GLP-1s, Pregnancy, and Birth Control: The 'Ozempic Babies' Guide

Three things matter if you can get pregnant and you take a GLP-1 drug. First, these drugs are not recommended in pregnancy, and every label says to stop once a pregnancy is recognized. Second, tirzepatide (Mounjaro, Zepbound) can make oral birth control less reliable for a few weeks after you start and after each dose increase; semaglutide (Ozempic, Wegovy) does not carry that warning. Third, the drugs can restore fertility you did not know you had, which is the main reason "Ozempic babies" became a story.

None of this means the drugs are unsafe to be on. It means the months around starting one, and the months around trying to conceive, need a contraception and timing plan. This guide lays out what is documented, what is still thin, and what to do.

"Ozempic babies" is the nickname for surprise pregnancies among people on GLP-1 drugs, some of whom had assumed they could not conceive. As a media trend it started in early 2024, with coverage in the Washington Post, CNN, and a wave of social-media accounts. As a measured phenomenon it is still thinly documented: a 2025 National Geographic review found reproductive endocrinologists describing plenty of anecdotes but almost no published data on changed pregnancy rates.

The best quantitative signal comes from an Australian general-practice cohort published in the Medical Journal of Australia in 2025. By 2022, 90.5% of new GLP-1 prescriptions to reproductive-age women went to women without diagnosed diabetes, only 21.2% of those women had contraception overlapping their first prescription, and 2.2% became pregnant within six months of starting. Two mechanisms explain the pattern, and they are not equally well supported.

This is the better-documented pathway and it applies to the whole class. Excess weight and insulin resistance suppress ovulation, most visibly in polycystic ovary syndrome (PCOS), the leading cause of ovulatory infertility. Losing 5 to 10% of body weight can restore regular ovulation, and GLP-1 drugs routinely clear that bar: in the SELECT trial, 44% of semaglutide patients lost more than 10% of their body weight within two years.

A meta-analysis of eleven PCOS trials found GLP-1 receptor agonists improved menstrual regularity and natural pregnancy rates, though a separate 2025 review of about a dozen small studies found gains in natural conception but not in IVF outcomes. The practical takeaway is simple: someone who was not ovulating reliably can start ovulating within weeks of starting one of these drugs, well before they would expect to be fertile again.

This is the part most coverage gets wrong by lumping all GLP-1s together. The interaction with oral birth control is specific to tirzepatide. Tirzepatide slows gastric emptying enough to reduce how much of an oral contraceptive the body absorbs. In the FDA pharmacology data, a single 5 mg dose cut the peak blood levels of the contraceptive hormones by more than half (up to 66% for one progestin) and total exposure by about 20%. The effect was largest after the first dose and faded as the body adjusted over subsequent weekly doses.

Semaglutide is different. The Wegovy and Ozempic labels state that semaglutide did not meaningfully affect the absorption of oral medications, and contraceptive hormone levels were if anything slightly higher when taken with it. Semaglutide carries no instruction to switch methods or add a backup. So the headline is precise: the pill-and-GLP-1 problem is a tirzepatide problem, not an Ozempic problem, even though the trend took Ozempic's name.

The tirzepatide label gives specific instructions. People using oral hormonal contraceptives should either switch to a non-oral method or add a barrier method (such as condoms) for four weeks after starting tirzepatide and for four weeks after each dose increase. Those are the windows when absorption is most disrupted.

Non-oral hormonal methods are not affected, because the interaction is about gut absorption rather than the hormones themselves. An IUD, implant, injection, patch, or vaginal ring stays reliable throughout. For someone who wants to stay on the pill, the cleanest approach is a barrier backup mapped to the start date and every titration step. None of this applies to semaglutide.

Because the drugs are not recommended in pregnancy, the planning question is how far ahead to stop, and that scales with how long each drug lingers. Semaglutide has the clearest instruction: its labels say to stop at least two months before a planned pregnancy, because of its long half-life. Tirzepatide clears faster, with a half-life of about five days and most of the drug gone within a month, so reviews and teratology services suggest stopping roughly four to five weeks ahead. Liraglutide (Saxenda, Victoza) has a half-life of about thirteen hours and needs only a few days. Orforglipron, the new oral GLP-1, is also not for use in pregnancy; its half-life sits between liraglutide's and the weekly injectables'.

One nuance worth raising with a clinician: rapid weight loss right before conception is not automatically good for the pregnancy. Reviews note that losing significant weight can help the mother but has been associated with smaller-for-gestational-age babies and preterm delivery, so the timing of both stopping the drug and stabilizing weight is a real conversation, more than a washout calculation.

The warnings rest mostly on animal studies, not on proven human harm. In pregnant animals, semaglutide caused embryofetal death and structural abnormalities, tirzepatide caused fetal growth reductions and abnormalities, and liraglutide caused fetal harm in rabbits at exposures below the human dose. Every label translates that into the same instruction: not recommended, discontinue when pregnancy is recognized.

The human data are limited but so far reassuring. The largest prospective series, drawn from six teratology information services and published in BMJ Open in 2024, followed 168 pregnancies exposed to GLP-1 drugs in early pregnancy and found major birth defects in 2.6%, versus 2.3% in a diabetes comparison group and 3.9% in an overweight group, with similar miscarriage rates. A separate registry analysis and a small Danish cohort of 32 semaglutide-exposed pregnancies found no clear malformation signal, though all of these studies are small with wide error bars. The honest reading is no detected increase in harm rather than demonstrated safety, with essentially no data yet on long-term outcomes for the children.

There are no human studies of GLP-1 drugs during breastfeeding. The molecules are large peptides that are likely broken down in the infant's gut and probably transfer into milk in small amounts, but "probably" is the operative word, and animal data on milk transfer vary widely by drug. The labels generally advise against use while breastfeeding until there is evidence, so this is another decision to make with a clinician rather than from the package insert alone.

If you take a GLP-1 and could become pregnant, assume your fertility may improve whether or not you intended it, and use reliable contraception. If that drug is tirzepatide and you rely on the pill, add a barrier method or switch to a non-oral method for four weeks after starting and after each dose increase; if it is semaglutide, the pill is unaffected. When you do want to conceive, stop semaglutide at least two months ahead, tirzepatide about a month ahead, and talk through the timing of weight stabilization. The drugs are not proven dangerous in pregnancy, but they are not proven safe either, and the data are too thin to treat an accidental exposure casually. The recurring theme in every careful source is the same: plan the contraception and the timing before you need them, not after.