Probably not, on the one study that measured it. Semaglutide met bioequivalence criteria for ethinylestradiol exposure in a 43-woman pharmacokinetic study, and levonorgestrel exposure ran 20% higher. Higher exposure is not reduced protection.
The question matters more than usual here because the population overlaps. These drugs are prescribed heavily to women of reproductive age, including for PCOS, ahead of the indication. And if a pregnancy is planned rather than prevented, stopping beforehand did not produce the benefit expected.
That study enrolled postmenopausal women with type 2 diabetes. It measured drug exposure, not pregnancies. Nobody has run the study that would answer the question directly.
A separate register finding is often misread as an answer to this. Among 249,634 Danish participants, every pattern of prior hormonal contraception was associated with later starting semaglutide[6]. The range ran from 1.42 for the combined pill to 2.11 for women who had used several types. That is an association between two prescriptions. It does not show contraception causes weight gain, and its authors do not claim it does.
Two questions come up constantly here and are answered more confidently than the evidence permits. Whether a GLP-1 interferes with oral contraception, and what it does to libido and sexual function. Both deserve a direct answer. In both cases the direct answer includes naming what has not been studied, which is the standard set out in the methodology.
Oral contraception: measured, but measured narrowly
The specific question has been tested. Forty-three postmenopausal women with type 2 diabetes took a combined oral contraceptive daily for eight days. That ran before and during semaglutide treatment at a 1.0 mg steady state[1]. The pill was ethinylestradiol 0.03 mg with levonorgestrel 0.15 mg, and the usual dose escalation came first.
The bioequivalence criterion was met for ethinylestradiol exposure, with a ratio of 1.11 (90% CI 1.06 to 1.15) against the prespecified 0.80 to 1.25 limits. Levonorgestrel exposure was 20% higher at semaglutide steady state, a ratio of 1.20 (90% CI 1.15 to 1.26). Peak concentration for both sat inside the bioequivalence range[1]. The conclusion the paper draws is the one in its title: semaglutide does not reduce the bioavailability of that combined oral contraceptive.
The broader interaction question has been reviewed systematically. A review covered 22 reports and six prescribing sheets on injectable GLP-1 receptor agonists given with oral medications[2]. Treatment left peak concentration unaffected or reduced it, and delayed the time to peak. That held across drugs of differing solubility and permeability, contraceptive pills among them. The review reports that these agents did not exert clinically significant changes in most of the cases examined[2]. A separate pharmacokinetic review of the approved GLP-1 agents and the dual GLP-1/GIP agonist covers the same interaction surface in more detail[3].
Two practical points follow, and neither is a reassurance. Delayed gastric emptying is the mechanism, so a drug with a narrow therapeutic index is the case to raise with a prescriber[2]. Severe vomiting is common enough on these drugs to appear in every tolerability cohort. It can interfere with the absorption of an oral pill for reasons no pharmacokinetic study captures.
Testosterone and sexual function in men: indirect, and improving
A systematic review and meta-analysis examined the effect of GLP-1 receptor agonists on testicular function in overweight and obese men. Seven studies, with 680 participants in total, entered the quantitative analysis, and treatment produced a significant increase in total serum testosterone[4]. The review frames it as weight loss inducing an indirect positive effect on testicular function. Recent evidence also suggests a possible direct influence on gonadal function[4].
Erectile function specifically is less settled. A review describes erectile dysfunction as a common and frequently underrecognized microvascular complication of diabetes[5]. Studies of various antidiabetic therapies have produced inconsistent results. Experimental and clinical research increasingly supports a positive effect for GLP-1 receptor agonists, while preliminary reports have raised concerns in the other direction.
Women, libido and the gap nobody fills
The corresponding question in women is thinner still. The testosterone meta-analysis above is in men, and no equivalent body of work exists for female sexual function on these drugs. Statements about libido in women on a GLP-1 are being extrapolated from weight change, from the male hormone data, or from nothing at all. That should be said plainly rather than filled with a plausible sentence.
The honest summary across this whole cluster is short. Contraceptive drug exposure has been measured once, narrowly, and looked reassuring. Male testosterone rises, in a small meta-analysis, probably through weight loss. Erectile function is contested. Female sexual function is unstudied. Anything more definite than that is coming from somewhere other than the literature. What each seller charges is a separate matter, recorded across the seller reviews.