CDC study on COVID-19 vaccination and spontaneous abortions did not contain major error

For many months, vaccine skeptics have been sounding false alarms about the effect of COVID-19 vaccines on pregnancy and fertility.

Concerns about the risk of spontaneous abortion, also known as miscarriage, have resurfaced following a November report in Science, Public Health Policy & the Law, a journal run by James Lyons-Weiler.

"Report Finds Increased Risk of Spontaneous Abortion Following COVID-19 Vaccination During Pregnancy," reads the headline on a post from Lyons-Weiler, who seeks to summarize the report’s findings. "Epidemiologists Find CDC Study Contains Major Error."

Lyons-Weiler, whose claims about COVID-19 we have previously fact-checked, claims in his piece that the authors of the report found a "major error" in a study conducted by the Centers for Disease Control and Prevention that was published in the New England Journal of Medicine in June. The miscarriage rate, the authors asserted, was seven to eight times higher than the CDC study presented.

The authors' calculations, however, misrepresent the CDC study findings by manipulating and cherry-picking data, using a similar method to previously circulating claims on social media about the same CDC study.

The blog post was flagged as part of Facebook’s efforts to combat false news and misinformation on its News Feed. (Read more about our partnership with Facebook.) 

The authors of the original CDC study, titled "Preliminary Findings of mRNA Covid-19 Vaccine Safety in Pregnant Persons," used data from three federal reporting systems monitoring vaccine safety. They contacted 3,958 people enrolled in one of these systems about three months after they had received the Pfizer-BioNTech or Moderna COVID-19 vaccine.

Of those people, 712 had given birth to healthy babies, 104 experienced miscarriages and one had a stillbirth. Ten people had ectopic pregnancies or induced abortions. The remaining 3,131 participants were either still pregnant or had not yet completed follow-up.

They found 104 miscarriages out of 827 completed pregnancies, which divides to a 12.6% miscarriage rate — comparable with the average miscarriage rate of 12.5% to 18.7% in the general population.

The authors of the report in Lyon-Weiler's journal claim that this figure is incorrect, presenting alternative calculations that result in a much higher miscarriage rate of 82% to 91%.

They arrived at these figures by using the same flawed method that was used in social media posts that we saw circulating in July. Those posts, which we fact-checked, had excluded 700 people vaccinated in the third trimester, when fetuses are more developed and the risk of miscarriage is reduced. Likewise, the authors excluded 700 people who were vaccinated after 20 weeks of pregnancy, the cutoff period that defines miscarriage. This left just 127 completed pregnancies. The 104 miscarriages out of 127 pregnancies amounted to 82%.

Featured Fact-check

Instagram posts

stated on November 7, 2024 in Instagram posts

“2,038 suicides have been reported (Nov. 6) following Donald Trump’s election win.”

This 82%, however, is not the miscarriage rate. Determining the effects of vaccination early in pregnancy means looking only at people who were vaccinated before 20 weeks, but it’s misleading to apply the narrow denominator of 127 to the preliminary data in the NEJM paper.

"[Y]ou can’t do that because the data set is looking at all pregnancy outcomes," said Dr. Jen Gunter, an OB-GYN, in a blog post. "Of course when you remove those 700 pregnancies the miscarriage rate looks artificially high, because the only way a person got into the data set … was if the pregnancy had ended and they had been vaccinated."

Likewise, the upper range of their estimate, 91%, used a denominator that only counted people with loss of pregnancy. This included 104 people who had spontaneous abortions and 10 people who had ectopic pregnancies or induced abortions, essentially dividing the number of miscarriages by itself. Determining the frequency of a certain outcome (i.e., miscarriages) using a sample that grossly overrepresents that outcome necessarily produces an inflated figure that cannot be applied to the general population.

A more precise assessment of whether vaccination before the third trimester increases miscarriage rates would require following these pregnancies to completion. "These pregnancies will eventually form the denominator to look at vaccination in the first and second trimesters, but these pregnancies have to end and all data needs to be collected before they can be used," said Gunter.

"There is no conspiracy to hide miscarriages here, just an abuse of statistics," said Gunter.

Since the CDC study was originally published, more data has become available.

The authors of the CDC study have both acknowledged that there was an error in the denominator that they used, because it included people vaccinated after the 20-week cutoff. They published another paper on Oct. 14 with updated data that now included 2,456 pregnant people. They found that the cumulative miscarriage risk for those at 6 to 20 weeks of pregnancy was 14.1%, which is still comparable to the average miscarriage rate in the general population.

Other studies corroborate these findings. A September paper in the Journal of the American Medical Association analyzed 105,446 pregnancies drawn from medical record data from eight health systems in the U.S. and found no increased risk for miscarriage following vaccination with the Pfizer-BioNTech or Moderna COVID-19 vaccine.

The CDC issued an urgent health advisory on Sep. 29 strongly recommending COVID-19 vaccination for pregnant and recently pregnant people, noting their increased risk of severe illness. An April multinational study found that people at any stage of pregnancy or delivery who contract COVID-19 are 22 times more likely to die than those without COVID-19, and their newborns are twice as likely to require intensive care or die after birth. The American College of Obstetricians and Gynecologists also updated its guidelines on Nov. 3, 2021, to recommend that pregnant and recently pregnant people receive a COVID-19 booster dose.

A blog post claims that a new report had found a major error in a CDC study, alleging that the miscarriage rate for pregnant people who received the COVID-19 vaccine was several times higher than presented in the study.

The authors of the report, however, arrived at these figures by cherry-picking and manipulating data from the CDC study, and failing to account for all completed pregnancies. 

The CDC study has since published follow-up data that supports its original findings that there is no increased risk of miscarriage following COVID-19 vaccination.

We rate the claim False.