A modified self‐controlled case series method for event‐dependent exposures and high event‐related mortality, with application to COVID‐19 vaccine safety

Abstract

We propose a modified self-controlled case series (SCCS) method to handle both event-dependent exposures and high event-related mortality. This development is motivated by an epidemiological study undertaken in France to quantify potential risks of cardiovascular events associated with COVID-19 vaccines.Event-dependence of vaccinations, and high event-related mortality, are likely to arise in other SCCS studies of COVID-19 vaccine safety. Using this case study and simulations to broaden its scope, we explore these features and the biases they may generate, implement the modified SCCS model, illustrate some of the properties of this model, and develop a new test for presence of a dose effect. The model we propose has wider application, notably when the event of interest is death.

Risk of death following SARS-CoV-2 infection or COVID-19 vaccination in young people in England: a self-controlled case series study

“…Among 12-29 year olds, the mortality rate was very low and unlikely to bias in the estimates of the relative incidences. Another limitation is that we could not control for SARS-CoV-2 infection [16], despite the infection being related to cardiovascular outcomes [8].…”

Thromboembolic events and hemorrhagic stroke after mRNA (BNT162b2) and inactivated (CoronaVac) covid-19 vaccination: A self-controlled case series study

“…The modified SCCS has been used and was recommended in a published covid-19 vaccine safety study. 15 , 44 The modified SCCS is based on a counterfactual that assumes exposures can never occur after an event, i.e. it will be able to take into account scenario of event-dependent exposure.…”

“…To account for deaths following the event, it is recommended to omit the date of death and set the planned observation end date for all subjects in the modified SCCS model. 44 Patients who were tested positive for SARS-CoV-2 were removed from the SCCS analysis to minimize bias from misclassification. Monthly seasonal effect was adjusted.…”

“…Even if it is unlikely to develop thromboembolic events or hemorrhagic stroke on the same day as vaccination or vice versa, we specified Day 0 from the risk period to check the results’ robustness. 44 , 45 5) Since the recommended vaccination schedule for BNT162b2 was 21 days (shorter than the 28-day risk period), the risk periods were stratified into 0-13 days, 14 to 20 days, and 21 to 28 days to examine the differences.…”

Characteristics associated with the residual risk of severe COVID-19 after a complete vaccination schedule: A cohort study of 28 million people in France

“…We conducted a cohort study using the National Health Data System (SNDS), which covers the entire French population, i.e., 67 million inhabitants, and which has been extensively used in France to conduct pharmacoepidemiology studies. 16 , 17 , 18 , 19 …”