Safety of mRNA COVID-19 vaccines during pregnancy

Study type

• Study topic:
  Other
• Study topic, other:
  Vaccines
• Study type:
  Non-interventional study
• Scope of the study:
  Disease epidemiology

  Effectiveness study (incl. comparative)
• Data collection methods:
  Secondary use of data

Non-interventional study

• Non-interventional study design:
  Cohort

Study drug and medical condition

• Name of medicine, other:
  mRNA-BNT162b2, mRNA-1273
• Medical condition to be studied:
  Deep vein thrombosis

  Pulmonary embolism

  Myocardial infarction

  Ischaemic stroke

  Bell's palsy

  Encephalitis

  Guillain-Barre syndrome

  Myelitis transverse

  Haemorrhagic stroke

  Myocarditis

  Pericarditis

  Immune thrombocytopenia

  Anaphylactic reaction

  Narcolepsy

  Disseminated intravascular coagulation

  Abortion spontaneous

  Stillbirth

  Premature labour

  Eclampsia

  HELLP syndrome

  Haemorrhage in pregnancy

  Abnormal labour

  Postpartum haemorrhage

  Endometritis

  Maternal death

  Thrombosis with thrombocytopenia syndrome

Population studied

• Short description of the study population:
  In Objective 1, all pregnancies starting between 1st January 2018, and up to 9 months before end of data will be eligible to enter the study. Pregnant women aged 12 to 55 years old at pregnancy start day and with 365 days or more of previous follow-up will be included. Pregnant women with an occurrence of the outcome of interest in the wash-out period relative to pregnancy start date will be excluded. ; In Objective 2, the source population will include everyone who is pregnant during the enrolment period* and who satisfy the following inclusion criteria: 1) aged 12 to 55 (both included) at pregnancy start date, 2) sex recorded as female, 3) ≥365 days of previous continuous observation at pregnancy start date, and 4) eligible to receive the exposure based on previous exposures: unvaccinated of COVID-19 (for first dose vs. unvaccinated), 16 days or more from first dose (for second dose vs. first dose), and 90 days or more since previous dose, with at least 2 previous doses (for booster vs. unboosted); *The enrolment period for SO 2.1 (1st vs. 0) goes from 01/04/2021 to 28/02/2022, aligning with the period when COVID-19 vaccination was recommended during pregnancy and when the majority started receiving their first dose. For SO 2.2 (2nd vs. 1st) enrolment period goes from 01/05/2021 to 31/03/2022, a 1-month extension from the enrolment in the first SO, as typically 2nd dose is ≥21 days after the 1st. Finally, enrolment period for SO 2.3 (Booster vs. 2nd) goes from when boosters were available, 01/10/2021, up until 12 months before data-cut. We will stop enrolling at that time for SO 2.3 to ensure there is enough follow-up time to see all the pregnant outcomes of interest. These periods are in-line with pregnant vaccination uptake seen in a previous study on COVID-19 vaccine effectiveness during pregnancy.;
• Age groups:
  • Adult and elderly population (≥18 years)
• Special population of interest:
  Pregnant women

Study design details

• Study design:
  Objective 1) estimate background rates of adverse descriptive epidemiology.; Objective 2) safety of mRNA COVID-19 vaccines following the Targe Trial Emulation framework.;
• Main study objective:
  Research Question 1: What are the background incidence rates of selected adverse events during pregnancy? Which are the differences in baseline characteristics between women who experience an adverse event during pregnancy vs. those who do not?; Research Question 2: What is the risk of adverse events associated with receiving a 1st, 2nd, or booster (3rd or subsequent doses) of an mRNA COVID-19 vaccine during pregnancy, compared to pregnant women eligible for the same dose but who have not yet received it?; Each research question corresponds to a study objective. To answer the first question (Objective 1), we aim to provide epidemiological context to adverse events occurring during pregnancy and post-partum. ; For the second research question (Objective 2), where the objective is to assess the safety of mRNA COVID-19 vaccination during pregnancy, we have three specific objectives (SOs):; • SO 2.1: To estimate the risk of pregnant AESI and MAE when receiving a 1st COVID-19 vaccine dose (mRNA type) during pregnancy compared to unvaccinated pregnant individuals.; • SO 2.2: To estimate the risk of AESI and MAE when receiving a 2nd COVID-19 vaccine dose (mRNA type) during pregnancy compared to individuals who are eligible for a 2nd dose during pregnancy but have not yet receive it.; • SO 2.3: To estimate the risk of AESI and MAE when receiving a booster (3rd or subsequent doses) COVID-19 vaccine dose (mRNA type) during pregnancy compared to individuals who are eligible for a booster dose during pregnancy but have not receive it.; ;
• Setting:
  For the first objective, all pregnancies meeting inclusion criteria from 01-01-2018 to the end of data availability will be included. ; For the second objective, all pregnancies meeting inclusion criteria during the dose-specific enrolment period (see “Study population”) will be eligible to enter the study. ; Risk Set Sampling will be used to create exposed and comparator cohorts. Each day during the enrolment period, when a pregnant woman from the source population becomes exposed (receives the vaccine dose of interest), we will select a group of unexposed pregnant women to enter the comparator cohort on the same day (index date). For each exposed woman, we will sample unexposed women matched by maternal age (within a 2-year range) and gestational age (within a 2-week range), ensuring they meet the inclusion and exclusion criteria at that date. Comparator sampling will be performed with replacement, and the sampling ratio will be determined based on prior exploratory analysis of the proportion of exposed to unexposed pregnant women in the database. An exposed woman will only be included in the cohorts if at least one suitable comparator can be identified from the source population. If a comparator later becomes exposed during the enrolment period, they will be allowed to re-enter the study as an exposed individual, provided that appropriate comparators are available.; ;
• Comparators:
  Not applicable for the first objective.; In the second objective, we will compare pregnant women who receive a first, second, or booster mRNA COVID-19 dose, with pregnant women eligible to receive the same dose but unvaccinated with it yet.;
• Outcomes:
  The study's outcomes include Adverse Events of Special Interest (AESI) and pregnancy-specific adverse events, Maternal Adverse Events (MAE). The outcome list is based on the prioritized COVID-19 vaccine AESI list by the Brighton Collaboration, and the Global Alignment on Immunization Safety Assessment in Pregnancy consortium. Outcomes include:; - 15 AESI: Deep vein thrombosis, Pulmonary embolism, Myocardial infarction, Ischaemic stroke, Bell’s palsy, Encephalitis, Guillain Barré Syndrome, Transverse Myelitis, Haemorrhagic stroke, Myocarditis or pericarditis, Thrombosis with Thrombocytopenia, Immune Thrombocytopenia, Anaphylaxis, Narcolepsy, Disseminated Intravascular Coagulation; - 13 MAE: Miscarriage, Stillbirth, Preterm labour, Eclampsia, HELLP syndrome, Antepartum Haemorrhage, Dysfunctional labour, Postpartum haemorrhage, Postpartum endometritis, Maternal death, Gestational diabetes, Ectopic pregnancy, Pre-eclampsia;
• Data analysis plan:
  In the first objective incidence rates per 100,000 person-years will be calculated as the number of incident cases divided by the total person-time at risk. Incidence rates will be estimated for calendar year and month using the R package `Incidence Prevalence, and stratified by age group (12-17, 18-34, and 35-55 years) and gestational trimester (0 to 90, 91 to 180, and 181 and more days). Incidence rates will not be estimated if there are less than 5 events in a given stratum.; In the second objective, Propensity Score with Overlap weighting will be used to account for observed confounding. Standardised mean difference on diagnosis and prescriptions before index date will be estimated between cohorts to assess covariate balance. Residual confounding will be assessed with a set of negative control outcomes. ; Incidence rate ratios (IRR) will be used to measure the relative incidence rate between exposed and comparator cohorts for each analysis and outcome of interest. 95% confidence intervals will be computed using the bootstrap method to account for non-independence of samples. IRR estimates will be calculated locally for each participating database and then combined using random-effects meta-analysis. Where heterogeneity (I2) is greater than 50%, estimates will not be meta-analysed and action will be taken to investigate which are the causes on this discrepancy. IRR will not be estimated if there are less than 5 events between exposed and comparator cohorts.; ;
• Summary results:
  At study completion, we will report incidence rates of adverse events occurring during pregnancy and the postpartum period. In addition, we will estimate the relative incidence of these adverse events among pregnant women who received a COVID-19 vaccine compared to those who did not.