Impact of Individual Case Reports for COVID-19 Vaccines on the Ability to Detect Future Safety Signals with Other Medicinal Products

Background: As of February 2022, the WHO global database of reported potential side effects of medicinal products, VigiBase, contains 3.2 million individual case reports of adverse events following COVID-19 immunization. This is more than 10% of all reports in VigiBase to date and more than 50% of all reports submitted during 2021. The reports on COVID-19 vaccines are a crucial asset in supporting their safe use. Meanwhile, this massive influx of reports increases overall relative reporting rates of some adverse events (AE), which may mask future safety signals for other medicinal products when disproportionality analysis is used to support and guide clinical review.

Objectives: To identify adverse events masked by reports on COVID-19 vaccines in VigiBase.

Methods: The approach for simple unmasking proposed by Juhlin et al in Pharmacoepidemiology and Drug Safety, 2013 was used to identify AEs for which the overall relative reporting rates increased by at least 50% when reports on COVID-19 vaccines were considered alongside reports on all other drugs and vaccines (data in VigiBase up to Dec 1, 2021). In this approach, statistical shrinkage is used to limit the impact of random variability and shift focus away from rare AEs. AEs in the System Organ Classes Investigations and Surgical and medical procedures were not considered.

Results: 132 AEs were masked by COVID-19 vaccine reports to the extent that their overall relative reporting rates in VigiBase increased by 50% or more. 21 of these are on the European Medicines Agency's list of Important Medical Events (IME), including: sweating fever (expected value up by 1300%), Bell's palsy (up 560%), vaccine breakthrough infection (up 680%), hemiparaesthesia (up 240%), myocarditis (up 140%), pericarditis (up 130%), COVID-19 pneumonia (up 120%), and postmenopausal haemorrhage (up 120%).

Conclusions: Inclusion of COVID-19 vaccine reports in disproportionality analysis for other medicinal products may hinder or delay the detection of statistical signals of many AEs, including IMEs that are of special interest in pharmacovigilance. An increase by 50% in the overall relative reporting rate of an AE increasing the expected count, from 10 to 15, would require 30 reports instead of 18 to detect a statistical signal. Strategies to mitigate this need to be developed and evaluated for example to exclude reports of COVID-19 vaccines from disproportionality analysis for other medicinal products or to apply a method for unmasking.