Evidence-based · GLP-1 & Metabolic

GLP-1s in Pregnancy and Before Conception
GLP-1 drugs are not recommended in pregnancy, and labels call for stopping them weeks to months before trying to conceive. Here's what's known, what isn't, and why 'Ozempic babies' are a real phenomenon.
Part ofThe GLP-1 Guide→GLP-1 receptor agonists have become a fixture of weight management and type 2 diabetes care, which means more people of reproductive age are on them than ever before. That raises a specific and underexamined question: what happens if you get pregnant while taking one, or want to conceive soon after stopping? The honest answer is that the data are thinner than most people assume, and the guidance is built more on caution than on certainty.

What the labels actually say
Every GLP-1 receptor agonist on the market — semaglutide (Ozempic, Wegovy, Rybelsus), tirzepatide (Zepbound, Mounjaro), liraglutide (Saxenda, Victoza), and dulaglutide (Trulicity) — carries labeling that recommends against use during pregnancy. Manufacturers advise discontinuing the drug before a planned pregnancy, with the lead time varying by how long the drug stays active in the body. Semaglutide and tirzepatide are long-acting, once-weekly injections, so their labels generally call for stopping roughly two months before attempting to conceive to let the drug clear. Shorter-acting agents like liraglutide, which is dosed daily, carry a shorter recommended washout. If a pregnancy is discovered while already on treatment, the standard advice is to stop the medication and discuss next steps with an obstetric provider — not to panic, but not to continue business as usual either.
This isn’t guidance based on a large body of controlled human trials. It’s a precautionary stance rooted mostly in animal data and the simple fact that pregnant people are almost never enrolled in the trials that got these drugs approved in the first place.
What the animal data showed — and why it doesn’t translate cleanly
In animal reproduction studies, GLP-1 receptor agonists given at doses well above typical human exposure were associated with structural birth defects, growth restriction, and pregnancy loss in some species. That’s a real signal that warranted the cautious labeling. But animal reproductive toxicology at supratherapeutic doses is a blunt instrument — it flags a possibility, not a confirmed human risk, and rodent and primate placental physiology doesn’t map onto human pregnancy in a one-to-one way. The FDA’s own drug labels reflect this: they describe the animal findings plainly and then note that human data are insufficient to establish a drug-associated risk one way or the other, which is a very different statement from “this drug causes harm in people.”
Observational data in humans — largely from registries and cohort studies of women who conceived while on or shortly after a GLP-1 — have so far not shown a consistent signal of major congenital malformations above the background rate, but these studies are underpowered, retrospective, and can’t fully separate the drug’s effect from the effects of obesity, diabetes, or rapid weight loss itself, all of which independently affect pregnancy outcomes. Uncontrolled maternal blood sugar and obesity carry their own well-documented pregnancy risks, which is part of why this area is genuinely hard to untangle rather than a simple “drug bad” story.
The caution around GLP-1s in pregnancy isn’t based on proof of harm in humans — it’s based on the absence of proof of safety, plus animal findings serious enough that no one wants to be the exception that confirms them.

The “Ozempic babies” phenomenon
Anecdotal reports and some fertility-clinic observations have described people conceiving unexpectedly after starting a GLP-1 drug, even in cases where they’d struggled with infertility for years — a trend that picked up enough steam online to earn the nickname “Ozempic babies.” The most plausible explanation isn’t that these drugs boost fertility directly. Obesity and insulin resistance are independently associated with ovulatory dysfunction and reduced fertility, particularly in conditions like polycystic ovary syndrome. Meaningful weight loss and improved insulin sensitivity can restore ovulation in people who weren’t ovulating regularly before, which means a pregnancy can happen while someone is still taking the drug — and is on the label’s “do not use” list at exactly the wrong moment.
There’s a second mechanical wrinkle specific to tirzepatide: its label notes that it can delay gastric emptying, and the manufacturer has flagged that this may reduce the absorption of oral contraceptives, particularly around the time of dose escalation. The current guidance for people on tirzepatide who rely on oral birth control is to add a barrier method or switch to a non-oral contraceptive for a period after starting or increasing the dose. This is a labeling-level caution based on pharmacokinetic reasoning rather than a large confirmed pregnancy-rate signal, but it’s a plausible and specific mechanism, which is why it’s taken seriously.

What this means in practice
| Situation | Common guidance |
|---|---|
| Planning a pregnancy | Stop the GLP-1 before conception; lead time varies by drug (often cited as around 2 months for long-acting agents) |
| Pregnancy discovered while on treatment | Stop the medication and contact your obstetric provider promptly |
| On tirzepatide and using oral contraception | Consider a backup or non-oral method, especially during dose escalation, per manufacturer guidance |
| Breastfeeding | Data are limited; discuss with a clinician, as recommendations vary by drug |
None of this is a substitute for an individualized conversation. Someone using a GLP-1 for type 2 diabetes has different risk-benefit tradeoffs around pregnancy planning than someone using it purely for weight management, and untreated diabetes in pregnancy carries its own serious risks that have to be weighed against drug discontinuation.
The takeaway
The evidence base here is genuinely incomplete: animal studies flagged real concerns, human pregnancy data are too sparse to confirm or rule out risk, and the “Ozempic babies” story is more likely a fertility side effect of weight loss than a direct pro-fertility action of the drug. What’s clear is that none of the major GLP-1 agonists are currently considered appropriate to continue through a planned or ongoing pregnancy. Anyone taking one of these drugs who is pregnant, trying to conceive, or relying on oral contraception while on tirzepatide should talk to their prescribing clinician and, ideally, an obstetric provider before making any changes — this is not a decision to navigate from forum threads or drug labels alone.
Sources
Stay current
Get evidence-based briefings in your inbox.