Skip to content
GLP Loss
← Research
Safety

GLP-1s and the Pill: Why Only One Label Says Use Backup

The tirzepatide labels carry a section headed Contraception and a four-week backup instruction. The semaglutide label does not contain the word — it lists ethinyl estradiol and levonorgestrel among the drugs nothing happened to. Five molecules were tested; here is what each one actually did.

Owen Castellanos9 min read
Five incretin drugs, one contraceptive pillFall in peak ethinyl estradiol concentration, each manufacturer’s own trialTirzepatide 5 mg, single dose59%Exenatide, pill 30 min after46%Liraglutide, pill 7 hours after12%Semaglutide 1 mg weeklyno reductionOral semaglutide at steady stateno reductionOne of the five carries a contraception instruction.Both tirzepatide labels name it. No semaglutide label mentions the word.No study in any of them counted a pregnancy.

Two prescriptions sit in the same drawer more often than any other pair this site covers: a weekly injection for weight, and a daily pill for birth control. Whether the first disturbs the second has a precise answer, and the answer depends entirely on which injection it is. One molecule’s label tells a reader to add a second method of contraception. The other molecule’s label does not use the word. That gap is not an oversight, and it is not a difference in how carefully the two companies wrote — it follows from what each company measured.

The instruction, quoted

The tirzepatide labels carry a section headed Contraception, and its first sentence states the reason: “Use of ZEPBOUND may reduce the efficacy of oral hormonal contraceptives due to delayed gastric emptying. This delay is largest after the first dose and diminishes over time.” The instruction that follows is to “switch to a non-oral contraceptive method, or add a barrier method of contraception for 4 weeks after initiation with ZEPBOUND and for 4 weeks after each dose escalation.”[1] The diabetes-branded product carries the same paragraph with its own name substituted.[2] Both add that hormonal contraceptives not administered orally should not be affected. Why the window reopens at every dose step, rather than closing after the first month, is worked through in the contraception and pregnancy article.

Now the semaglutide side. Searched as a document, the weight-management label does not contain the string “contracept” anywhere. What it contains instead is this, inside its clinical pharmacology section: “No clinically significant differences in the pharmacokinetics of the following drugs were observed when used concomitantly with semaglutide: lisinopril, S-warfarin, R-warfarin, metformin, digoxin, ethinyl estradiol, levonorgestrel, furosemide or rosuvastatin.”[3] The two hormones in a combined pill are named — in a list of drugs that nothing happened to. The general caution in the same label is the one that covers every swallowed drug in the class, and it asks for extra monitoring where the therapeutic index is narrow. A contraceptive is not in that category, so no instruction attaches.

Four molecules were tested and nothing happened

The asymmetry is easier to trust once the other studies are laid out, because there are more of them than most summaries admit. Forty-three postmenopausal women with type 2 diabetes took ethinyl estradiol 0.03 mg with levonorgestrel 0.15 mg for eight days without semaglutide and eight days with it at a steady-state 1.0 mg weekly. Exposure to the estrogen met the bioequivalence criterion at a ratio of 1.11 (1.06 to 1.15); the progestin came in 20% higher, at 1.20 (1.15 to 1.26). Peak concentration stayed inside the criterion for both.[4] Note the direction. The one component that moved moved upward.

The swallowed form of semaglutide was tested the same way, and it is the harder test, because a tablet carrying an absorption enhancer has two ways to interfere rather than one. Twenty-five healthy postmenopausal women received the same combined pill alone, with the enhancer alone, and with oral semaglutide. Exposure ratios came back at 1.06 for the estrogen (1.01 to 1.10) and 1.06 for the progestin (0.97 to 1.17), with peak concentration unaffected and the enhancer alone doing nothing.[5] The route distinction that makes that a separate question is set out in the comparison of the two delivery routes.

Dulaglutide was given alongside a norgestimate pill in a study whose conclusion is embedded in its title: no dose adjustment is recommended. Peak concentrations were generally lower and time to peak generally later, while total exposure was similar.[6] And liraglutide, with the pill taken seven hours after the injection, lowered peak concentrations of the estrogen and the progestin by 12% and 13%, left estrogen exposure untouched, raised progestin exposure 18%, and pushed both peaks back an hour and a half.[7] Liraglutide is the instructive control here: a real, measured, published effect on a contraceptive, attached to no instruction at all, because the size of it did not warrant one.

Exenatide shows that the clock matters more than the drug

The oldest study in this set is also the one that best explains the mechanism, because it varied the timing on purpose. Thirty-two healthy women took a combined pill in three arrangements: alone, one hour before an exenatide injection, and thirty minutes after one.

Taken an hour before the injection, nothing substantive changed. Taken thirty minutes after, single-dose peak concentrations fell 46% (90% CI, 42 to 51) for the estrogen and 41% (35 to 47) for the progestin, with peaks arriving three to four hours late. Across repeated daily dosing the estrogen peak was still 45% lower. Bioavailability was not altered, and daily trough concentrations did not fall.[8] The authors’ own conclusion is a scheduling note rather than an alarm: the reduction is of limited importance given unchanged bioavailability, but for drugs that depend on a threshold concentration, take them at least an hour before the injection.

That is the whole mechanism in one experiment. A slowed stomach moves when a pill arrives, not how much of it arrives. The same pill, the same drug, the same person — a ninety-minute difference in timing turns a 46% effect into none.

So why does tirzepatide get a rule?

Because its numbers are larger, and because the falls in peak concentration are accompanied by falls in total exposure rather than rises. In the single published study, a 5 mg dose given with a combined pill cut peak concentrations of the estrogen and both progestin components by more than half, cut total exposure by about a fifth for each, and delayed the peak by up to four and a half hours.[9] A 2024 review that compared all six available trials side by side reached the obvious conclusion: one agent showed a statistically significant reduction, the other five did not, and the difference tracks the steeper delay in gastric emptying after a first or an increased dose.[10]

That review also names the real driver, which is the dosing calendar rather than the molecule. Tirzepatide’s effect on emptying is largest at a dose the body has not met before and fades as the dose repeats, which is why the labeled precaution is four weeks long and attached to each step of the escalation schedule rather than to the whole course. A reader who has been at the same dose for six months is not in the same situation as one who stepped up last Tuesday, even though they are taking the same two medicines.

A lower peak is not a failed pill, and nobody has counted one

The number most often quoted from the tirzepatide study is the largest one, and it is the least informative. A 2026 review makes the point carefully: exposure to the active progestin metabolite fell far less than its peak did, which demonstrates that a fall in peak concentration cannot be equated with reduced effectiveness.[11] Combined pills work by maintaining a concentration across the day, not by hitting a spike each morning, and trough concentrations are the quantity that speaks to that. Exenatide’s troughs did not fall.[8]

Here is the sentence that belongs at the end of every article on this subject and appears in almost none of them. No study of any drug in this class has measured a contraceptive outcome. Not a pregnancy rate, not an ovulation rate, not breakthrough bleeding. Every figure above is a plasma concentration in a small group of volunteers, several of them postmenopausal and therefore incapable of demonstrating the endpoint that matters. The tirzepatide instruction is a precaution derived from a pharmacokinetic result, which is a defensible thing for a regulator to require and a different thing from evidence that pills have failed.

Two further gaps deserve naming. Progestogen-only pills, which a great many people take, have never been studied with any of these drugs, and the 2026 review states in terms that findings from one contraceptive formulation cannot be carried over to a structurally different one.[11] And the labeled reassurance about non-oral methods rests on mechanism rather than measurement: an implant or an intrauterine device never passes through a stomach, so there is nothing for a slowed stomach to do, but no trial confirmed it.

What this changes for someone buying online

The practical difference between the two molecules is real and it is narrow. Someone on compounded or branded tirzepatide who also takes a daily pill has a labeled precaution that reopens at each dose increase. Someone on semaglutide has a label that lists their contraceptive among the drugs nothing happened to. Choosing between the molecules on any other basis is covered in the molecule comparison, and this is one more line in that ledger rather than a reason to reconsider either.

One caution specific to this market. Compounded semaglutide and compounded tirzepatide are not approved by the FDA and are not reviewed by the agency for safety, effectiveness or quality before a pharmacy ships them, which means the labeled contraception text above travels with the branded product and not with a compounded vial. The pharmacology is the same molecule; the document that would have told a buyer about it is not in the box. Nothing here is a reason to start, stop or change any medication on your own, and how claims on this site are established is set out in the methodology.

Frequently asked

Does the semaglutide label say anything about birth control pills?
No. The word "contraceptive" does not appear in the Wegovy label at all. What appears instead is ethinyl estradiol and levonorgestrel inside a list of drugs whose pharmacokinetics showed no clinically significant differences alongside semaglutide. The general caution about swallowed medicines is there, but it attaches to drugs with a narrow therapeutic index, which a combined pill is not.
What exactly does the tirzepatide label instruct?
It carries a section headed Contraception which says that use of the drug may reduce the efficacy of oral hormonal contraceptives because gastric emptying is delayed, that the delay is largest after the first dose and diminishes over time, and that patients should switch to a non-oral method or add a barrier method for four weeks after starting and four weeks after each dose escalation. Both the weight-management and the diabetes product carry it.
Were the other GLP-1 drugs ever tested against a pill?
All of them were. Weekly semaglutide met the bioequivalence criterion for the estrogen and came in 20% higher for the progestin. Oral semaglutide produced ratios of 1.06 for both. Dulaglutide produced lower peaks with similar total exposure and a published conclusion of no dose adjustment. Liraglutide lowered both peaks about 12% and raised progestin exposure 18%, and still carries no instruction.
Does it matter when I take the pill relative to the injection?
In the one study that varied the timing deliberately, it mattered more than the drug did. With exenatide, a pill taken an hour before the injection showed no substantive change; the same pill taken thirty minutes after showed peak concentrations down 46% and 41%. Bioavailability and daily trough concentrations were unchanged either way, and the authors' advice was a scheduling note rather than a warning.
Has anyone shown that a contraceptive actually failed?
No. Every published figure in this area is a plasma concentration measured in a small group of volunteers, many of them postmenopausal. No study of any drug in this class has counted a pregnancy, an ovulation or breakthrough bleeding. The tirzepatide precaution is derived from a pharmacokinetic result, which is a reasonable basis for a label instruction and is not the same as a measured failure rate.
What about progestogen-only pills, implants or an IUD?
Progestogen-only pills have never been studied with any of these drugs, and a 2026 review states plainly that a result from one contraceptive formulation cannot be extrapolated to a structurally different one. For methods that are not swallowed, both tirzepatide labels say they should not be affected — which follows from the mechanism, since nothing that bypasses the stomach can be delayed by one, though no trial tested it directly.

Sources

  1. [1] Eli Lilly and Company (2026). ZEPBOUND (tirzepatide) injection — Use in Specific Populations 8.3: Contraception, and Drug Interactions 7.2: Oral Medications DailyMed, U.S. National Library of Medicine. Source
  2. [2] Eli Lilly and Company (2026). MOUNJARO (tirzepatide) injection — Use in Specific Populations 8.3: Contraception DailyMed, U.S. National Library of Medicine. Source
  3. [3] Novo Nordisk Pharmaceutical Industries LP (2026). WEGOVY (semaglutide) injection — Clinical Pharmacology 12.3: Other Drugs, and Drug Interactions 7.2 DailyMed, U.S. National Library of Medicine. Source
  4. [4] Kapitza C, Nosek L, Jensen L, et al. (2015). Semaglutide, a once-weekly human GLP-1 analog, does not reduce the bioavailability of the combined oral contraceptive, ethinylestradiol/levonorgestrel. J Clin Pharmacol. PMID 25475122
  5. [5] Jordy AB, Albayaty M, Breitschaft A, et al. (2021). Effect of Oral Semaglutide on the Pharmacokinetics of Levonorgestrel and Ethinylestradiol in Healthy Postmenopausal Women and Furosemide and Rosuvastatin in Healthy Subjects. Clin Pharmacokinet. PMID 33782832
  6. [6] de la Peña A, Cui X, Geiser J, et al. (2017). No Dose Adjustment is Recommended for Digoxin, Warfarin, Atorvastatin or a Combination Oral Contraceptive When Coadministered with Dulaglutide. Clin Pharmacokinet. PMID 28357715
  7. [7] Novo Nordisk Pharmaceutical Industries LP (2026). SAXENDA (liraglutide) injection — Clinical Pharmacology 12.3: Oral Contraceptives DailyMed, U.S. National Library of Medicine. Source
  8. [8] Kothare PA, Seger ME, Northrup J, et al. (2012). Effect of exenatide on the pharmacokinetics of a combination oral contraceptive in healthy women: an open-label, randomised, crossover trial. BMC Clin Pharmacol. PMID 22429273
  9. [9] Eli Lilly and Company (2026). ZEPBOUND (tirzepatide) injection — Clinical Pharmacology 12.3: Drug Interaction Studies, Oral Contraceptives DailyMed, U.S. National Library of Medicine. Source
  10. [10] Skelley JW, Swearengin K, York AL, et al. (2024). The impact of tirzepatide and glucagon-like peptide 1 receptor agonists on oral hormonal contraception. J Am Pharm Assoc (2003). PMID 37940101
  11. [11] Viana DPDC, Invitti AL, Jacobsen L, et al. (2026). Tirzepatide and oral progestogens: A hypothesis-generating review of a biologically plausible pharmacokinetic interaction in gynaecologic disease control and menopausal hormone therapy. Maturitas. PMID 42721915

More in Safety