[NSIAD-00-36] Medical Readiness: DOD Faces Challenges in Implementing Its Anthrax Vaccine Immunization Program , Part 1/2
in...@www.gao.gov
Message-ID: <GAORPTns...@us.govnews.org>
MIME-Version: 1.0
Medical Readiness: DOD Faces Challenges in Implementing Its Anthrax
Vaccine Immunization Program (Letter Report, 10/22/1999,
GAO/NSIAD-00-36).
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
program.
--------------------------- Indexing Terms -----------------------------
REPORTNUM: NSIAD-00-36
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: **
** **
** <in...@www.gao.gov> **
** **
** 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
Program
*****************
<Graphic -- Download the PDF file to
view.>
*****************
GAO/NSIAD-00-36
Letter 3
Appendixes
Appendix I:Scope and Methodology
38
Appendix II:Packaging and Shipping Protocol
41
Appendix III:Survey of Servicemember Views of the Anthrax Vaccine
Immunization Program
43
Appendix IV:Comments From the Department of Defense
48
Appendix V:GAO Contacts and Staff Acknowledgments
53
Related GAO Products
54
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-
99-148,
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
infections
National Security and
International Affairs Division
B-283133
October 22, 1999
The Honorable Arlen Specter
Chairman
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
anthrax.
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
manufacturer
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.
Background
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
vaccinations./Footnote6/
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
vaccinations.
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
protocol.
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
view.>
*****************
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
view.>
*****************
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
Possible
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
improvement./Footnote22/
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
view.>
*****************
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
data.
Possible Adverse Events Are Monitored, but DOD's Use of Data May Be
Misleading
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
reactions.
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
age.
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
surveyed).