Estimation of background incidence rates of coagulation disorders and the association with COVID-19 vaccines in pregnant population: a multi-database study from 3 European countries

Study type

• Study topic:
  Disease /health condition

  Human medicinal product
• Study type:
  Non-interventional study
• Scope of the study:
  Disease epidemiology
• Data collection methods:
  Secondary use of data

Non-interventional study

• Non-interventional study design:
  Other
• Non-interventional study design, other:
  Self-controlled case series and background incidence rates

Study drug and medical condition

• Medicinal product name, other:
  EMA-approved COVID-19 vaccines:; -BioNTech/Pfizer; -Moderna; -AstraZeneca; -J&J
• Anatomical Therapeutic Chemical (ATC) code:
  (J07BN) Covid-19 vaccines

  Covid-19 vaccines
• Medical condition to be studied:
  Embolism venous

  Cerebral venous sinus thrombosis

  Arterial thrombosis

  Thrombosis with thrombocytopenia syndrome

  Deep vein thrombosis

  Pulmonary embolism

  Haemorrhagic stroke

  Ischaemic stroke

  Disseminated intravascular coagulation

  Microangiopathy

Population studied

• Short description of the study population:
  The source population comprises all persons entered in the data sources from 01-Jan-2019, through the most recent data availability. The study population consists of pregnant individuals, identified using the IMI-ConcePTION pregnancy algorithm which is used for background rate analysis. From this study population, we will define the study population for Self-controlled case series (SSCS) - Eligible for inclusion in the SCCS analysis are all pregnant women (which has one or more pregnancies) who received at least one dose of a COVID-19 vaccine and were diagnosed with a coagulation disorder during the follow-up period.
• Special population of interest:
  Pregnant women

Study design details

• Study design:
  A background incidence rate will be applied before COVID-19 vaccination. Then, a self-controlled case series (SCCS) analysis will be conducted with all pregnant women with at least one COVID-19 vaccine dose (BioNTech/Pfizer, Moderna, AstraZeneca, J&J) during pregnancy.
• Main study objective:
  1. To estimate the background incidence rates of coagulation disorders—including thromboembolic and hemorrhagic events— during pregnancies and prior to vaccination with the four-EMA approved COVID-19 vaccines (BioNTech/Pfizer, Moderna, AstraZeneca, J&J). ; 2. To evaluate the association between the exposure to each of the four EMA-approved COVID-19 vaccines (BioNTech/Pfizer, Moderna, AstraZeneca, J&J) and the coagulation disorders* in pregnant women through a self-controlled case series (SCCS) design ;
• Setting:
  This study will use data from secondary electronic health record databases that are population-based. The following data sources were included: SIDIAP (access provided by the Fundació Institut Universitari per a la Recerca a l'Atenció Primària de Salut Jordi Gol i Gurina (IDIAP JGol)), VID (access provided by the Foundation for the Promotion of Health and Biomedical Research of Valencia Region; (FISABIO)); Danish National Registries (DHR) (access provided by Aarhus University) and Finnish registers (access provided by University of Eastern Finland).; Each data source will produce a study-specific data instance. A data instance is a subset of the data source that has been ETL’ed into the CDM at a certain point in time. This instance does not necessarily contain data from all databanks in the data source, but data required for one or more studies.; Pregnancies will be identified using the IMI-ConcePTION pregnancy algorithm.
• Outcomes:
  Coagulation disorders are the main outcome of this study. We defined coagulation disorders as a composite outcome of any one of the following:; 1. Venous thromboembolism (VTE); 2. Cerebral Venous Sinus Thrombosis (CVST); 3. Arterial Thrombosis; 4. Thrombosis with thrombocytopenia syndrome (TTS); 5. Deep vein thrombosis (DVT); 6. Pulmonary Embolism (PE); 7. Hemorrhagic stroke; 8. Ischemic stroke (IS); 9. Disseminated intravascular Coagulation (DIC); 10. Microangiopathy; These events will be captured from the EVENTS table using diagnosis code lists. A separate table of algorithms linked to diagnostic code list that will be submitted with the publication; Code lists to identify events have been created using the VAC4EU Code Mapper tool29, which maps concepts across medical vocabularies based on the Unified Medical Language System. The output of the Code Mapper is a CSV list presented in Annex 1.; For the sub-analysis, thromboembolic outcome will be composed by the following events: VTE, DVT, PE, CVST, IS, Arterial thrombosis, and TTS. ; For the sub-analysis of haemorrhagic disorders, we will combine the following events: DIC, Hemorrhagic stroke and Microangiopathy.;
• Data analysis plan:
  The background Incidence rates of outcomes among pregnant women with coagulation disorders will be calculated using total cases and person-time at risk. The incidence rates will be adjusted for the covariates (age, gestational age, prior Venous thromboembolism and anti-thrombotic agents) via Poisson regression. A 95% confidence intervals will be computed using exact methods.; A modified SCCS analysis using conditional Poisson regression will be conducted to estimate the relative incidence, stratified by data source and gestational trimester. A 14-day risk window, pre-exposure transition window of 14 days and two-day post-exposure transition windows are proposed for this study. Sensitivity analyses using 7-day interval will be applied. Country specific estimates will be pooled via random-effects meta-analysis.;