Causality assessment of adverse events reported to the Vaccine Adverse Event Reporting System (VAERS)

Johns Hopkins Bloomberg School of Public Health, United States. Electronic address: .
Vaccine (Impact Factor: 3.62). 10/2012; 30(50). DOI: 10.1016/j.vaccine.2012.09.074
Source: PubMed


Adverse events following immunization (AEFI) reported to the national Vaccine Adverse Event Reporting System (VAERS) represent true causally related events, as well as events that are temporally, but not necessarily causally related to vaccine. OBJECTIVE: We sought to determine if the causal relationships between the vaccine and the AEFI reported to VAERS could be assessed through expert review. DESIGN: A stratified random sample of 100 VAERS reports received in 2004 contained 13 fatal cases, 19 cases with non-fatal disabilities, 39 other serious non-fatal cases and 29 non-serious cases. Experts knowledgeable about vaccines and clinical outcomes, reviewed each VAERS report and available medical records. MAIN OUTCOME MEASURES: Modified World Health Organization criteria were used to classify the causal relationship between vaccines and AEFI as definite, probable, possible, unlikely or unrelated. Five independent reviewers evaluated each report. If they did not reach a majority agreement on causality after initial review, the report was discussed on a telephone conference to achieve agreement. RESULTS: 108 AEFIs were identified in the selected 100 VAERS reports. After initial review majority agreement was achieved for 83% of the AEFI and 17% required further discussion. In the end, only 3 (3%) of the AEFI were classified as definitely causally related to vaccine received. Of the remaining AEFI 22 (20%) were classified as probably and 22 (20%) were classified as possibly related to vaccine received; a majority (53%) were classified as either unlikely or unrelated to a vaccine received. CONCLUSIONS: Using VAERS reports and additional documentation, causality could be assessed by expert review in the majority of VAERS reports. Assessment of VAERS reports identified that causality was thought to be probable or definite in less than one quarter of reports, and these were dominated by local reactions, allergic reactions, or symptoms known to be associated with the vaccine administered.

Download full-text


Available from: Christine Casey, Sep 18, 2015
61 Reads
  • [Show abstract] [Hide abstract]
    ABSTRACT: Comprehensive surveillance of adverse events following immunization (AEFI) is required to detect potential serious adverse events that may not be identified in prelicensure vaccine trials. Surveillance systems have traditionally been passive, relying upon spontaneous reporting, but increasingly active surveillance and supplemental strategies are being incorporated into vaccine safety programs. These include active screening for targeted conditions of interest (e.g., hospitalization), monitoring of new data sources and real-time methodologies to detect changes in vaccine safety data in these sources. The role of improved causality assessment in AEFI surveillance is discussed, with its important role in determining whether a temporal association may have occurred by chance alone. Strong local vaccine safety networks are required to support national immunization programs, with recent progress in developing a framework for low- and middle-income countries. Global collaboration is increasingly required to address challenges in active AEFI surveillance, particularly for rare serious adverse events.
    Expert Review of Vaccines 12/2013; 13(2). DOI:10.1586/14760584.2014.866895 · 4.21 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: To evaluate epidemiological features of post vaccine acute disseminated encephalomyelitis (ADEM) by considering data from different pharmacovigilance surveillance systems. The Vaccine Adverse Event Reporting System (VAERS) database and the EudraVigilance post-authorisation module (EVPM) were searched to identify post vaccine ADEM cases. Epidemiological features including sex and related vaccines were analysed. We retrieved 205 and 236 ADEM cases from the EVPM and VAERS databases, respectively, of which 404 were considered for epidemiological analysis following verification and causality assessment. Half of the patients had less than 18 years and with a slight male predominance. The time interval from vaccination to ADEM onset was 2-30 days in 61% of the cases. Vaccine against seasonal flu and human papilloma virus vaccine were those most frequently associated with ADEM, accounting for almost 30% of the total cases. Mean number of reports per year between 2005 and 2012 in VAERS database was 40±21.7, decreasing after 2010 mainly because of a reduction of reports associated with human papilloma virus and Diphtheria, Pertussis, Tetanus, Polio and Haemophilus Influentiae type B vaccines. This study has a high epidemiological power as it is based on information on adverse events having occurred in over one billion people. It suffers from lack of rigorous case verification due to the weakness intrinsic to the surveillance databases used. At variance with previous reports on a prevalence of ADEM in childhood we demonstrate that it may occur at any age when post vaccination. This study also shows that the diminishing trend in post vaccine ADEM reporting related to Diphtheria, Pertussis, Tetanus, Polio and Haemophilus Influentiae type B and human papilloma virus vaccine groups is most likely due to a decline in vaccine coverage indicative of a reduced attention to this adverse drug reaction.
    PLoS ONE 12/2013; 8(10):e77766. DOI:10.1371/journal.pone.0077766 · 3.23 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Masking is a statistical issue by which true signals of disproportionate reporting are hidden by the presence of other products in the database. Masking is currently not perfectly understood. There is no algorithm to identify the potential masking drugs to remove them for subsequent analyses of disproportionality. The primary objective of our study is to develop a mathematical framework for assessing the extent and impact of the masking effect of measures of disproportionality. We have developed a masking ratio that quantifies the masking effect of a given product. We have conducted a simulation study to validate our algorithm. The masking ratio is a measure of the strength of the masking effect whether the analysis is performed at the report or event level, and the manner in which reports are allocated to cells in the contingency table significantly impact the masking mechanisms. The reports containing both the product of interest and the masking product need to be handled appropriately. The proposed algorithm can use simplified masking provided that underlying assumptions (in particular the size of the database) are verified. For any event, the strongest masking effect is associated with the drug with the highest number of records (reports excluding the product of interest). Our study provides significant insights with practical implications for real-world pharmacovigilance that are supported by both real and simulated data. The public health impact of masking is still unknown. Copyright © 2013 John Wiley & Sons, Ltd.
    Pharmacoepidemiology and Drug Safety 02/2014; 23(2). DOI:10.1002/pds.3530 · 2.94 Impact Factor
Show more