Canadian Immunization Guide
Seventh Edition - 2006

[Previous] [Table of Contents] [Next]

Part 2
Vaccine Safety and Adverse Events Following Immunization

Vaccine Safety

• Introduction
• Vaccine evaluation and regulation
• Vaccine safety surveillance and assessment in Canada
• Evidence pertaining to vaccine safety: where to find it, how to interpret it
• Vaccine safety controversies
• Suggested reading and resources

Introduction

Vaccine safety is of the highest importance and concern for all vaccine stakeholders. As vaccine-preventable infections have decreased, the spotlight of public and mass media concern has shifted to vaccine safety. Since vaccines are usually given to healthy people, especially children, tolerance for adverse events is low. Perceived vaccine safety risks get as much attention as real ones and can be difficult to dispel despite credible scientific evidence. Loss of confidence threatens the continued success of immunization programs.

Health care providers have essential and pivotal roles to play in gaining and maintaining public confidence in the safety of vaccines These include providing evidence-based information on the benefits and risks of vaccines; helping clients and patients to interpret media and Internet vaccine safety messages; and identifying and reporting adverse events following immunization. Any single occurrence of an unusual event following immunization may be coincidental or caused by the vaccine. An accumulation of reports, sometimes as few as four or five, may signal a risk due to the vaccine. Thus, each and every report submitted by vaccine providers is important.

This new chapter has been added to the Canadian Immunization Guide for the following reasons:

• To highlight the critical importance of ongoing post-marketing vaccine safety surveillance by describing how vaccines are evaluated and regulated, and the scientific limitations of pre-marketing assessments
• To provide an overview of Canada's vaccine safety surveillance system with specific information not only on how to report adverse events but also on how such information is used to ensure that immunization programs, in Canada and internationally, remain as safe as possible
• To provide an overview of the type and quality of evidence available to inform vaccine safety
• To summarize the current status of key vaccine safety issues
• To provide a list of key resources and references on vaccine safety

Vaccine evaluation and regulation

The development of a new vaccine starts with pre-clinical laboratory testing to ensure that vaccine candidates produce the immune response needed to prevent disease and have no toxicities that would prevent their use in people. Human studies then proceed through several phases involving progressively more subjects. Table 1 describes the phases of vaccine evaluation in terms of how many subjects are studied and what is learned.

Depending on the specific vaccine, it may take years to decades to gather the scientific immunogenicity, safety and efficacy data needed to obtain authorization for marketing. However, pre-marketing vaccine studies do not have sufficient numbers of subjects to detect rare or very rare adverse events, the frequency of which is shown in Table 2. Furthermore, all potential target populations have not been fully studied prior to marketing approval. Thus, ongoing post-marketing studies of vaccine safety and effectiveness are essential, not only to gather data on new vaccines but also to monitor existing vaccines for any change in the frequency of known events that might occur if newly released vaccine lots do not perform as expected. Post-marketing data help to refine the benefit-risk assessment of a given vaccine as well as add to key information regarding contraindications, warnings and concomitant use with other vaccines.

The Biologics and Genetic Therapies Directorate (BGTD) of Health Canada is the regulatory authority responsible for establishing the safety, efficacy and quality of all biologics for human use, including vaccines (http://www.hc-sc.gc.ca/dhp-mps/brgtherap/index_e.html). BGTD reviews the clinical and chemistry/manufacturing information of vaccine submissions, and conducts on-site evaluations of manufacturing facilities and laboratory analysis of vaccines. The clinical information includes data from clinical trials, and post-marketing safety and efficacy information. BGTD will grant a marketing authorization for the vaccine if the evidence to support the safety, efficacy and quality of the vaccine is considered adequate and sufficient.

Table 1. Stages of Clinical Vaccine Assessment and Detectable Adverse Events

Subsequently, if there are any changes in chemistry/manufacturing procedures or new clinical information pertaining to approved products, vaccine manufacturers must submit information for BGTD approval. The nature, extent and importance of the changes affecting the approved vaccine will determine whether additional clinical testing is required and whether the changes must be communicated to vaccine users through updated labeling and revisions of the product monograph.

A product monograph is the official labeling document for a vaccine and must be approved by Health Canada when the vaccine is first authorized for marketing and each time the information is updated. It is a factual, scientific document that, devoid of promotional material, describes the properties, claims, indications, conditions and any other information required for optimal, safe and effective use of the vaccine. It must accurately reflect important information and results from clinical trials and other relevant information submitted to Health Canada for evaluation. The product monograph consists of three parts:

• Health professional information: contains prescribing information, including indications, contraindications, warnings and precautions, adverse reactions, interactions, dosage, administration and storage instructions.
• Scientific information: contains a summary description of the pre-clinical, toxicological and clinical testing of the vaccine and any other pertinent scientific information with relevant references.
• Consumer information: contains an abbreviated summary, written in simplified language, to communicate essential information to the vaccine or product recipient/user.

Table 2. Description of Terms Used for the Frequency of Adverse Events Following Immunization

Product monographs may contain proprietary information and thus are not generally made available in their entirety, although many manufacturers now publish them on their Web sites. The package insert in marketed vaccines is an abbreviated form of the product monograph and usually contains the same prescribing information as is found in part 1 of the full monograph. Information on a specific vaccine found in other publications, including the Compendium of Pharmaceuticals and Specialties (CPS), is not controlled by Health Canada. The vaccine manufacturer may choose to include all, selected or modified parts of the information from the product monograph.

To further establish ongoing quality, safety and efficacy, all vaccines are released on a lot-by-lot basis. For each lot, BGTD reviews production protocols submitted by the manufacturer and performs selective confirmatory testing as appropriate to each vaccine.

Vaccine safety surveillance and assessment in Canada

In 2005 the name of Canada's vaccine safety surveillance system was changed from the Vaccine Associated Adverse Event Surveillance System (VAAESS) to the Canadian Adverse Event Following Immunization Surveillance System (CAEFISS). This change harmonizes Canadian terminology with what is used by the World Health Organization (WHO) and many other countries. Furthermore, it describes the nature of such events more accurately, in that adverse events do indeed follow immunization, but the temporal association is not proof that the event was caused by the vaccine.

In Canada a standard report form is available on the Internet (http://www.phac-aspc.gc.ca/im/aefi-form_e.html ), through public health units and in the CPS. This form includes check boxes to facilitate the reporting of selected adverse events of special interest (e.g., anaphylaxis, injection site reactions, neurologic events) but also should be used to report all other severe, unusual or unexpected adverse events that are of concern to the vaccine provider, health care provider, vaccine recipient or his/her parent(s)/ caregiver(s). Vaccinees and/or their parents/caregivers should therefore be advised to notify their health care provider about any adverse event of concern. Information on the report form facilitates monitoring and follow-up of adverse events at the local/provincial level, and causality assessment and signal detection at the national level. Confidentiality and privacy of health information are maintained throughout.

Surveillance

• Passive surveillance: this encompasses all spontaneous adverse event reporting. Health care providers complete reports and submit them to their local health unit or Medical Officer of Health. From there, reports are sent to the central provincial/territorial health department, which in turn forwards them to the Immunization and Respiratory Infections Division within the Centre for Infectious Disease Prevention and Control at the Public Health Agency of Canada (PHAC). This federal office is responsible for maintaining a national database of all reported adverse events. The database also includes reports from vaccine manufacturers, which are required by law to submit all adverse event reports to PHAC and, if serious, to do so within 15 calendar days of receipt (http://www.phac-aspc.gc.ca/publicat/ccdr-rmtc/00vol26/26s1/26s1e_e.html). In some jurisdictions, reports related to vaccines that are not publicly funded are submitted by vaccine providers directly to PHAC. If in doubt it is best to check with the provincial/territorial public health department as to where the report should be sent.
• Active surveillance: for severe adverse events following childhood immunizations this type of surveillance has been conducted in Canada since 1991 by the Immunization Monitoring Program ACTive (IMPACT). This is a pediatric, hospital-based network funded by PHAC and administered by the Canadian Paediatric Society. The 12 IMPACT hospitals encompass approximately 90% of tertiary care pediatric beds in Canada. Details on the network and lists of relevant publications can be found at http://www.cps.ca/English/Surveillance/IMPACT/IMPACT.htm. All serious adverse events detected by IMPACT are to be reported to the vaccinee's home provincial/territorial public health authorities, as well as to PHAC.
• Ad hoc studies: additional surveillance, as well as epidemiologic or clinical studies, may be undertaken by public health or academic investigators to further characterize adverse events of concern, assess whether or not there is a causal link between the vaccine and a given adverse event or learn about risk factors that increase the likelihood that an adverse event will occur. Examples of such studies include those done following the recognition of oculorespiratory syndrome (ORS) following influenza vaccination.

Causality assessment

Special review of serious and unusual adverse events (life-threatening, such as anaphylaxis or those associated with 3 or more days of hospitalization, congenital abnormality, residual damage or death) is conducted by the Advisory Committee on Causality Assessment (ACCA), which comprises pediatricians, immunologists, epidemiologists and other experts. In addition, any unusual events or cluster of events may be reviewed by ACCA. The primary mandate of ACCA is to evaluate the degree to which such events are linked to the implicated vaccine (for more information see http://www.phac-aspc.gc.ca/im/vs-sv/acca_e.html). The process of causality assessment requires sufficient case detail to be sure the adverse event diagnosis is accurate and to judge the potential contribution of underlying disease, intercurrent illness or concomitant medication(s). Since details are often missing in the submitted report it is usually necessary to contact the original reporter for additional information before ACCA can review a case. Plausible biologic mechanisms, as well as the availability and strength of existing scientific evidence to support or reject a causal association between the vaccine and a given adverse event, are all taken into consideration. The findings of ACCA are communicated back to the provinces/territories from which the report originated.

Global partners in vaccine safety

Canada actively participates in several international endeavours to monitor and improve vaccine safety on a global scale. Adverse event reports are forwarded to the WHO Uppsala Monitoring Centre for entry into a global pharmacovigilance database (http://www.who-umc.org/DynPage.aspx?id=13140&mn=1514). These data are regularly scanned to identify any safety signals of potential concern. Canada is an active participant in the Brighton collaboration (http://www.brightoncollaboration.org), which seeks to standardize and harmonize adverse event definitions for use in all phases of vaccine testing, as outlined in Table 1. Canada is represented on the WHO Global Advisory Committee on Vaccine Safety (http://www.who.int/vaccine_safety/en/) and also participates in ad hoc consultations and committees set up by the WHO to review specific issues in vaccine safety. Canada also cooperates with the Council for International Organizations of Medical Sciences (CIOMS, http://www.cioms.ch/), which is an international, non-governmental, non-profit organization established jointly by WHO and UNESCO in 1949 to facilitate and promote international activities in the field of biomedical sciences, including making recommendations on the assessment and monitoring of adverse reactions.

Evidence pertaining to vaccine safety: where to find it, how to interpret it

Temporal associations

Since vaccines are usually given to healthy people, any event that follows soon after immunization may be perceived as being due to the vaccine. This is particularly true for events with no proven cause, such as autism, most encephalopathies and multiple sclerosis. Multiple immunizations are given during early childhood because that is the period of greatest human vulnerability to vaccine-preventable morbidity and mortality. However, it is also a critical period of growth and development during which damage due to genetic, in utero and/or other post-natal influences may first become apparent. Consideration should always be given to the possibility of an association between the vaccine and an adverse event. However, other possibilities must also be considered. These include infections and concomitant medications, as well as diseases due to genetic, environmental or other factors. Adverse events due to these other causes may simply occur by chance after the administration of a vaccine.

Chance associations illustrate the greatest vulnerability of universal immunization programs. If a vaccine truly causes a given event, even if rare, the association can be proven by a well-designed study with sufficient subjects. In contrast, the absence of association or "zero risk" cannot be proven by epidemiologic methods. Even if no association is repeatedly shown in a number of studies, it is always possible to theorize that an association might be found in another group of individuals who have not been studied. It is not possible to demonstrate that there is a 100% certainty that no person has ever had the adverse event of interest. An element of doubt will always remain, although it can be stated that the risk is very close to zero.

Clinical trials have repeatedly shown that placebo recipients experience adverse events, which clearly cannot be due to the vaccine. In a randomized placebo-controlled trial of varicella vaccine among healthy children aged 1 to 14 years, the vaccinees (n = 491) and placebo recipients (n = 465) had a similar frequency of irritability (24% and 20%, respectively), tiredness (20%, 22%), headache (15%, 16%), cough (45%, 48%), common cold (63%, 65%), poor sleep (12%, 13%) and loss of appetite (11%, 13%) during the 8-week period after immunization. Rigorous trials such as this are very helpful because they allow the assessment of the degree to which adverse events are attributable to the vaccine as opposed to other factors.

Vaccine attributable risk

This is defined as the difference between the frequency of adverse events in otherwise comparable vaccinated and unvaccinated individuals. Figure 1 illustrates that not all health problems noted after immunization are caused by vaccine. In a population of immunized children, the number of illnesses or clinical symptoms compatible with an adverse event increased in the week after hepatitis B immunization but returned to pre-vaccination levels thereafter. The vaccine can be implicated only for this "excess" of illness (or attributable risk [AR]).

As another example, in a Finnish study of cross-over design, each twin of 581 pairs was given either measles, mumps and rubella (MMR) vaccine or placebo in a blinded fashion, and 3 weeks later was administered the other substance. Adverse events were monitored for 21 days after immunization. Table 3 clearly shows that some children in the placebo group experienced fever throughout the follow-up period, but the only significant differences (AR) between placebo and MMR groups occurred from days 7 to 12.

Table 3. Percentage of Children with Fever after MMR Immunization or Placebo Injection in 581 Twin Pairs*

Randomized, placebo-controlled trials

These trials provide the most reliable and valid evidence pertaining to vaccine safety. Unfortunately such trials are not done for all vaccines nor are they usually large enough to detect rare adverse events.

Population-based epidemiologic studies

Such studies use cohort (i.e., they compare the adverse event rate in immunized versus non-immunized populations) or case-control methodologies (i.e., they compare the proportion of cases with an adverse event and controls without an adverse event who were exposed to vaccine) to test hypotheses regarding a causal association between a given vaccine and an adverse event. However, the validity, generalizability and utility of data from such studies are highly dependent on study design. Since exposure to vaccine is not random in the study populations, several sources of bias exist that may confound the results.

Ecologic studies

Ecologic studies take advantage of "natural experiments" to test hypotheses regarding vaccines and adverse events. For example, the occurrence of autism might be compared during two separate periods of time in a country that switched from thimerosal-containing to thimerosal-free vaccines. As another example, the prevalence of multiple sclerosis might be compared in a country that has never introduced hepatitis B vaccine to one that has been using the vaccine for decades. A major methodologic problem with such studies is the inability to control for multiple confounding factors that may not be equally distributed or applicable to the time periods or geographic areas being compared. For example, differences in diagnostic criteria, standards of health practice and/or health-seeking behaviour could confound the results in favour of or against the hypothesis.

Reports of single or multiple cases

These reports often represent the first evidence of a possible link between a vaccine and an adverse event. As discussed earlier, chance temporal associations between vaccine(s) and subsequent adverse events are relatively common occurrences given the frequency of disease that occurs in any given population. In certain rare instances, a well-documented case report can establish a causal relation such as death due to disseminated BCG or unrelenting measles infection following administration of BCG vaccine or measles vaccine, respectively, to a severely immunocompromised host. The vaccine strains are distinguishable from naturally circulating disease strains (commonly referred to as "wild type"). Thus recovery of the vaccine types from body tissue(s) in conjunction with histopathological changes consistent with severe infection is usually considered proof of causality. However, the vast majority of case reports represent unproven temporal associations that require confirmation using scientifically sound methodologies.

Spontaneously submitted reports to passive surveillance systems

Reports sent to CAEFISS or the Vaccine Adverse Event Reporting System (VAERS) in the United States provide the weakest evidence of a causal association between a vaccine and the reported adverse event. It is essential to understand that proving causality is not the intent of passive surveillance. Rather, such systems are put in place to identify signals of concern as early as possible. Subsequently, specific studies must be designed to test the hypothesis that the adverse event is truly caused by vaccine. Illustrative of this process is the recent US experience with rotavirus vaccine. Regular analysis of VAERS data revealed an increased frequency of reports of intussusception (the "signal") in infants following the introduction of live attenuated rotavirus vaccine. Subsequently, several case-control studies confirmed the hypothesis of a link between rotavirus vaccine and intussusception, and the vaccine was withdrawn from the market. In Canada oculorespiratory syndrome (ORS) following immunization with influenza vaccines used in 2000 was recognized through an increased frequency of reports to the passive surveillance system. Following those observations several studies were done to characterize the causes and determinants of ORS, and modifications were made to one of the implicated vaccines before the 2001 influenza vaccine campaign.

There is currently public access to data from VAERS. Public access to CAEFISS data is planned, with announcements to be posted at the PHAC vaccine safety web site, http://www.phac-aspc.gc.ca/im/vs-sv/index.html, as soon as it is initiated. This is essential, given the need for transparency and openness regarding reported adverse events in order to maintain public confidence in immunization programs. However, the data are frequently misinterpreted and used to draw inappropriate conclusions regarding risks associated with immunization. Since many of the allegations presented on the Internet and in mass media result from inappropriate use of such data, health care professionals need to clearly understand both the purpose (as described above) and the limitations of passive surveillance systems:

• Lack of an appropriate control group
• Reporting bias. This stems from several factors that increase the rate of reporting other than a true increased frequency of a given adverse event. Examples include the following:
  
  
  • Major media focus on allegations such as a link between autism and MMR vaccine
  • Markedly increased frequency of immunization such as occurs during mass vaccine campaigns for outbreaks of infection due to Neisseria meningitidis
  • Enhanced awareness following recognition of clusters of specific adverse events such as ORS
• Lack of use of and/or adherence to standard case definitions
• Incomplete detail to support a given diagnosis and/or to consider other possible causes, such as intercurrent infection or concomitant medication
• Underreporting
• Iinability to determine the frequency of association since the total number of persons immunized is unknown:
  
  
  • The Internet often contains allegations of "hot lots" based on data from VAERS, which publishes the lot numbers associated with adverse event reports. However, the number of actual vaccine doses distributed for a given lot can vary from thousands to millions. If the lot size is unknown it is impossible to determine the lot-specific incidence of a given adverse event.

Expert-based reviews of vaccine safety issues

Vaccine safety is an issue of global concern. Although there are variations in vaccine products used in different countries, the similarities in terms of immunogen are such that much can be learned from expert reviews of specific issues, including examination of published and unpublished data. Currently, there are three sources of such reviews: the WHO, the Institute of Medicine and the Cochrane Collaboration.

World Health Organization: Details of the many WHO initiatives involving vaccine safety are available on the Internet (http://www.who.int/immunization_safety/en/). Two initiatives deserve further mention here as sources of reliable information on vaccine safety issues. In 1999 the Global Advisory Committee on Vaccine Safety (GACVS) was established to provide prompt, scientific evidence-based responses to safety issues of global concern. The expert committee meets every June and December, and soon afterwards posts reports of its deliberations on the WHO Web site (http://www.who.int/vaccine_safety/en/) and publishes these in the Weekly Epidemiological Record (http://www.who.int/wer/en/). The Web site also has a "topics" page that not only summarizes committee conclusions and recommendations but also provides links to other key publications or information on the specific issue. The other initiative, Vaccine Safety Net (http://www.who.int/immunization_safety/safety_quality/vaccine_safety_websites/en/), has been developed by GACVS to promote and identify Web sites on vaccine safety that adhere to good information practices.

Institute of Medicine (IOM): Formed in 1970 by the U.S. National Academy of Sciences (NAS), the IOM functions as an independent, expert professional body that examines issues of relevance to the health of the public (http://www.iom.edu). From 1977 through 1994 the IOM committees reviewed childhood vaccines and other vaccine safety issues. In 2001 a new IOM Immunization Safety Review Committee was assembled and included 13 individuals with broad expertise. To avoid real or perceived conflict of interest, an absolute criterion for membership was lack of any association with vaccine manufacturers or their parent organizations and no prior function as a legal expert witness. From 2001 through 2004 the committee reviewed and published its findings on eight specific vaccine safety issues (http://www.iom.edu/; a search on <immunization safety> will lead to all activities since 2001). For each issue studied, the Committee reviewed all pertinent theoretical, experimental, clinical and epidemiologic evidence and heard presentations from the public and health professionals. The Committee started from a neutral position, with no prior assumption regarding a positive or negative connection between the vaccine and the issue at hand. The scientific evidence was then reviewed, and biologic mechanisms for a possible causal association were carefully considered. Prior to publication, each report was reviewed by an independent expert panel, chosen by the NAS and IOM but anonymous to the committee. Reviewer's comments are given due consideration, but ultimately the final published report represents the consensus of the IOM safety panel alone. To view reports online and/or purchase copies see the National Academies Press site, http://lab.nap.edu, and search on vaccine or immunization safety.

The Cochrane Collaboration (www.cochrane.org) also conducts systematic reviews of vaccines, which may include information on vaccine safety. Since reviews are limited to randomized controlled trials, information regarding rare adverse events is unlikely to be covered.

Vaccine safety data in the Canadian Immunization Guide

In each chapter of this Guide, pre-licensure and post-marketing evidence-based safety data are presented for specific vaccines, as appropriate. At the time of publication of the Guide post-marketing surveillance of reports submitted to the CAEFISS has demonstrated continued vaccine safety and no unexpected serious adverse events. Detailed summaries of Canadian safety surveillance data for all reports by year, as well as for subgroups by vaccine and specific adverse event, will be published periodically on the Internet, in the Canada Communicable Disease Report and in peer reviewed publications as appropriate to the content. An updated list of published materials can be found at the PHAC Vaccine Safety Web site (http://www.phac-aspc.gc.ca/im/vs-sv/index.html, see "Safety data and publications").

Vaccine safety controversies

Space does not permit a detailed discussion of past or current controversies. Table 4 summarizes the conclusions of the IOM safety panel on several recent vaccine safety issues. Topical information on new, as well as past, controversies can be found at the PHAC's Vaccine Safety Web site. Additionally, in the suggested reading and resources given later, Web addresses are provided for the IOM's detailed reports as well as for meeting reports from the GACVS. Also, see the WHO Vaccine Safety Net Web site for a list of sites whose content on vaccine safety has been judged to meet the necessary criteria for credibility, content, accessibility and design.

Table 4. Events Judged Not To Be Linked to Vaccines*

Suggested reading and resources

Final report: National Immunization Strategy, 2003: http://www.phac-aspc.gc.ca/publicat/nat_immunization_03/index.html

WHO Global Advisory Committee on Vaccine Safety

Folb PI, Bernatowska E, Chen R et al. A global perspective on vaccine safety and public health: the Global Advisory Committee on Vaccine Safety. American Journal of Public Health 2004;94(11):1926-1931. This is an overview of the committee's role and activities. Downloadable pdf file at http://www.who.int/vaccine safety/about/en/vaccine.pdf.

GACVS. Bi-annual meeting reports, including summary lists of the topics discussed and full text details, can be found at http://www.who.int/vaccine_safety/reports/en/

WHO Vaccine Safety Net

http://www.who.int/immunization_safety/safety_quality/vaccine_safety_websites/en/

Institute of Medicine

To read and/or purchase reports on vaccine and immunization safety issues, see the National Academies Press site, http://lab.nap.edu, and search on vaccine or immunization safety.

Guidelines for preparing core clinical safety information on drugs - report of the Council for International Organizations of Medical Sciences (CIOMS) Working Group III. Geneva: World Health Organization (WHO) , 1995. (Chapter 5, Good Safety Information Practice).

Canadian Coalition for Immunization Awareness and Promotion

Tips for assessing Web sites (usually for health professionals and the public) can be found at the Canadian Coalition for Immunization Awareness and Promotion

English version: http://www.immunize.cpha.ca/english/poster/intip_e.htm
French version: http://www.immunize.cpha.ca/francais/posterf/intip_f.htm

[Previous] [Table of Contents] [Next]