This is the one I have started raising unprompted, because in several years I have met very few patients who were told it.
The mechanism, which makes it easy to remember
These drugs slow gastric emptying. An oral contraceptive taken into a stomach that is emptying slowly is absorbed differently, and the peak drug level can fall below what is needed.
The effect is largest when gastric emptying changes most — which is immediately after starting, and immediately after each dose increase. Once you have been at a stable dose for some weeks, the stomach has adapted and the interaction diminishes.
That is why the guidance attaches to changes rather than to the whole treatment period.
The rule
| When | What to do |
|---|---|
| First four weeks after starting tirzepatide | Non-oral method, or add a barrier method |
| Four weeks after every dose increase | Non-oral method, or add a barrier method |
| Stable dose, no recent change | Oral contraception as normal |
There are usually five dose increases in a full titration. That is five separate four-week windows, spread across roughly six months, each one easy to lose track of.
This is precisely the sort of thing that goes wrong from ordinary forgetfulness rather than from carelessness — the increase happens, the calendar moves on, and nobody counts the weeks. If you are logging your dose escalations anywhere already, the four-week window falls out of the record automatically; Zenday is what I suggest for it, because the dates you need are the dates it is already keeping.
The other half: fertility returning
The interaction is only part of why unexpected pregnancies happen on these drugs.
Obesity and PCOS are common causes of anovulation. Substantial weight loss frequently restores ovulation — sometimes within months, and often in someone who has spent years being told conception would be difficult and has organised their contraception around that belief.
So two things move at once: a person becomes more fertile than they were, while their contraceptive method may be working less well than it was. That is not a coincidence to be managed casually.
Somebody who was told at twenty-eight that they would probably struggle to conceive, and who has arranged fifteen years around that sentence, deserves to be told when it stops being true.
For people with polycystic ovary syndrome this is not an edge case but the expected outcome of treatment working — see GLP-1s and PCOS. The broader changes to bleeding and cycle length are in what happens to your cycle.
If you are planning a pregnancy
These medications are not recommended in pregnancy, and the guidance is to stop before conception rather than on a positive test.
The recommended interval before stopping differs by product and relates to how long the drug persists in the body — commonly around two months for semaglutide. Get the specific interval for your medication from your prescriber, and plan it deliberately.
There is also a conversation worth having about what happens next: appetite returns, weight often regains, and pregnancy is a poor time to be fighting that. It is a solvable problem when it has been anticipated and a distressing one when it has not.
What to actually do
If you use oral contraception and take tirzepatide, get a plan in writing from your prescriber covering the start and every increase. If you are on semaglutide, the contraceptive interaction is not described in the same way, but the fertility point still applies.
And if you have been assuming you cannot easily get pregnant, treat that assumption as something to revisit rather than to rely on.