The short version: stop before pregnancy, and stop if you become pregnant
The Ozempic label states that the available data are insufficient to evaluate the risk of birth defects or miscarriage, and directs you to discontinue it before a planned pregnancy. As a diabetes drug the framing is "insufficient data" rather than the "no benefit" language the weight-loss versions use — but the instruction to stop is the same.
What the Ozempic label says about pregnancy
The Ozempic label states that the available data are insufficient to evaluate the risk of birth defects or miscarriage, and directs you to discontinue it before a planned pregnancy. As a diabetes drug the framing is "insufficient data" rather than the "no benefit" language the weight-loss versions use — but the instruction to stop is the same.
One point worth being precise about, because most pages get it wrong: Ozempic is not a formal contraindication in pregnancy — that label wording is reserved for Saxenda. The instruction to stop sits in the pregnancy section rather than the contraindications list, which is a real distinction even though the practical advice is the same.
When to stop Ozempic before trying to conceive
The label says to stop Ozempic at least 2 months before a planned pregnancy, because of semaglutide’s roughly one-week half-life. This is label-mandated wording.
What the warning is actually based on: animal data
Every warning on these labels comes from animal studies — rats and rabbits at clinically relevant doses showed embryo loss and structural abnormalities. Pregnant women were excluded from all the registration trials, so there is no human trial data behind the label caution.
What the human data show so far
The human evidence has started to arrive, and so far it is reassuring rather than alarming. A Danish nationwide cohort published in 2026 (756,636 pregnancies, 529 exposed to a GLP-1) found no increased risk of major malformations after matching. A 2024 registry study and a 2025 systematic review of over 1,000 semaglutide-exposed pregnancies reached the same conclusion. But these are observational, the numbers are small, they are weighted toward pregnancies with diabetes, and miscarriage is under-captured — so the honest read is "no red flag has emerged yet", not "proven safe".
What to do if you are already pregnant
If you find out you are pregnant while taking a GLP-1, the guidance from the labels, ACOG and the ADA is the same: stop the medication now and tell your prescriber. Inadvertent early exposure before you knew is not, on its own, a reason for any intervention beyond stopping and normal prenatal care — the human data on early exposure are reassuring. What matters next is managing the underlying condition without the drug.
Stopping the drug is not the end of the plan. Observational work in 2025 found that stopping a GLP-1 around pregnancy was linked to more gestational weight gain and higher rates of preterm birth, gestational diabetes and high blood pressure — not because stopping is wrong, but because the underlying metabolic condition still needs active management. Have that conversation with your clinician before you conceive, not after.
“Ozempic babies”: why unplanned pregnancies happen on these drugs
The unexpected-pregnancy stories are real, and there are two separate mechanisms behind them — and they are not the same across drugs.
- Restored ovulation. Weight loss and improved insulin sensitivity can restart ovulation in people who were not ovulating regularly — especially with PCOS. Fertility can return faster than expected, and it applies to every GLP-1. If you are not trying to conceive, this is the reason to keep contraception in place even if you were not before.
- The pill can fail — but only on tirzepatide. Ozempic contains semaglutide, which was specifically tested and does NOT carry this warning — it does not meaningfully reduce oral-contraceptive absorption. This risk is unique to tirzepatide (Zepbound and Mounjaro), so if you are on Ozempic the pill is not the concern; restored fertility is.
Breastfeeding
There is no human data on Ozempicin breast milk. The labels state this plainly and defer to a benefit-versus-risk judgement with your clinician rather than an outright ban. Clinicians often note that these are large peptide molecules, unlikely to pass into milk in quantity and likely to be broken down in an infant's gut — but that is pharmacologic reasoning, not human evidence, and the labels are clear that the data are simply absent.
Trying to conceive, or preventing it, on a GLP-1? See GLP-1 and fertility and GLP-1 and birth control.
For the cross-drug comparison and the full evidence, see GLP-1 and pregnancy.