Role of Background Incidence in Signal Detection;Observed/ Expected Analysis of Classical Thromboembolism with COVID 19 Vaccines in The Netherlands: An International Journal of Medical Toxicology and Drug Experience

ABSTRACT

Introduction: Spontaneous reporting (SR) is a key method for monitoring the safety of COVID-19 vaccines used in the pandemic in 2021. SR led to early warnings of vaccine-induced-thrombosis-withthrombocytopenia (VITT) with the vector vaccines of AstraZeneca and Janssen [1-3]. In addition, an increase of reports of other thromboembolic events with both vector and mRNA vaccines was seen. For now, venous thromboembolism (VTE) is labelled for Janssen and cerebral-venous-thrombosis (CVST) for the AstraZeneca vaccines [4,5]. Because a large population is vaccinated in a short period of time and the fact that venous (VTE) and arterial thrombosis (ATE) are quite common conditions, the number of reports should be compared with background incidence rates as part of signal detection. Objective: We evaluated the use of Standardised Morbidity Rates (SMR) to compare spontaneously reported cases of thromboembolism with COVID-19 vaccines with stratified Dutch background incidence rates from 2019 in the vaccine exposed population. Methods: SMRs are calculated by dividing the reported as observed (O) cases by the expected (E) number within risk windows of 14/28 days following each dose and vaccine. SMR >1 indicates that more cases were reported than expected [6]. Results: Until 9 December 2021, 2080 reports were received concerning at least one thromboembolic event with 24 million administered COVID-19 vaccines. These concerned 1000 of VTE, 956 of ATE, 19 of CVST and 134 reports of miscellaneous forms of thrombosis. Cases of VITT (TTS) were excluded. In the table the main results are summarized and highlighted. For AstraZeneca, SMRs reach or exceed 1 for VTE, ATE and CVST, especially in people aged < 60 years. With the second dose of mRNA vaccines more cases of the rare condition CVST were observed in men, although absolute numbers are low;n = 3 for men with 2nd dose of Pfizer (14 days) and n = 1 for men with 2nd dose of Moderna (28 days). Conclusion: SMRs can be used when the exposed population is defined, stratified background incidence rates are available and the reporting rate is substantial. The rate of underreporting is unknown, differs in population subgroups, and is influenced by media attention. Therefore, SMRs near 1 are also considered to be 'high'. Our data cannot confirm nor exclude a potential signal of other types of thromboembolism with COVID-19 vaccines. But the potential signal initiated further epidemiological research to confirm and quantify a potential increased risk of thromboembolism with COVID-19 vaccines.