[Table of Contents]

Volume: 23S4 - May 1997

Canadian National Report on Immunization, 1996

9. Surveillance of Adverse Events Temporally Associated with Vaccine Administration

9.1 Background

As indicated in the section, "Immunization in Canada," all vaccines undergo extensive testing for safety and efficacy before being licensed. Quite often, large-scale clinical trials have already taken place, and sometimes the product is already in use in other countries as well. However, pre-market assessments may not be able to detect all the potential side effects of a vaccine, nor can they detect the rare conditions that occasionally arise once a product is in use on a larger scale and in the general population. This does not suggest that unsafe products are being licensed, only that rare adverse events which do not affect the overall benefit/risk of the vaccine may still occur. For this reason, most countries have systems in place for the post-marketing surveillance of drugs. Vaccines are biologic products that require lot-by-lot testing before being released. This potential lot-by-lot variation further strengthens the need for post-marketing surveillance.

The VAAE surveillance system was developed to monitor vaccine safety and maintain public confidence in vaccines and immunization programs. The specific objectives of the system are as follows:

• to identify illnesses of infrequent occurrence that may be caused by a vaccine;
• to develop estimates of rates of occurrence of more serious illnesses following immunization by type of vaccine;
• to monitor for any unusually high rates of adverse events, both with individual vaccines and individual lots of vaccine;
• to provide timely information that can be made available to potential recipients as well as to health-care providers so that they can weigh the risks and benefits of immunization; and
• to identify areas that require further epidemiologic investigation and research, or problems that require immediate investigation and intervention.

9.2 Surveillance systems

The cornerstone of surveillance activities is a voluntary system in which health-care providers (mainly public-health nurses and physicians) report events that they feel are temporally associated with an immunization to local, provincial, and territorial public-health authorities. These authorities, as well as a small proportion of vaccine manufacturers, forward all such reports to the Division of Immunization, LCDC, for compilation on a national level. Particular VAAEs that have been cited in the literature are included on a special reporting form (with modifications) used by all provinces and territories. Other severe or unusual events are also solicited and reported if the health-care provider feels that they may have been due to the administration of a vaccine.

The Canadian Immunization Guide provides more information on the nature of adverse events occurring with specific immunizing agents⁽⁵⁾. A copy of the special reporting form can be found in current issues of Compendium of Pharmaceuticals and Specialties published by the Canadian Pharmaceutical Association. Reports which are received by LCDC are computerized after immediate scrutiny. The database includes epidemiologic and medical information on the reported events. To calculate VAAE reporting rates, the Division of Immunization obtains lot-specific data from vaccine manufacturers on the number of doses of their products which are distributed across the country. These "vaccine distribution" data are used as an approximation of the actual number of doses of aministered vaccine. However, because of varying reporting practices, differences in lot-specific VAAE rates require cautious interpretation. Although these denominator data are limited in reliability, they are very useful in generating signals that should be investigated. Further estimates of denominator data are obtained from studies of vaccine coverage.

The reporting of VAAEs by health-care providers is voluntary, except in Ontario which has specific mandatory reporting requirements. However, there is no evidence of a higher reporting rate with this approach. This is partly explained by the fact that immunization in this province is usually provided by physicians who have lower reporting rates than public-health nurses.

In addition to this spontaneous voluntary reporting system, Canada also has an active surveillance system, IMPACT, for serious VAAEs, vaccination failures, and selected infectious diseases. The system is operated through a contract with CPS. It involves a network of 11 pediatric centres across Canada which comprise over 2,000 pediatric beds with over 85,000 children admitted annually. This represents over 80% of all pediatric tertiary-care admissions in the country.

At each centre, a nurse monitor and clinical investigator actively look for cases based on a daily review of admission records; they search for diagnoses of illnesses that are potentially linked to immunization. They are helped by an informal network, which includes admitting-department staff, infection-control nurses, neurology-ward staff and physicians, infectious-diseases staff, and medical-records technicians.

Since IMPACT monitors are so visible in institutions, they have become a focal point for vaccination-related questions from other medical staff and therefore learn of any concerns about vaccine safety that may arise. The IMPACT monitoring program was designed to actively and more systematically detect the most serious VAAEs that are temporally linked to the administration of vaccines, such as cases of neurologic illness. In addition, any hospitalization that is discovered to have followed an immunization is also recorded. Under the more recent CPSP, practising pediatricians actively report selected disease conditions and may also contribute to VAAE monitoring. The first related target disease is AFP, but other more specific VAAEs may be added in the future (see the section on the "Canadian Paediatric Surveillance Program" for more details).

Several aspects of the post-marketing surveillance program also serve to enhance signal-generating capabilities. In 1994, an external and multidisciplinary advisory group, the Advisory Committee on Causality Assessment (ACCA), was formed to help evaluate all cases involving serious events and identify signals for in-depth investigation. The specific mandate of this group is to review, in a systematic fashion, all serious cases of VAAEs temporally associated with immunization that are reported either through IMPACT or through the spontaneous reporting system⁽²⁵⁾. In addition, provincial public-health authorities can refer selected cases for special review by ACCA, although most often such reports would automatically meet the selection criteria for ACCA review.

An assessment that a particular report was possibly due to the administration of a vaccine is not an indication that there should be a concern about immunization, but more often represents a rare but expected event that is known to occur with the vaccine. By reviewing a large number of cases in a careful manner, a potential signal (indicating a suspicion that a vaccine may be causing a particular event that previously has not been well described) can be identified.

ACCA meets twice a year. Approximately 50 cases are reviewed during each meeting. A specially designed causality assessment form, developed according to the principles of adverse drug reaction causality assessment, is used. The outcome of the assessment is described using WHO International Drug Monitoring Program criteria.

A number of signals have been generated and evaluated by the committee during these reviews. None has led to the need for immediate action but, in one case, an initial concern was later refuted when further cases were evaluated. Should an urgent issue arise, ACCA would be immediately involved through a conference call. The issues generated or reviewed by ACCA are outlined in the section below on trends in vaccine safety. ACCA actively collaborates with many other national and international agencies, as well as public-health authorities in the provinces and territories.

9.3 Trends in vaccine safety

The VAAE system receives between 4,000 and 5,000 case reports per year through the reporting network. The trend is increasing as provinces begin to promote VAAE reporting more heavily. This has been most noticeable in Alberta where VAAE reporting has been actively solicited from all public-health immunization clinics since 1980. As a result, the number of case reports involving vaccine products contained in the Canadian Drug Adverse Reaction Reporting Program database rose from an average of 50 to 60 per year, from 1965 until 1980, to an average approaching 2,000 by 1987. Although the upward trend levelled off, other provinces began to report more frequently when the program was implemented in 1987 as a separate reporting system apart from that reporting other drug products. By examining the distribution of case reports by reaction category, it can be seen that the majority are, in fact, minor reactions. The VAAE program solicits such events in order to monitor lots on the market. As expected, the number of local, minor reactions is a more sensitive indicator of some problems than is the number of rare, serious reactions. This was highlighted by discovering, retrospectively, a dramatic increase in parotitis with the Urabe strain of mumps virus in the late 1980s, which heralded an increase in the risk of viral meningitis from that product. Although meningitis rates were clearly higher with the Urabe strain than with the Jeryl-Lynn strain, the absolute numbers were not so high; however, the number of cases of parotitis were much more readily detected. The vaccine was quickly withdrawn from use in Canada following identification of the problem. Canada was the first country in the world at that time to detect such a problem with this specific vaccine and most other countries later followed Canada's action. Still, at the time, the Canadian reporting system was far from being as effective as it is today.

Table 3 shows the distribution of reported VAAEs by reaction category.  Table 4 shows the distribution of case reports by vaccine. As expected, the majority of reports pertain to VAAEs which occur in temporal coincidence with vaccines that are given most frequently, according to the immunization schedule.

Since 1991, IMPACT has reported a total of 274 cases to the VAAE system. These represent the results of systematic case finding; therefore, it is not surprising that just under 20% involved seizures; neurologic events are a major focus of surveillance. Table 5 shows the distribution of reports by year and Table 6 shows their distribution by nature of the VAAEs reported.

Table 3 Distribution of Overall Reported VAAEs, by Reaction Category

Table 4 Distribution of Overall Reported VAAEs, by Vaccine Administered

Table 5 Cases Reported by IMPACT, by Year of Vaccine Administration, 1991-1995

Table 6 VAAEs Reported by IMPACT, by Nature of the Event

9.4 Discussion

The safety of vaccines marketed in Canada is monitored on a continuous basis. Each lot of vaccine is tested and released on an individual basis and, once in use, spontaneous reporting, active surveillance, and the systematic review and evaluation of serious concerns are all undertaken. Despite the usual limitations shared by all spontaneous reporting systems, namely that of underreporting and poor documentation of case reports, Canada's post-marketing surveillance system for vaccines has been a model for other countries. This is due to two major factors:

1. cooperation and collaboration with provincial public-health networks, leading to very high reporting rates; and
2. multiple active and passive components of surveillance systems, along with an expert advisory committee as mentioned above.

The improvement of surveillance activities has also been made possible by extraordinary VAAE funding allocated by Treasury Board specifically for post-marketing surveillance. This money is scheduled to sunset in 1998. It is hoped that funding will be continued (although there is no indication that it will) so that Canadians will continue to benefit from one of the best vaccine-surveillance systems in the world.

Reports of adverse events are solicited for any event that is felt to be temporally related to the administration of an immunization, but not necessarily absolutely causally related. In addition, reports are encouraged for "minor" events in order to monitor lot-by-lot performance. These two factors account for the fact that the 4,000 to 5,000 case reports received every year represent a higher reporting rate than that in most other countries that have well-developed post-marketing surveillance programs. This rate is a credit to the reporting system and not a marker for poor vaccines. Over 12 million doses of vaccine are distributed every year and very few concerns arise despite intense searching. Until diseases are eradicated, immunization remains our best defence.

[Previous] [Table of Contents] [Next]