How the adverse effect counting window affected vaccine safety calculations in randomised trials of COVID‐19 vaccines

Abstract

Several articles published recently in the Journal of Evaluation in ClinicalPractice raise important questions about how the efficacy and safety of COVID-19 vaccines have been and are being measured. First, Fung, Jones, and Doshi explained that statistical biases, particularly the case counting window bias, can greatly exaggerate the perceived effectiveness of COVID-19 vaccines in observational studies. 1 A follow-up article by myself explained that this effect may be understated, due to the often accompanied definitional bias, and also that these biases could likewise exaggerate the perceived safety of COVID-19 vaccines in observational studies. 2 Doshi and Fung then added further to the discussion by finding issues with case counting windows in COVID-19 vaccine clinical trials, relating to vaccine effectiveness. 3 In this final article, I shall provide an example justifying my prior concern about counting windows in observational studies concerning safety, and further explain that there are also concerns around adverse effect counting windows in COVID-19 vaccine clinical trials, which may have exaggerated claims about vaccine safety.

Anti-science case study: COVID-19 vaccines’ effectiveness and safety exaggerated

“…This has been decisively argued in an unofficial series in the Journal of Evaluation in Clinical Practice , involving BMJ editor Peter Doshi, with the authors collectively finding: infections being overlooked in the ‘partially vaccinated’; such infections being ascribed to unvaccinated groups; numerous suspected infections overlooked as ‘unconfirmed’ (divided roughly equally between vaccinated and unvaccinated); adverse effects being overlooked in the ‘partially vaccinated’; adverse effect reporting reliant on solicited reports; longer-term adverse effects overlooked; numerous trial participants lost to follow-up; long-term impacts impossible to discern due to unblinding; and financial conflicts of interest [ [2] , [3] , [4] , [5] ]. Also discussed were vaccine-related myocarditis, with recent research on this one adverse effect alone showing incident rates far exceeding UK government estimates on the numbers needed to vaccinate in various groups to prevent a severe COVID-19 hospitalisation; and some of the evidence for perceived negative effectiveness, where the vaccines are associated with increased COVID-19 infections, hospitalisations, and even deaths.…”