News & Events
FDA's Vaccine Adverse Event Reporting System
Susan S. Ellenberg, Ph.D.
Division of Biostatistics and Epidemiology
Center for Biologics Evaluation and Research
Food and Drug Administration
Department of Health and Human Services
the Subcommittee on National Security, Veterans Affairs, and International Relations
House Committee on Government Reform
July 21, 1999
Mr. Chairman and Members of the Committee, I am Susan S. Ellenberg, Director of the Division of Biostatistics and Epidemiology, Center for Biologics Evaluation and Research (CBER), Food and Drug Administration (FDA or the Agency). I appreciate the opportunity to discuss FDA's Vaccine Adverse Event Reporting System (VAERS), which is designed to receive and evaluate reports of adverse events following vaccinations, and its interface with the Department of Defense's (DOD) Anthrax Vaccine Immunization Program (AVIP). As requested by the Committee, I will provide an overview of the VAERS system, the evaluation and review of the information that is obtained through these reports, and the Agency's experience with adverse events reports for the anthrax vaccine.
The Importance of Vaccine Safety
Vaccines are among the most significant public health interventions of all time, and have been responsible for saving millions of lives and preserving health worldwide. Nevertheless, like all other medical products, vaccines are not entirely risk-free. While serious complications are extremely rare, they can occur. Vaccines are unique in that they are administered to healthy individuals and there is virtually universal exposure of our population to vaccines. Therefore, it is important to identify even these very rare adverse reactions.
VACCINE ADVERSE EVENT REPORTING SYSTEM
The Vaccine Adverse Event Reporting System (VAERS) resulted from enactment of the National Childhood Vaccine Injury Act of 1986 (NCVIA), 42 U.S.C. ( 300aa-1 et seq., as amended, which was aimed at improving childhood vaccine safety and mandated reporting of certain adverse events associated with vaccines. The NCVIA led to the creation of a unified national system to collect, manage and evaluate these adverse event reports. The VAERS system was initiated in 1990 and is jointly managed by FDA and the Centers for Disease Control and Prevention (CDC). VAERS receives reports from vaccine manufacturers, private practitioners, State and local public health clinics, and vaccinees themselves (or their parents or guardians). Certain of these reports are required to be filed under a mandatory reporting requirement. Vaccine manufacturers, however, must report to FDA all reports of adverse events of which they are aware.
Occasionally, reports are filed with the MedWatch system, which is FDA's larger adverse event reporting system for medical products other than vaccines. Adverse event reports to MedWatch involving vaccines are transferred immediately to the VAERS system, regardless of whether they originate by phone, fax or mail.
VAERS is similar in intent and operation to surveillance systems for other types of FDA-regulated products maintained by the Agency and to safety surveillance programs in other countries. While the motivation for creating VAERS came from the NCVIA, VAERS accepts all reports of suspected adverse events after administration of any U.S. licensed vaccine to individuals in any age group.
POST-MARKETING SURVEILLANCE SYSTEMS
VAERS is a "passive" surveillance system. This means that it relies on health professionals, patients or guardians to submit reports of adverse reactions following vaccination. (An "active" surveillance system, in contrast, would follow all individuals in a defined population to determine their responses to vaccination.) To encourage reporting of any adverse event suspected of being vaccine-induced, the criteria for reporting to VAERS are non-restrictive. In effect, the system accepts and includes any report submitted, no matter how tenuous the possible connection with vaccination might seem.
These types of systems are essential to the discovery of potential rare adverse consequences of medical products that may not become evident until many thousands or millions of people have been exposed to them. While they are critical to FDA's post-marketing surveillance, there are important limitations to the interpretation of the data, however, as discussed below.
OVERVIEW OF VAERS ACTIVITIES
VAERS receives 11,000 to 12,000 reports per year. Approximately 15 percent of the reports describe a "serious" event, which is considered to be either fatal, life-threatening, or resulting in hospitalization or permanent disability. Most of the remaining reports describe self-limited, transient events such as injection site reactions, irritability, prolonged crying and fever.
All reports are entered into a computer database. Reports of serious events and fatalities are followed up individually by a health professional. Autopsy reports and other relevant medical records are sought and retrieved for review. Medical staff carefully monitor individual reports and trends in adverse event reporting for vaccines, with particular attention to newly licensed vaccines.
VAERS data are available to the public through the National Technical Information Service and also through requests to FDA's Freedom of Information office. Patient identifiers are removed from all data provided to the public. General information and the VAERS form itself are available on the VAERS Internet website.
OBJECTIVES OF VAERS
Spontaneous report-based surveillance programs, such as VAERS, perform a critical function by generating signals of potential problems that may warrant further investigation. As such, VAERS is the "front line" of national vaccine safety surveillance. It is especially valuable in assessing the safety of newly marketed vaccines. Careful review of reports during the initial months following licensure can provide additional assurance about the safety of a new vaccine, uncover previously unexpected events which occur when a vaccine is used in a larger and more diverse population than was studied in clinical trials, or rapidly identify potential problems not observed pre-licensure.
Although VAERS has methodological limitations inherent in passive surveillance systems, VAERS is essential to the U.S. vaccine safety monitoring system. It is the only surveillance system that covers the entire U.S. population and therefore includes the largest number of case reports of events temporally associated with vaccination in the U.S. It provides timely availability of data from a geographically diverse population, allowing rapid detection of possible new, unusual or rare adverse events. Such detection generates hypotheses that may then be tested in other databases.
Based on careful review, analysis and further investigation of spontaneous reports, FDA can initiate various actions: manufacturers' labeling or packaging change(s), conducting or requesting manufacturer-sponsored post-marketing epidemiological investigations (hypotheses testing in more rigorous databases); issuing a Safety Alert or "Dear Health Professional" letter, inspecting manufacturers' facilities/records, or working with a manufacturer regarding possible withdrawal of vaccine from the market (for safety or efficacy reasons). Keeping vaccine labeling/package inserts up-to-date is an ongoing, dynamic process that depends on new information gleaned from spontaneous adverse event reports as well as other sources. Dissemination of safety-related information to healthcare professionals and the public is an important health goal of post-marketing surveillance.
LIMITATIONS OF VAERS
While assessment of VAERS data is often the first step in identifying potential new information about the safety of vaccines, it is important to recognize that VAERS data alone are usually inadequate for drawing firm conclusions or providing a basis for regulatory actions. Many reports omit important data and/or contain obvious errors that may not be easily identifiable or correctable. Multiple vaccines are frequently administered simultaneously, according to currently recommended vaccine schedules, making it difficult or impossible to determine which (if any) of the vaccines administered was the possible cause of the event. The extent of under-reporting of events occurring after vaccination is unknown, and the number of individuals in subgroups of interest (for example, infants) receiving the vaccine during specific time intervals is not known, so that incidence rates cannot be calculated. In addition, because VAERS accepts and encourages reports of all temporal associations, regardless of the rationale for the vaccine being the cause of the outcome reported, there is also "over-reporting" since many events reported, and entered in the database, are most likely not attributable to vaccination.
Probably the most important limitation of VAERS, as it is for any passive reporting system, is its inability to establish causality for most reports it receives. Adverse events occurring in unvaccinated individuals are not reported, so there is no "control group" to study. Most of the types of serious adverse events reported to VAERS can occur in unvaccinated as well as vaccinated individuals. Without an unvaccinated group it is usually impossible to assess whether the number of reported events is different from the number that would have been observed in the absence of vaccination.
Even if a vaccine is not the cause of certain rare medical problems, when a vaccine is used in a large population, it is a certainty that some number of these events will occur within a short interval following a vaccination. For this reason, the fact that an event --even a very serious event such as a death -- occurs shortly after a vaccine has been administered cannot by itself lead to the conclusion that the event was caused by the vaccine.
An adverse event can be causally attributed to a vaccine more readily if:
- The event conforms to a specific clinical syndrome whose association with vaccination has strong biologic plausibility (e.g., anaphylaxis within 30 minutes after vaccination).
- A laboratory result confirms association (e.g., isolation of vaccine strain varicella vaccine from skin lesions of a patient with rash).
- The event recurs on re-administration of vaccine ("positive rechallenge").
- A controlled clinical trial or well-designed epidemiological study shows greater risk of adverse events among vaccinated than unvaccinated (control) groups.
Because few of the serious adverse events reported to VAERS meet any of the first three criteria (one such example, however, is described below), and because clinical trials are almost always too small to provide useful information on rare events, methodologically more rigorous epidemiological studies must be conducted to assess causality for most serious adverse events that are investigated. A determination that the vaccine caused the post-vaccination event usually cannot be made on the basis of information acquired from individual VAERS reports.
CONTRIBUTIONS OF VAERS DATA TO UNDERSTANDING VACCINE SAFETY
Several investigations of VAERS data have uncovered previously unrecognized problems that may occur rarely in vaccine recipients. FDA investigators noted occasional instances of life-threatening thrombocytopenias (low platelet counts) following the administration of MMR (measles, mumps, rubella) vaccine, a previously unappreciated level of severity of a known side effect. Other FDA investigators documented a series of cases in which hair loss followed immunizations (primarily hepatitis B vaccine), a rare effect not previously reported. Because some of these cases exhibited "positive rechallenge," as defined earlier, there is a greater level of confidence that these outcomes truly may have been caused by the vaccine. In another study, FDA staff identified a series of cases of severe injuries resulting from vaccination-induced fainting or syncope. These outcomes did not appear related to any specific vaccine, but were most probably attributable to the act of vaccination itself. Sometimes VAERS data may provide the useful and reassuring information that new problems have not been identified after additional experience with a vaccine, as with our review of reports for Hepatitis A and B vaccines.
Trends in Reporting
VAERS data also have been used to compare reporting patterns over time and investigate changes in reporting rates that might be due to changes in vaccine practices. For example, CDC epidemiologists reviewed reports of fever, seizures, and hospitalizations following administration of a newly licensed combination of diphtheria, tetanus and acellular pertussis vaccine (DTaP). The rate of such reports was about one-third lower than the reporting rate following the standard DTP vaccine, consistent with-and confirming in the context of general practice-the safety findings of the pre-licensure clinical trials.
As we discussed in FDA's testimony before this subcommittee on April 29, anthrax vaccine was licensed by the National Institutes of Health's Division of Biologic Standards (the predecessor agency to CBER) in 1970 for protection against anthrax, a highly infectious and often fatal disease caused by spores of the Bacillus anthracis bacterium. Experience has shown that inhalation anthrax has a very high mortality rate, with estimates ranging from 80 percent to 90 percent or higher.
Since licensure, the anthrax vaccine has been used by livestock workers, veterinarians, lab workers and researchers at risk for infection, and more recently, as a preventive measure against a possible biologic weapons attack utilizing anthrax against U.S. armed forces. According to the manufacturer, from 1974 to 1989, approximately 68,000 doses of the vaccine were distributed. In 1990, approximately 268,000 doses were distributed. Between 1991 and the present, we understand that approximately 1,200,000 doses were distributed.
It is not possible to give a precise number of persons who received the vaccine prior to 1990. We estimate that between 1966 and 1971, approximately 7,000 at-risk persons received approximately 16,000 doses of the vaccine in a study conducted by the CDC. In addition, between 1974 and 1989, our files show approximately 68,000 doses were distributed. This is sufficient to vaccinate about 11,000 people with the full six-dose regimen of the currently approved anthrax vaccine. It is possible, however, that some doses distributed were not used, or that some individuals did not receive the full course of the vaccine.
Adverse Event Reporting for Anthrax Vaccine
FDA receives adverse event reports on the anthrax vaccine through a system similar to other adverse event reporting systems within the Agency. They are filed directly by health professionals as well as by patients or families. Reporting of adverse events associated with the use of anthrax vaccine is voluntary for individual healthcare providers but, as stated above, the vaccine manufacturer must report to FDA all reports of adverse events of which they are aware. It should be emphasized that adverse event reports can be made by a healthcare professional, a patient or anybody else. If a patient's physician does not file a VAERS report, the patient can do so. FDA protects the confidentiality of patients reporting adverse events. FDA encourages individuals to report to VAERS any clinically significant adverse event occurring after the administration of any vaccine licensed in the United States. Reports to VAERS may be made in writing or by calling a toll-free number, 1-800-822-7967. Reporting instructions are available on the Internet at FDA's website.
CBER handles numerous inquiries from individuals concerning the anthrax vaccine. Individuals who believe they have experienced an adverse reaction are encouraged to report and provide information on filing a VAERS report. Forms are mailed and faxed to individuals upon request and individuals also are referred to FDA's website.
VAERS Reports on Anthrax Vaccine
Since the beginning of VAERS operations in 1990, through July 1, 1999, 215 reports of adverse events associated with use of the anthrax vaccine have been reported to VAERS. Of those, 22 are considered serious events, as defined earlier. These reports are for diverse conditions, with no clear patterns emerging at this time. Some of these events are described below. The remaining reports describe a variety of symptoms, including injection site edema (swelling with fluid in tissue), injection site hypersensitivity, rash, headache and fever.
The 22 serious events were reported to have occurred or been diagnosed at times ranging from 45 minutes to 4 1/2 months after vaccination. Some individuals experienced adverse events following the first dose; others received up to 5 doses before event onset. Most of these individuals reporting adverse events during the current anthrax vaccination program have recovered. Five patients were hospitalized for severe injection site reactions. One individual experienced a more widespread allergic reaction. One individual was hospitalized with a confirmed case of aseptic meningitis nine days after vaccination. Two individuals experienced Guillain-Barr syndrome. Three weeks after receiving the vaccine, another individual was diagnosed with bipolar disorder and has not recovered. One individual experienced onset of multi-focal inflammatory demyelinating disease and has since recovered. Another individual experienced onset of lupus and has not recovered.
None of these events, except for the injection site reactions, can be attributed to the vaccine with a high level of confidence, nor can contribution of the vaccine to the event reported be entirely ruled out. It should be emphasized once again that it is not always possible to attribute a cause and effect relationship between a reported event and a vaccination. With the exception of injection site reactions, all of the adverse events noted above do occur in the absence of immunization.
While the data gathered from the VAERS system can serve as a useful tool in identifying potential problems, the reports on anthrax vaccine received thus far do not raise any specific concerns about the safety of the vaccine. As more people receive the vaccine, the numbers of adverse events reported will increase. FDA continues to view the anthrax vaccine as safe and effective for individuals at risk of exposure to anthrax.
DOD's Anthrax Vaccine Immunization Program
FDA has not had an official role in the development or operation of the Department of Defense's (DOD) Anthrax Vaccine Immunization Program (AVIP), including the AVIP tracking system or the program's adverse event reporting system. In March 1997, DOD briefed FDA about their draft plan for the possible use of the anthrax vaccine to inoculate U.S. military personnel. Subsequently, FDA learned that the plan had been adopted by DOD.
Anthrax Vaccine Expert Committee
CDC was approached by DOD with a request to conduct additional reviews of adverse events reported for the anthrax vaccine. In July 1998, CDC requested that FDA participate in a program to evaluate VAERS reports for the anthrax vaccine. In response to the request by DOD, a group of non-government medical experts was convened in the fall of 1998 as the Anthrax Vaccine Expert Committee (AVEC). AVEC, which is coordinated by the Health Resources and Services Administration, has met six times since 1998. These experts have been reviewing all reports for the anthrax vaccine. Representatives of FDA, CDC and DOD attend meetings and FDA has provided information to assist the committee in its deliberations. AVEC is unique in that it provides an independent civilian expert assessment of adverse events reported for the anthrax vaccine.
Deployment of Partially-Vaccinated Service Personnel
The Committee requested that FDA "discuss the implications under FDA regulations of DOD policy declaring military personnel eligible for deployment to threat areas after only two of the six inoculations in the FDA-approved AVA immunization program." The Committee's concern centers on the question of whether a deployment before the full course of vaccinations is complete constitutes regulated activity.
FDA continues to view the anthrax vaccine as safe and effective for individuals at high risk of exposure to anthrax, when used in accordance with the approved labeling. That labeling calls for a six dose series of inoculations, with the initial dose followed by doses at two and four weeks, and at six, twelve and eighteen months. Data from the Brachman study submitted in support of licensure, and from CDC surveillance, suggest that individuals who receive less than the full series of inoculations may be at a higher risk of becoming infected than persons receiving the full series of doses.
FDA is not aware of any decision by DOD on deployability which postulates that less than the full course of six inoculations would provide full protection against anthrax. Because DOD has undertaken a program of full force protection against anthrax but has a continuing need to deploy service personnel in support of national defense, the issue of deployability involves a balancing of interests that is well outside of FDA's jurisdiction and expertise.
FDA evaluates the risks and benefits, both known and potential, for all FDA-regulated medical products. So far, the data gathered from VAERS reports on anthrax vaccine do not signal concerns about the safety of the vaccine. The Agency will continue to closely monitor and investigate reports of serious adverse events received on all vaccines, including anthrax.
Vaccine safety is a high priority of FDA and the Agency considers all of its safety programs, including VAERS, as critical to carrying out its goals. Thank you for this opportunity to discuss VAERS and our efforts to monitor and ensure the safety of the anthrax vaccine.