In medicine, few assurances carry as much emotional weight—or ethical responsibility—as declaring something “safe during pregnancy.”
History has taught us that such declarations, when premature or poorly grounded, can have devastating consequences. Thalidomide. Diethylstilbestrol (DES). The list is long, and the lessons should have been unforgettable.
So when a study published in Pediatrics – the flagship journal of the American Academy of Pediatrics – claimed that COVID-19 vaccination in early pregnancy was safe, it understandably made headlines. It was embraced by public health bodies around the world. It reassured millions of expectant mothers.
But behind the authority of that journal, and the confidence of those headlines, lies a study built on sand.
A Glaring Omission
The study reviewed 78,052 pregnancies that ended in live births. What it did not include were the 20,341 pregnancies that ended in miscarriage, stillbirth, or other non-live outcomes.
Let that sink in.
The primary purpose of evaluating vaccine safety in pregnancy is to understand whether there are risks to the developing fetus. Yet the study excluded over 20,000 pregnancies that may have ended because of adverse outcomes. This isn’t a minor oversight. It’s a statistical sleight of hand known as live-birth bias, and it fundamentally undermines the validity of the findings.
Even two of the authors (Dr Stacey Rowe and Dr Annette Regan) acknowledged that this exclusion “could lead to an underestimation of identified outcomes.”
Yet they moved forward with a conclusion of safety.
How?
The Black Box of Commercial Data
As if that weren’t enough, the study relied on a commercial database known as Merative® MarketScan® – a vast collection of de-identified insurance claims, prescriptions, and hospital records compiled by a private company.
MarketScan data is not open-access. It is not peer-verifiable. It is not transparent. Researchers using it cannot confirm if the patients represented are real, if records are complete, or if any cleaning or filtering has biased the data set.
It’s a black box. And the study’s authors never verified what was inside.
When questioned about the authenticity of the dataset—whether the patient records were real or representative—the authors simply did not respond.
This is not how science is supposed to work. Especially not science that informs decisions affecting pregnant women and unborn children.
Red Flags and Statistical Impossibilities
The authors conducted 93 separate statistical tests to assess the presence of adverse outcomes. Not one came back statistically significant.
Not one.
That might sound reassuring—unless you understand statistics. With that many tests, you’d expect at least a few false positives purely by chance. In fact, the probability of none reaching significance is just 0.8%.
That improbability suggests either a dataset with highly unusual properties—or one that cannot be trusted.
We’ve been here before. Think back to the Surgisphere scandal, in which entirely fabricated hospital data was used to publish major papers in The Lancet and The New England Journal of Medicine. Those papers were eventually retracted, but not before influencing global medical decisions.
The Pediatrics study carries the same warning signs.
Trust Without Evidence
And yet, the conclusions are being used by governments around the world to reassure pregnant women. In Australia, official health guidance states that recommendations for pregnant women are the same as for the general population.
This, despite:
- The lack of randomized clinical trials including pregnant women
- The known limitations of the original Pfizer trial
- And now, a deeply flawed observational study being used as a cornerstone of reassurance
The original Pfizer trial excluded pregnant women entirely. Later pregnancy-related data was based on voluntary follow-up and passive surveillance—far from the rigor of randomized trials.
When the lives of unborn children are at stake, reassurances without access to the data are not just reckless—they’re unforgivable.
A Dangerous Precedent
We are entering an era where scientific conclusions are drawn from proprietary data, inaccessible to the public, and often not even understood by the researchers using them.
An era where we’re told to “trust the science” — but denied the transparency necessary to verify that science.
Pregnant women are being asked to make shared decisions with their doctors based on studies like this one. But what does shared decision-making even mean when the evidence base is broken?
This isn’t just bad science. It’s a betrayal of scientific ethics. And when the consequences involve birth defects, miscarriage, or long-term harm to a child, the cost of getting it wrong is simply too high.
The Critical Mindshift
This article stands alone, but it doesn’t stand apart. It is intimately connected to the growing list of examples we’ve examined in our Medical Hoax series—a series that exposes how authority, funding, and media narratives can warp science beyond recognition.
This particular case raises the question:
How many of our public health recommendations are based on science we aren’t allowed to question, using data we aren’t allowed to see?
Pregnant women deserve more than glossy reassurances. They deserve access to unfiltered data, transparency about risk, and the chance to make decisions based on real evidence—not curated narratives.
When trust is demanded but never earned, what we’re left with isn’t science.
It’s marketing—masquerading as medicine.
Further Reading: Beyond the Headlines
If you’re not satisfied with surface-level reassurances, you’re not alone. The following studies and investigative reports provide deeper insight into the integrity of COVID-19 vaccine research and the challenges of data transparency in medicine:
COVID-19 Vaccination During Early Pregnancy: Is It Safe?
American Academy of Pediatrics March 2025
This is the original study that excluded over 20,000 non-live pregnancies from its safety analysis—raising red flags over methodological bias.
The Surgisphere Scandal: What Went Wrong?
The Scientist
A powerful exposé on how fake datasets influenced some of the world’s most respected medical journals before being retracted.
COVID-19 Vaccination During Pregnancy Not Linked to Birth Defects
Healio
A media summary of the Pediatrics study, illustrating how the flawed findings were widely publicized without scrutiny.
Covid-19: Lancet Retracts Paper That Halted Hydroxychloroquine Trials
The Guardian
Coverage of the fallout from one of the most infamous data scandals of the pandemic.
These are just the tip of the iceberg. As the floodgates slowly begin to open, expect to see more buried data, flawed methodologies, and institutional silence exposed. What was once dismissed as fringe concern is now finding daylight—and it’s only the beginning.
Image Acknowledgement
We’re grateful to the talented photographers and designers on Unsplash for providing beautiful, free-to-use images. The image on this page is by Devon Divine. Check out their work here: https://unsplash.com/@lightrisephoto.
