A systematic scoping review and consensus guideline published in Obesity Reviews on August 4, 2026 put a question that has been sitting in the fine print of a drug label in front of a general audience, and by August 18 it was running in national consumer press under headlines about birth control failing. The underlying facts are narrower and more useful than the headlines: one compound in this class has a measured, labeled contraceptive interaction, one has a dedicated study showing none, and several of the compounds people actually buy through the research channel have neither.
That split is the whole story. It is also the part that gets flattened when "GLP-1s" is treated as a single thing, because the interaction is not a class effect in the labeling — it is a tirzepatide effect, extended by precaution to a new oral pill, and absent from the semaglutide record.
Research-context information only. This article reports what approved drug labeling, published pharmacokinetic trials and a new consensus guideline state, as issued. It is not medical advice and not a contraception protocol. Retatrutide, cagrilintide and other unapproved compounds have no reviewed labeling of any kind; research-use-only material sold by peptide vendors is not FDA-approved for human use and has not been evaluated for safety, purity or potency as sold. Contraception is a clinical decision that belongs with a licensed prescriber.
What the new consensus paper actually says
The paper is Maslin and colleagues, "Incretin-Based Medications in Women and Reproduction: A Systematic Scoping Review and Consensus Guidelines for Clinical Practice," published in Obesity Reviews on August 4, 2026 (DOI 10.1111/obr.70203). It was produced by authors across 23 institutions in Europe, North America and the Middle East, and it screened 34 studies — randomised trials, observational cohorts, pharmacovigilance analyses and case reports — to cover fertility, pregnancy, breastfeeding and postpartum management.
Three findings carried the coverage. The review reported that available evidence "did not identify an increased risk of major congenital anomalies following inadvertent early pregnancy exposure," while noting substantial gaps on longer-term maternal and child outcomes. It found that incretin-based medications may improve fertility-related outcomes in women with obesity and with the condition now formally renamed polyendocrine metabolic ovarian syndrome — a renaming we covered separately in GLP-1s and PMOS. And it concluded that current evidence "remains insufficient to support their use during pregnancy or breastfeeding," which is why the authors landed on contraceptive counselling as the practical recommendation rather than a dosing rule.
The authors' own summary of the state of the field is the most honest line in the paper: nearly half of the clinically important questions surrounding incretin-based medications and reproductive health remain unanswered. That is a scoping review telling you where the map ends, not a trial telling you what to do.
The label matrix: which compound carries which instruction
This is where the class generalisation breaks down. The instruction is not uniform, and the differences are documented rather than inferred.
| Compound | Contraception instruction in labeling | Measured in a dedicated OC study? |
|---|---|---|
| Tirzepatide | Switch to a non-oral method, or add a barrier method, for 4 weeks after initiation and 4 weeks after each dose escalation | Yes — reduced exposure measured |
| Orforglipron (oral) | Switch to a non-oral method, or add a barrier method, for 30 days after initiation and 30 days after each dose escalation | No — labeling states the effect "has not been evaluated in a clinical trial" |
| Semaglutide | No equivalent contraception instruction | Yes — oral semaglutide study found no effect |
| Liraglutide | No equivalent contraception instruction | Covered in the pharmacist review; no significant effect reported |
| Retatrutide, cagrilintide | No approved labeling exists | No published dedicated study |
The tirzepatide numbers are the anchor for the whole discussion. In the interaction study reflected in the approved labeling, a combined oral contraceptive containing 0.035 mg ethinyl estradiol and 0.25 mg norgestimate was given alongside a single 5 mg tirzepatide dose. Mean peak concentration fell by 59% for ethinyl estradiol, 66% for norgestimate and 55% for norelgestromin. Mean total exposure — the AUC, which is the figure that generally matters more for contraceptive efficacy than the peak — fell by 20%, 21% and 23% respectively.
The labeling attributes this to delayed gastric emptying and states that the delay "is largest after the first dose and diminishes over time." The same tachyphylaxis shows up in the acetaminophen probe used to characterise gastric emptying: after a first 5 mg dose, acetaminophen peak concentration was reduced by 50% and arrived an hour later, but by week 4 there was no meaningful impact on peak or timing, and total exposure was never affected. That decay pattern is exactly why the instruction attaches to initiation and to each escalation rather than running for the whole course of treatment.
The 2024 review in the Journal of the American Pharmacists Association (PMID 37940101) put the class comparison on the record. It included 6 studies. The one investigating tirzepatide showed a statistically significant reduction in oral contraceptive concentration; the other 5, covering GLP-1 receptor agonists, showed no statistically or clinically significant difference. The authors attributed the divergence to tirzepatide's rapid dose escalation and greater delay in gastric emptying.
For semaglutide specifically, the dedicated evidence points the other way. A crossover trial in healthy postmenopausal women given combined ethinylestradiol and levonorgestrel with oral semaglutide, published in Clinical Pharmacokinetics in 2021 (PMID 33782832), concluded that co-administration did not affect the pharmacokinetics of either hormone.
Where there is no label at all
The compounds with the largest research-channel demand right now — retatrutide, cagrilintide, and the cagrilintide-retatrutide pairing — sit outside this entire framework, because none of them has been through a regulatory review that produces labeling.
That absence cuts in a specific direction rather than a reassuring or an alarming one. There is no measured contraceptive interaction for retatrutide because no dedicated study has been published, not because a study came back clean. What is documented is that retatrutide acts on GLP-1 and GIP receptors — the same two tirzepatide hits — plus the glucagon receptor, and that gastric-emptying delay is a reported feature of the class across compounds. The tirzepatide finding is the closest labeled analogue anyone has. Whether it transfers is an open question that a scoping review would file under the half that remains unanswered.
Two other structural points are worth naming. The label instruction is written for a supervised titration schedule with a defined escalation calendar; a self-directed research protocol does not necessarily have one, which makes "4 weeks after each dose escalation" harder to map onto. And the instruction is normally delivered by a dispensing pharmacist at the counter, which is a step the research-supply channel does not include by design. The information exists publicly either way — it is in the labeling, and it is above.
Vendor-side, this is a pricing and sourcing question like any other. (Affiliate disclosure: the vendor links in this paragraph and below earn The Peptide Catalog a commission.) Current per-milligram comparisons across tracked vendors sit on the tirzepatide buying surface, the semaglutide surface and the retatrutide surface, and active vendor discounts across the whole class are consolidated on the deals page.
