discussed but are not yet funded or underway.

Program Recently Established a Communications Division

The program has recently established a communications division to focus on
servicemembers' information needs. The division updated the program's
Internet site and established a toll-free information line and a traveling
speakers' bureau of experts on anthrax and the vaccine. The communications
division was also instrumental in updating briefings for installation
leaders and medical personnel to provide more detailed information on the
threat of anthrax. DOD expects these briefings to respond effectively to
commanders' and medical staff's needs by countering misinformation in the
media and on the Internet.

The communications division plans to periodically obtain feedback on
implementation of its plan, which includes surveys carried out by DOD and
service program staff while on site visits to convey key messages and
ensure consistency of information. Program staff, including some from the
communications division, conducted the first survey in July 1999 and plan
to conduct surveys at seven other sites to be visited by December 31,
1999. The surveys will not be projectible but are expected to provide
useful information on the implementation of the communications plan. In
July 1999, the program submitted a budget proposal for program evaluation
and research to include an annual evaluation of communications
effectiveness and clinical issues. The proposal did not include linking
vaccine refusals to program effectiveness. 


DOD's policy decision to vaccinate the entire force against anthrax has
presented many challenges. DOD has made progress in implementing the
anthrax vaccination program, but several challenges remain. As of July
1999, DOD had administered more than 1 million vaccinations to over
315,000 servicemembers. DOD has taken steps to ensure that vaccine lots
are recently tested for purity, potency, sterility, and safety before they
are released for use. Vaccinations are recorded in a central database (an
improvement over past record keeping); data on the program's
implementation progress is collected; reported adverse events are
monitored; servicemembers receive information on the program; and the
manufacturer's contract has been restructured to help improve its
financial condition. 

The first challenge, however, is to develop a formal plan for vaccinating
servicemembers should the anthrax vaccine supply not be available as
currently anticipated. If BioPort, the sole-source supplier of the
vaccine, is unable to obtain FDA approval to release stockpiled or newly
produced vaccine, DOD will not be able to vaccinate the entire force as
planned. Developing a formal plan would help DOD consider (1) various
contingencies, including options for altering the three phases of the
program, should the vaccine supply become limited and (2) strategies to
mitigate the risk of loss of the sole-source manufacturer, including
strategies to acquire a second production source or develop a
second-generation vaccine.

Second, while DOD has improved its recording and tracking of vaccinations,
shortcomings remain in documenting vaccinations in paper medical records
and in establishing a DOD-wide database useful to commanders for tracking
vaccinations. To ensure that servicemembers obtain the health care they
need, especially if they experience short- or long-term adverse events
associated with vaccinations, DOD must keep paper and electronic medical
records accurate and current. Also, because the anthrax immunization
regimen requires several vaccinations over a short period and annual
boosters, it is critical that commanders have timely information about
servicemembers in their units who are scheduled for vaccinations. Because
the DOD-wide database, the Defense Enrollment Eligibility Reporting
System, lacks current data on servicemembers' duty stations, commanders do
not find it useful for scheduling individual vaccinations or determining
the status of vaccinations for their unit as a whole. DOD's plan to
incorporate vaccine tracking in an upgrade to its Composite Health Care
System program will be of limited use to commanders if it does not give
them some of the capabilities of the service-level systems.

Third, measures used to track program implementation omit important data
needed to assess overall performance such as refusals. Program officials,
however, have discontinued monitoring refusals, even though such data
would help monitor possible lack of acceptance of the program. Moreover,
previous reports of refusals did not include personnel leaving the
services because of concern about the anthrax vaccine. If collected during
exit interviews scheduled in 2000, this data could provide another
indicator of possible resistance to the program. 

Fourth, data on adverse events may be underreported, making it difficult
to continuously monitor vaccine safety. DOD has updated educational
material on reporting adverse events, and monitoring the effectiveness of
efforts to distribute this information to servicemembers would help ensure
adverse events are consistently reported. 

Fifth, servicemembers clearly want more information on the possibility of
long-term side effects. Because the vaccination program is a mandatory,
servicewide program, it is essential that servicemembers be given the
fullest information possible on these side effects. Although DOD officials
have recently discussed potential studies on possible long-term side
effects of the vaccine, none have been designed or funded.

Finally, program officials have not systematically monitored their
education efforts. Informing servicemembers about the risks of anthrax,
the protection the vaccine affords, and the vaccine's safety and efficacy
is critical to the long-term success of the program. While the program has
provided information on some of these topics and has established a
communications division dedicated to improving communications with and
education of servicemembers, monitoring the effectiveness of such efforts
is important for allocating education resources. Officials plan to obtain
feedback on their new efforts but have not yet designed and implemented a
systematic strategy to help assess overall progress in meeting
communications goals. Further, because data on refusals to receive the
vaccine is no longer being collected, it is difficult to better target
educational efforts and address emerging concerns. 

These problems need to be resolved if the program is to succeed in
vaccinating the entire force against anthrax. 


To address the challenges DOD faces in vaccinating its total force against
anthrax, we recommend that the Secretary of Defense direct the Secretary
of the Army, as Executive Agent for the anthrax vaccination program, to

o   prepare a formal, written plan that addresses strategies to deal with
  (1) contingencies for vaccinating servicemembers if the supply of
  anthrax vaccine is not augmented with new production and (2) the
  risks associated with reliance on a single vaccine manufacturer; 

o   routinely collect and report, among other program performance
  measures, data on the number of servicemembers refusing to take the

o   improve DOD guidance and training on how to report adverse events to
  the Vaccine Adverse Event Reporting System and refrain from
  inappropriately using data from the system to report an adverse
  reaction rate; 

o   design and conduct a study on possible long-term side effects of the
  anthrax vaccine and develop a communications plan to provide
  servicemembers information on the status of this effort; and 

o   continue improvements in educational efforts by regularly surveying
  vaccine recipients and addressing their educational needs. 

In addition, we recommend that the Secretary of Defense direct the Defense
Manpower Data Center to 

o   assess the timeliness of personnel duty station data in the Defense
  Enrollment Eligibility Reporting System to determine where time lags
  occur in obtaining data and take or recommend steps to resolve untimely

o   review service requirements for recording and tracking medical data
  and incorporate plans to address these requirements in future upgrades
  of the Composite Health Care System, and

o   include the response "to avoid the mandatory anthrax vaccine" (or
  words to that effect) among answers to questions on the reasons for
  resigning from the military in the DOD-wide exit survey to be
  administered in 2000.

Agency Comments

In written comments on a draft of this report, DOD generally concurred
with the report findings and recommendations, emphasized several areas of
concern, and described recent or proposed actions to implement
recommendations made in our report. DOD also provided technical comments
which we incorporated as appropriate.

DOD commented that we did not fully discuss some key aspects and successes
of the anthrax immunization program. For example, DOD stated that it keeps
three paper records to ensure that immunizations are documented and that
no other organization in the world can match this accomplishment. Our
report recognizes that DOD has made improvements to its systems for
recording and tracking vaccinations but notes that further improvements
are needed to ensure that data are recorded in an accurate and timely
manner. DOD also stated that the report, "did not mention the excellent
long-term safety record of the vaccine examined over a period of 44
years." Our report notes that GAO's recent work on this issue found that
data on the vaccine's long-term safety is limited. In our previous work,
we found that while some studies have spanned many years, they focus on
short-term reactions to the vaccine. For example, a 20-year study on
reactions to the vaccine only reported on symptoms that began within 48
hours of the vaccination. Moreover, DOD has indicated that additional data
on the vaccine's long-term safety would be beneficial and has established
a committee to identify and plan additional research on this issue.

Finally, DOD noted several actions it has taken or plans to take to
implement our recommendations such as using existing data to develop a
written plan to address possible vaccine shortages and improving DOD
guidance and training on how to report adverse events to the Vaccine
Adverse Event Reporting System. Regarding our recommendation that DOD use
a DOD-wide exit survey to query members whether the requirement to receive
the vaccine affected their decision to resign, DOD noted that it is not
appropriate to single out anthrax vaccinations as a potential reason for
departing the military because it is a "leading" question and would
produce survey bias. Rather, DOD believes that focus groups and surveys of
individuals who refuse to take the vaccine are more appropriate assessment
tools. We believe that DOD should pursue other methods, such as focus
groups, to determine the possible impact of the anthrax vaccine program on
retention but believe that a response category about the anthrax vaccine
could be included on DOD's exit survey since it will be one of many
possible reasons for leaving the military. 

We are sending copies of this report to Representative Bob Stump,
Chairman, and Representative Lane Evans, Ranking Minority Member, House
Committee on Veterans' Affairs. We are also sending copies to the
Honorable William S. Cohen, Secretary of Defense; the Honorable Louis
Caldera, Secretary of the Army; the Honorable Richard Danzig, Secretary of
the Navy; the Honorable F. Whitten Peters, Secretary of the Air Force;
General James L. Jones, Commandant of the Marine Corps and Dr. Jane E.
Henney, Commissioner of Food and Drugs. Copies will also be made available
to others upon request. 

Please contact me at (202) 512-3958 if you have any questions concerning
this report. Key contacts and major contributors to this report are listed
in appendix V.

<Graphic -- Download the PDF file to

Carol R. Schuster
Associate Director, National Security
Preparedness Issues

/Footnote1/-^Apr. 29, 1999).
/Footnote2/-^Contract Management: Observations on DOD's Financial
  Relationship With the Anthrax Vaccine Manufacturer (GAO/T-NSIAD-99-214,
  June 30, 1999).
/Footnote3/-^Medical Readiness: Issues Concerning the Anthrax Vaccine
  (GAO/T-NSIAD-99-226, July 21, 1999).
/Footnote4/-^Before some of the stockpiled lots can be released, FDA must
  approve the results of its required lab tests. Other stockpiled lots
  received FDA approval some years ago but must now pass supplemental
  tests before DOD can use them.
/Footnote5/-^Each lot includes roughly 200,000 doses.
/Footnote6/-^DOD had planned to begin vaccinations in southwest and
  northeast Asia in the summer of 1998. However, in March 1998, when
  hostilities in southwest Asia seemed likely, DOD began vaccinating
  personnel stationed there ahead of schedule.
/Footnote7/-^Medical Readiness (GAO/T-NSIAD-99-148, Apr. 29, 1999).
/Footnote8/-^According to the Code of Federal Regulations (21 C.F.R.
  section 600), purity is the relative freedom from extraneous matter in
  the finished product; potency is the specific ability or capacity of a
  product as indicated by appropriate laboratory tests or adequately
  controlled clinical data; sterility is the freedom from viable
  contaminating microorganisms; and safety is the relative freedom from
  harmful effects to persons affected, directly or indirectly, by a
  product when prudently administered, taking into consideration the
  recipient's condition at the time.
/Footnote9/-^In April 1999, 59 Marines were notified that they had
  received vaccine three weeks after its expiration date. Both the FDA and
  the Armed Forces Epidemiological Board determined that there was no
  concern over the safety or effectiveness of the vaccine. Those notified
  were nonetheless given an option of receiving an additional vaccination
  if they had concerns about the vaccine's efficacy. The Marine Corps
  followed up with a message reminding Marine commanders of the procedures
  for checking expiration dates on all vials of vaccine. Further,
  refresher training was implemented at the base in question and was
  strongly recommended for other medical units.
/Footnote10/-^At the start of the program in March 1998, some of these 31
  lots contained fewer than 20,000 vials because of previous commercial
  sales and military use.
/Footnote11/-^As we noted in our April 1999 testimony, quality cannot be
  guaranteed from final tests alone, only from a combination of in-process
  tests, end-product tests, and strict controls of the entire
  manufacturing process.
/Footnote12/-^In June 1998, on the basis of temperature testing, BioPort
  increased the temperature range for safe shipment of the vaccine from
  2DEG to 8DEG Celsius to 1DEG to 25DEG Celsius. 
/Footnote13/-^The Marine Corps uses the Navy's database.
/Footnote14/-^Although the original stockpile contained 31 lots, we use
  the term "stockpile" to refer to all anthrax vaccine--40 lots in all--
  stored at BioPort before production restarted in May 1999.
/Footnote15/-^BioPort has discussed with FDA completing studies that would
  enable the manufacturer to request FDA approval of release of those lots
  with less preservative (phemerol) than currently required. If these
  studies show that lower amounts of the preservative are effective, and
  if FDA, after reviewing the data, approves lowering the standard, DOD
  may be able to use some or all of these nine lots.
/Footnote16/-^The team suspected but could not confirm that at least some
  of the variances were due to changes in (1) the size, age, and sex of
  the test subjects (guinea pigs); (2) a saline solution used in the
  tests; and (3) the strain of anthrax used in the control group.
/Footnote17/-^The period measured was from the date the manufacturer
  completed lot potency tests to the date FDA approved the results of
  those tests. BioPort needs approval of potency test results as well as
  approval of its renovations which are separate FDA approval processes.
/Footnote18/-^Contract Management (GAO/T-NSIAD-99-214, June 30, 1999).
/Footnote19/-^BioPort sells these doses at a significantly higher price
  than the DOD contract price. DOD has approved the sale of 30,000 doses
  to the Canadian Armed Forces, and BioPort intends to sell the remaining
  40,000 doses to other potential customers. These sales would also
  require approval under export control regulations.
/Footnote20/-^90.6 percent of shipped vials and 99.2 percent of all
/Footnote21/-^Medical Readiness (GAO/T-NSIAD-99-226, July 21, 1999).
/Footnote22/-^Defense Health Care: Medical Surveillance Improved Since
  Gulf War, but Mixed Results in Bosnia (GAO/NSIAD-97-136, May 13, 1997).
  Our comparison of a centralized list of vaccine recipients with their
  medical records in five units revealed that vaccinations had not been
  recorded in 24 percent of medical records. Three of the five units
  failed to record vaccinations in more than 30 percent of medical records. 
/Footnote23/-^We visited one location per service where a large number
  (more than 1,000) of vaccinations had been given: Fort Stewart in
  Hinesville, Georgia, for the Army; the USS Eisenhower, Norfolk Navy
  Shipyard, Portsmouth, Virginia, for the Navy; Langley Air Force Base,
  Hampton, Virginia, for the Air Force; and Camp Lejuene, Jacksonville,
  North Carolina, for the Marine Corps. Our sample of records cannot be
  generalized. See appendix I for more information on our scope and
/Footnote24/-^DOD's policy is to adhere to the approved immunization
  schedule and to make deviations to the schedule the exception rather
  than the rule. According to DOD policy, the effect of deviations from
  this schedule on the efficacy of the vaccine is unknown, but in general,
  the greater the deviation, the less certain the protective effect in
/Footnote25/-^See section 581 of Public Law 100-65, October 5, 1999.
/Footnote26/-^Adverse events are adverse outcomes for which a cause and
  effect relationship with an exposure (to a vaccine or a medication) has
  not yet objectively been determined. An adverse event becomes an adverse
  reaction once objective evidence is available to establish a cause-and-
  effect link between an exposure and an adverse outcome.
/Footnote27/-^Military medical personnel reported 109 of these. 
/Footnote28/-^Of 174 reports reviewed by DOD, 20 met this criteria.
/Footnote29/-^As noted in appendix I, respondents were not randomly
  selected, and thus the data cannot be projected beyond those surveyed.
/Footnote30/-^As we testified in April 1999, data from this study was
  based on a different vaccine than the one eventually licensed. FDA
  reported that the method of preparing the licensed product was similar
  but not identical to the vaccine used in the study and that production
  changes for the licensed vaccine were "minor."
/Footnote31/-^In active monitoring, vaccine recipients are contacted to
  ascertain if there were any adverse reactions to the vaccine after
  vaccine administration. See Medical Readiness (GAO/T-NSIAD-99-148, Apr.
  29, 1999).
/Footnote32/-^(3 respondents, or 3 percent). Respondents were not limited
  to one response.
/Footnote33/-^The symptoms reported by these five individuals included
  burning sensations, colds, need for more sleep, memory problems, fevers,
  headaches, nausea, lower blood pressure, viral infections, fainting
  spells, chronic sinus problems never previously experienced, fevers, and
  blood in the stools.


To conduct our review, we interviewed officials and obtained documents
from the Army Office of the Surgeon General's Anthrax Vaccine Immunization
Program; the Joint Program Office for Biological Defense; the Naval
Medical Information Management Center; the Offices of the Judge Advocates
General for the Army, the Navy, Marine Corps, and the Air Force; and the
Joint Staff. We also obtained information and discussed the program with
officials from the Defense Manpower Data Center (DMDC) in Seaside,
California, and Arlington, Virginia; U.S. Air Force Air Combat Command,
Langley, Virginia; U.S. Navy Space and Warfare Systems Command,
Chesapeake, Virginia; medical and command personnel at Fort Stewart,
Georgia; USS Eisenhower, Norfolk Naval Shipyard, Portsmouth, Virginia;
Langley Air Force Base, Virginia; and Camp Lejeune, Jacksonville, North
Carolina. In addition, we interviewed officials and obtained documents
from BioPort Corporation in Lansing, Michigan; and the Food and Drug
Administration (FDA) in Rockville, Maryland.

To determine the availability of the vaccine and its impact on program
schedules, we reviewed and summarized data on vaccine lot status,
including supplemental test results, lot quantities, lot expiration dates,
and results of initial lot release testing. We analyzed assumptions of
projections for vaccine production and usage and compared them with
program schedules and past testing data. We also discussed measures for
securing and shipping the vaccine with officials from BioPort, the U.S.
Army Medical Materiel Agency, and one installation at each service.

To assess systems for recording and tracking vaccinations, we selected one
installation from each service where a large number of vaccinations had
been given (at least 1,000) and randomly selected 300 service members who
had received at least one injection of the vaccination series./Footnote1/
We then compared the information on the paper records with data from the
Defense Enrollment Eligibility Reporting System (DEERS). Table 3
summarizes the installations visited, records reviewed, and time frames of
our collection of DEERS and paper data.

Table****Helvetica:x11****3:    Collection and Review of Electronic and
                                Paper Records

| Service         : Populati : Medica : Yellow :      Date :     Date(s)  |
| location visited:  on that :      l :   shot :     DEERS : paper record |
|                 : received : record : record :  data was :    data was  |
|                 :       at :      s :      s : received  :    reviewed  |
|                 :    least : review : review :           :              |
|                 :      one :     ed :     ed :           :              |
|                 : vaccinat :        :        :           :              |
|                 :      ion :        :        :           :              |
| Army: Fort      :    8,751 :    200 :    197 :    1 Dec. :  14-17 Dec.  |
| Stewart, GA     :          :        :        :      1998 :        1998  |
| Navy: USS       :    2,108 :    238 :      1 :    2 Feb. :  16-17 Feb.  |
| Eisenhower, VA  :          :        :        :      1999 :        1999  |
| Air Force:      :    1,273 :    186 :    143 :    9 Nov. : 30 Nov. 1998 |
| Langley AFB, VA :          :        :        :      1999 :              |
| Marines: Camp   :    1,842 :    134 :      4 :   10 Mar. :  15-18 Mar.  |
| Lejeune, NC     :          :        :        :      1999 :        1998  |

Source: GAO.

We compared the vaccination number, date, and lot number contained in the
DEERS database with data on paper records--the medical record and yellow
shot records available on site. A mismatch of any vaccination for each
category was considered a mismatch for the entire record. Because our
samples included only those who had received at least one injection, our
analyses did not examine the possible condition that a servicemember
received an injection but did not have it recorded in DEERS. Further,
although our initial sample of records was designed to project our results
to the installations we visited with a precision of +/-5 percent at a 95-
percent confidence level, operational limitations in the field--most
notably the unavailability of some records because of deployments and
transfers--did not allow us to review sufficient records to generalize our
results to all personnel at the four installations with a reasonable level
of confidence.

To evaluate the reporting of vaccine-related adverse events, we reviewed
FDA requirements for the Vaccine Adverse Event Reporting System (VAERS),
obtained reports of adverse events from the program, discussed reporting
procedures with medical and command personnel at the four military
installations we visited, and reviewed additional Department of Defense
(DOD) studies on adverse events. In addition, we attended the May 1999
Annual DOD Conference for Biological Warfare Defense Immunizations.

To assess education initiatives of the program, we reviewed guidance and
service plans to determine education requirements; collected and reviewed
educational material used at the military installations we visited,
discussed education efforts with command and medical personnel at each
installation and with FDA officials, and surveyed a total of 249
servicemembers at those installations. We did not evaluate the accuracy of
information provided to vaccine recipients but used the survey to
determine what information was available to servicemembers and how helpful
they found the information. Questionnaire respondents were, with three
exceptions, vaccine recipients who were available at the time of our site
visits. Because the respondents were not randomly selected, their
responses cannot be projected. Details of the questionnaire and responses
are in appendix III. We also discussed program plans for future
communications and education initiatives with program officials.

We conducted our review from July 1998 through July 1999 in accordance
with generally accepted government auditing standards.

/Footnote1/-^Files for the Fort Stewart location inadvertently included
  the records for the first 300 social security numbers, and therefore
  were not random.


This appendix describes DOD's packaging and shipping protocol for
transporting anthrax vaccine from BioPort to military sites. DOD's packing
and shipping goals are to have zero defects (such as package damage that
would ruin the vaccine) and zero loss of accountability (such as packages
disappearing due to mishandling or theft).


BioPort packages vials of anthrax vaccine according to the protocol
designed by DOD and BioPort to maintain doses within an acceptable
temperature range (1-25DEGC). The vials are shipped in an insulated
container along with gelatin cold-packs, a digital monitor that records
the temperature every 5 minutes throughout transit, an addressed envelope
for return of the monitor, and an address label for return of the
packaging materials to BioPort. In tests of the temperature monitor, DOD
found its failure rate to be just under 1 percent****Symbol:xbe****usually
due to a mechanical or electrical problem. There are several layers in
each container:

o   The first layer is composed of two gelatin cold-packs. In spring and
  fall, one of the packs is frozen before packing; in summer, both are
  frozen. In winter, neither is frozen.

o   The second layer is made of cold-packs that are never frozen before
  shipping. The vaccine vials and the temperature monitor are packed
  between the second and third layers.

o   The third layer holds two more cold-packs cooled to 4DEGC. 

The highest temperature recorded since use of this protocol began (in a
shipment sent to southwest Asia) has been 16DEGC.


DOD's shipments of anthrax vaccine are managed by the U.S. Army Medical
Materiel Agency (USAMMA). Shipments in the continental United States,
nearly all of which are by air, are performed by Federal Express. Some
overseas shipments are also carried by Federal Express, but most are
delivered by DHL World Wide Express. Should either Federal Express or DHL
World Wide Express go on strike, the other carrier would take over
delivery of shipments.

The shipping label on each box has a code to track the package, giving DOD
"total asset visibility." As part of its Priority Alert program, Federal
Express gives DOD's shipments priority and aggressively pursues solutions
to problems that arise. The shipping box carries fluorescent "Priority
Alert" labels on all sides to notify handlers that the box must be moved
first and never bumped. If a Priority Alert shipment is held up by
problems with Federal Express' transportation vehicles, the company
immediately arranges with a common carrier to move the shipment. Federal
Express employees take procedural problems uncovered through this program
directly to the company's managers for priority resolution.

Federal Express has given USAMMA a computer system to track shipments, and
pagers are used for the two organizations to maintain
24-hour communication. BioPort enters information on an outgoing vaccine
shipment into the Federal Express system, establishing instant visibility.
The program can also generate reports that identify, among other things,
systemic problems with shipments to a particular military installation.
USAMMA, thus alerted, can check with the site and clarify the situation.
Special software, PC Track, will soon link USAMMA to Federal Express'
mainframe computer and provide more communication regarding shipments.

USAMMA notifies military recipients beforehand of imminent shipments and
gives instructions to alert local security about the shipment and verify
that proper refrigeration will be available in the receiving area. USAMMA
also faxes them a checklist to be used when the shipment arrives. Upon
receipt, the recipient visually inspects the package for damage. If
damaged, the recipient is to refuse shipment and contact USAMMA. The
military recipient then refrigerates the vaccine at 2-8DEGC in a
restricted area and returns the monitor to USAMMA. The recipient awaits
authorization from USAMMA, which checks that temperature data recorded by
the monitor did not exceed temperature tolerances before releasing the
vaccine. If the package's interior temperature has been too high or low at
any point in transit, it shows up on the monitor's read-out as a positive
or negative spike (if the box were opened en route, for example, a
positive spike would be recorded). Any deviation is recorded on a special
form and sent to BioPort for assessment.

When a shipping problem occurs, USAMMA conducts a risk analysis that runs
through an "if/then" protocol. Also, whenever a route is changed, USAMMA
runs a test shipment of one vial.


We surveyed vaccine recipients in all four services about the anthrax
vaccine program and obtained responses from 249 active duty
servicemembers: 18 percent (44) in the Army, 12 percent (31) in the Navy,
34 percent (85) in the Air Force, and 36 percent (89) in the Marine Corps.
Because our survey participants were not randomly selected, the survey
results cannot be projected to a larger military population.

o   About 89 percent (220) were enlistees and 11 percent (28)

o   56 percent (140) were between the ages of 18 and 25, the other
  44 percent were almost equally distributed between the ages of 26
  and 33 and 34 and 49. Most participants in the Army, the Navy, and
  the Marine Corps were between 18 and 25, while those in the Air
  Force tended to be older. 

o   About 86 percent (213) were men. 

o   Approximately 65 percent (162) identified themselves as Caucasian,
  22 percent (54) as Black, and the remaining 13 percent (32) as
  either Hispanic American, Native American, or Asian American. One
  participant did not respond to the question.

o   The number of respondents for each question varied because they were
  instructed to skip questions that did not apply to their individual

Servicemembers Wanted Information on Possible Long-term Effects of the
Vaccine and Other Issues

Two-thirds (164) of survey participants said they wanted information they
had not received, including information on temporary or short-term side
effects of the vaccine, possible long-term side effects, the vaccination
routine after active duty, the anthrax threat, or other information.
Participants from all four services also said they wanted information they
had not received, especially on possible long-term side effects (about
43 percent--106--of all participants). Relative to their peers from the
other services, a higher proportion of Air Force participants expressed a
need for information they had not received, particularly on possible long-
term side effects (see fig. 5).

Figure****Helvetica:x11****5:    Percentage of Participants Wanting More

<Graphic -- Download the PDF file to

Source: GAO.

Examples of topics not listed in the survey about which respondents wanted
more information included why more than three vaccinations are necessary,
whether the vaccine has been tested by a qualified source, the history of
the vaccine, the anthrax disease, and the extent to which the vaccine has
been used to immunize humans.

Figure 6 shows the percentage of participants, by service, who responded
that they experienced reactions. The Air Force had the highest rate (68
percent, or 58 out of 85 respondents).

Figure****Helvetica:x11****6:    Percentage of Respondents Reporting Short-
                                 term Adverse Effects, by Service

<Graphic -- Download the PDF file to

Note: Does not include three respondents who had not received their first

Source: GAO.

Of the 111 survey participants who said they had experienced short-term
reactions, 57 percent (59) said they wanted information on possible long-
term adverse effects.

Figure 7 shows the percentage of participants in three age groups who
reported having adverse effects and who said they wanted more information
on possible long-term effects. 

Figure****Helvetica:x11****7:    Percentage of Respondents Reporting
                                 Adverse Effects and Wanting Information,
                                 by Age Group

<Graphic -- Download the PDF file to

a Excludes three respondents who had not yet received their first
vaccination. The numbers in each group are: 18 to 25 years, 138
respondents; 25 to 33 years, 55 respondents; 34 to 49 years, 53 respondents.

b Includes all respondents. The numbers in each group are: 18 to 25 years,
140 respondents; 25 to 33 years, 56 respondents; 34 to 49 years, 53

Source: GAO.

As shown in figure 8, participants in all race categories said they had
experienced adverse effects and wanted information on possible long-term
adverse effects. 

Figure****Helvetica:x11****8:    Percentage of Respondents Reporting
                                 Adverse Effects and Wanting Information,
                                 by Race

<Graphic -- Download the PDF file to

a Does not include one respondent who did not specify race and three who
had not received their vaccination. The numbers for each group are: Black,
53 respondents; Caucasian, 161 respondents; other, 31 respondents.

b Does not include one respondent who did not specify race. The numbers
for each group are: Black, 53 respondents; Caucasian, 162 respondents;
other, 32 respondents. 

Source: GAO.

/Footnote1/-^ One participant in the survey did not indicate military rank.


<Graphic -- Download the PDF file to

<Graphic -- Download the PDF file to

<Graphic -- Download the PDF file to

<Graphic -- Download the PDF file to

<Graphic -- Download the PDF file to


GAO Contacts

Norman J. Rabkin (202) 512-5140
Christine Fossett (202) 512-2956


In addition to those named above, Margaret Best, Jack Edwards, Julia
Kennon, Melissa McDowell, MaeWanda Michael-Jackson, and Michael Whitlock
made key contributions to this report.



Medical Readiness: Issues Concerning the Anthrax Vaccine
(GAO/T-NSIAD-99-226, July 21, 1999).

Contract Management: Observations on DOD's Financial Relationship With the
Anthrax Vaccine Manufacturer (GAO/T-NSIAD-99-214, June 30, 1999).

Combating Terrorism: Observations on Growth in Federal Programs (GAO/T-
NSIAD-99-181, June 9, 1999).

Medical Readiness: Safety and Efficacy of the Anthrax Vaccine
(GAO/T-NSIAD-99-148, Apr. 29, 1999).

Gulf War Illnesses: Questions About the Presence of Squalene Antibodies in
Veterans Can Be Resolved (GAO/NSIAD-99-5, Mar. 29, 1999).

Combating Terrorism: Observations on Biological Terrorism and Public
Health Initiatives (GAO/T-NSIAD-99-112, Mar. 16, 1999).

Combating Terrorism: Observations on Federal Spending to Combat Terrorism
(GAO/T-NSIAD/GGD-99-107, Mar. 11, 1999).

Chemical and Biological Defense: Observations on DOD's Plans To Protect
U.S. Forces (GAO/T-NSIAD-98-83, Mar. 17, 1998)

Combating Terrorism: Efforts to Protect U. S. Forces in Turkey and the
Middle East (GAO/T-NSIAD-98-44, Oct. 28, 1997).

Combating Terrorism: Status of DOD Efforts to Protect Its Forces Overseas
(GAO/NSIAD-97-207, July 21, 1997).

Defense Health Care: Medical Surveillance Improved Since Gulf War, But
Mixed Results in Bosnia (GAO/NSIAD-97-136, May 13, 1997)


*** End of document. ***