mining whether an adverse health outcome is associated with receiving a vaccine. Several Institute of Medicine reports (IOM, 1991, 1994, 1997) have examined the complex issues involved in vaccine safety in greater depth.
Postmarketing surveillance of licensed vaccines in the United States relies on the voluntary reporting of adverse events. In 1990, the Vaccine Adverse Event Reporting System (VAERS) became operational and is overseen jointly by the Centers for Disease Control and Prevention (CDC) and the Food and Drug Administration. VAERS reports are open-ended forms that allow for description of the symptoms, time course, laboratory tests, vaccine(s) received, and treatment provided. While health care providers and manufacturers are obligated to report specific adverse effects to vaccines covered by the National Vaccine Injury Compensation Program, anyone can file a VAERS report. For the 65,720 VAERS reports received between January 1, 1991, and December 31, 1996, the sources of reports were health care providers (47.3 percent), manufacturers (39 percent), parents (2.4 percent), and others or unknown (11.3 percent) (CDC, 1999b). There is no long-term follow-up mechanism for VAERS reports.
VAERS is a passive reporting system in that it relies on incoming reports. Adverse events are therefore likely to be underreported (IOM, 1997). Further, some reports have incomplete medical information, and the same case may be reported by different sources. VAERS data are useful in signaling potential new adverse events but are limited in their usefulness for assessing the rate or causality of adverse events (IOM, 1994). Although the number of doses distributed is usually available, the number of doses administered is not. Further, the extent of underreporting of adverse events is unknown. FDA and CDC are responsible for monitoring VAERS data to detect unusual trends and occurrences of adverse health effects. This monitoring assists FDA and CDC in responding appropriately to adverse events.
Studies of vaccine safety use either active or passive methods of surveillance in assessing the extent of adverse events. Active surveillance methods involve direct follow-up by investigators of all individuals in the study. At a minimum, active surveillance seeks to systematically contact all vaccine recipients at prespecified intervals following vaccination. Often, in addition to posing open-ended questions about possible adverse effects, active surveillance asks explicitly about specific symptoms, and sometimes specific physical or laboratory examinations are conducted. Passive surveillance methods rely on the vaccine recipient to provide information (e.g., self-reports, surveys) or use other information that may indicate adverse outcomes (e.g., days missed from work, number of visits to the clinic following vaccination). Studies on the botulinum toxoid and anthrax vaccines have relied primarily on passive surveillance approaches and have involved only relatively short periods of follow-up.