Higher-than-Expected Fetal Loss Associated with mRNA COVID-19 Vaccination During Early Pregnancy: A Comprehensive Observational Study in Israel

Analysis of over 226,000 pregnancies reveals increased fetal loss rates linked to COVID-19 vaccination in gestational weeks 8-13, highlighting the need for further safety evaluations and dedicated cli

Photo by CDC on Unsplash

---

Summary

This extensive observational study analyzed 226,395 singleton pregnancies in Israel from 2016 to 2022 to investigate the relationship between mRNA COVID-19 vaccination during pregnancy and fetal loss. The focus was on vaccinations administered during early (weeks 8-13) and mid-pregnancy (weeks 14-27), primarily involving Pfizer’s BNT162b2 vaccine.

Link to study: https://www.medrxiv.org/content/10.1101/2025.06.18.25329352v1.full-text

Background

Pregnant women were excluded from initial COVID-19 vaccine trials, leaving a gap in understanding vaccine safety during pregnancy—especially early pregnancy when risks of teratogenic effects are highest. Observational studies so far often suffered from biases such as healthy vaccinee bias, limiting their reliability.

Methods

The study utilized Israel’s Maccabi Healthcare Services (MHS) electronic health records, linking pregnancy data with vaccination and clinical information. An observed-to-expected analysis compared actual fetal loss rates among vaccinated pregnant women to expected rates based on a model trained on historical data from 2016-2018. This model adjusted for individual risk factors including maternal age, health status, and gestational timing.

To reduce confounding, control cohorts were established: pregnant women vaccinated for influenza during pregnancy, and women vaccinated for COVID-19 or influenza prior to pregnancy. The main outcome was eventual fetal loss — a composite including spontaneous and induced abortions and stillbirths.

Key Findings

• Significantly Higher Fetal Loss After Early COVID-19 Vaccination: Women vaccinated with dose 1 during gestational weeks 8-13 experienced approximately 3.85 additional fetal losses per 100 pregnancies above the expected rate (~13 observed vs. 9 expected fetal losses per 100 pregnancies). Dose 3 during the same period was associated with 1.9 additional fetal losses per 100 pregnancies.

• Timing Matters: Vaccinations administered later in pregnancy (weeks 14-27) showed lower-than-expected fetal loss rates, likely influenced by healthy vaccinee bias rather than a protective effect.

• Control Cohorts Show Contrasting Patterns: Influenza vaccination during pregnancy was consistently linked with fewer-than-expected fetal losses, supporting the presence of healthy vaccinee bias in observational vaccine studies.

• Late Fetal Losses Confirm Biological Impact: A substantial portion of excess fetal losses after early COVID-19 vaccination occurred after week 20, including after week 25, suggesting these were not merely elective abortions but involved biological mechanisms such as stillbirth or medically indicated pregnancy termination.

• No Similar Signal for Vaccination Prior to Pregnancy or SARS-CoV-2 Infection: Women vaccinated before pregnancy or infected with SARS-CoV-2 during pregnancy did not show increased fetal loss rates.

• Dose-response Relationship: The excess fetal loss was higher for dose 1 and 2 administered during early pregnancy compared to dose 3 after two doses prior to pregnancy, which may indicate a biological dose-response effect.

Interpretation and Context

The findings raise concerns about vaccination with mRNA COVID-19 vaccines during early gestation (weeks 8-13). They align with known risks of teratogens affecting early development stages and possible biological mechanisms such as transplacental transfer of vaccine components or vaccine-induced thrombotic events that could contribute to fetal loss.

Healthy vaccinee bias likely explains the lower-than-expected fetal loss rates observed with influenza vaccination and COVID-19 vaccination later in pregnancy. The observed-to-expected analysis offers an advantage over traditional observational methods by matching vaccinated women to synthetic controls based on individual risk profiles derived from historical data.

Limitations

The study design cannot prove causality but identifies a safety signal requiring further investigation. It excludes weeks 1-7 due to inconsistent pregnancy documentation and cannot fully separate elective from medically driven induced abortions. Data are limited to Israel, and replication in other populations is needed.

Conclusion

This study provides evidence of a higher-than-expected rate of fetal losses following mRNA COVID-19 vaccination during early pregnancy (gestational weeks 8-13). The results underscore the urgent need for regulatory review, further biological research on mechanisms, and dedicated prospective clinical trials to evaluate vaccine safety in pregnant women, especially in early gestation.