Contraception: when oestrogen is the wrong answer

Clinical Rounds, the GPAtlas podcast. Sexual Health. 18 minutes. Published 31 August 2026. Free to listen.

Clinical source: our own Ocean article "Contraceptive Method Selection", last reviewed 5 August 2026.

Transcript

Sarah: Hello, and welcome to GPAtlas Clinical Rounds. I'm Sarah, and this is the podcast where we take one thing you'll genuinely see this week and work out what to do with it.

Ben: And I'm Ben. Today, choosing a contraceptive method, and one question in particular that changes the answer completely.

Sarah: Chloe is twenty-six and she wants to restart the pill. She was on Microgynon years ago, got on fine with it, and she's come in for a repeat, basically. Ten-minute appointment, and she's expecting it to take two. And somewhere in that conversation I have to ask about her headaches, because she gets migraines, and the answer to one question decides whether the prescription I'm about to write is safe or dangerous.

Ben: Three things. The question that has to be asked before any combined method. What the UKMEC numbers actually mean when you're sitting there. And the situations where the usual answer stops working: emergency contraception, interacting drugs, and the woman in her forties who thinks she's finished.

Sarah: So, the question. Does she get any warning before the headache. Visual disturbance, zigzags, a blind spot, numbness, pins and needles, speech trouble. Because our article is blunt about this: the single most common high-stakes error is overlooking an absolute contraindication to combined hormonal contraception, especially migraine with aura.

Ben: And the part that gets missed is that severity is irrelevant. The presence of aura, not its frequency or severity, is what makes combined contraception a UKMEC four, because the stroke risk is driven by the oestrogen, not by how bad the headaches are. So a woman who gets one mild aura a year is in the same category as one who's disabled by them monthly. That's counter-intuitive and it's the whole point.

Sarah: And it runs forwards too, which is the bit I've had to train myself on. If a woman develops new aura while she's on the pill, you stop the oestrogen and switch to a progestogen-only or non-hormonal method. That's not a review-in-a-month conversation. Somebody ringing to say her migraines have changed and she's now getting funny lights beforehand is a same-day prescribing decision.

Ben: Aura is a yes or no question. Not a how bad question.

Sarah: Let's do UKMEC properly, because the numbers get quoted and not always understood. It grades safety, not effectiveness. One, no restriction. Two, benefits outweigh risks. Three, caution, and our article's wording is expert judgement or specialist input, use only if better options are unacceptable. Four, unacceptable risk, don't use it.

Ben: And there's a trap in the arithmetic, which our article states outright: the categories grade safety, not efficacy, and are not additive. So two threes do not make a four. Two separate category-three conditions is a conversation and a judgement, not an automatic ban. People add them up, and they shouldn't.

Sarah: The numbers worth memorising for combined contraception, because they come up weekly. Smoking fifteen or more a day at thirty-five or over is a four. Under fifteen a day at that age is a three. Stage two or three hypertension, so a hundred and sixty or above, or a hundred or above diastolic, is a four. And a BMI of thirty-five or over is a three.

Ben: And blood pressure is mandatory before any combined method, with a technique detail: repeat it if the first reading is a hundred and forty over ninety or higher, and record the lower of the last two. What you don't need is an examination. Our article says no routine breast or pelvic examination is needed to start hormonal contraception, and a careful history is the main assessment tool.

Sarah: Threes do not add up to a four. And you do not need to examine her.

Ben: Emergency contraception, where the hierarchy is not the one most patients expect. The copper coil is the most effective, within five days of unprotected sex or of the earliest estimated ovulation, and crucially it's unaffected by BMI or by enzyme-inducing drugs. That's the one to offer first, even though the conversation is harder.

Sarah: And if it's a tablet, there's a dose rule that gets missed. Levonorgestrel one point five milligrams within seventy-two hours, but you double it to three milligrams if the BMI is over twenty-six, or the weight is over seventy kilos, or she's on an enzyme-inducer. Over seventy kilos is a lot of women. If you're not asking, you're under-dosing.

Ben: And then the timing rule that only works in one direction, which is a genuinely easy mistake. After ulipristal, you wait five days before starting or restarting hormonal contraception, because the progestogen reduces its effectiveness. After levonorgestrel, hormonal contraception can begin immediately. So the two emergency pills have opposite follow-on rules, and mixing them up either wastes the emergency dose or leaves her unprotected for five days.

Sarah: Ulipristal, wait five days. Levonorgestrel, start now.

Ben: Interacting drugs, and here the answer is unusually clean. Enzyme-inducers, so rifampicin and rifabutin, carbamazepine, phenytoin, topiramate at two hundred milligrams a day or more, efavirenz and St John's wort, reduce the effectiveness of combined contraception, the progestogen-only pill and the implant. But not the injectable, and not the coils. So when an interacting drug is unavoidable, the injectable or an intrauterine method is the preferred option.

Sarah: And the reverse of that, which is us over-warning. Most antibiotics do not interact, unless they're enzyme-inducers or they cause vomiting or diarrhoea. That advice about a week of extra precautions with every course of amoxicillin has been wrong for years, and patients still get told it, and some of them stop trusting the rest of what we say because of it.

Ben: One more that our article flags and I'd not have thought to check: the newer drospirenone- only pill is an antimineralocorticoid, so there's a risk of hyperkalaemia with ACE inhibitors, ARBs or potassium-sparing drugs. Worth a look at the repeat list before you start it in an older woman.

Sarah: Enzyme-inducers spare the injection and the coils. Antibiotics mostly do not matter.

Ben: And the last group, the women in their forties and fifties, where the commonest error is stopping too early. Our article is direct: a fifty-year-old is still potentially fertile, and she should switch to a progestogen-only or non-hormonal method, not simply stop.

Sarah: And the numbers for actually stopping. Continue for two years after the last period if she's under fifty, one year if she's fifty or over. And all methods can stop at fifty- five, when natural conception is assumed lost even if she's still bleeding. Fifty-five is the clean answer and most people don't know it.

Ben: FSH has one narrow use here and it is easy to misuse. In a woman of fifty or over on a progestogen-only pill, check it on two occasions one to two months apart, and if both are above thirty, continue for one more year and then stop. But it's unreliable on combined contraception because that suppresses it, and erratic on the injectable. So focus on symptoms, not numbers. And separately: hormone replacement does not prevent pregnancy.

Sarah: Fifty-five is when it stops. And HRT is not contraception.

Sarah: Two things about the coil that come up constantly, and both have changed. First, a negative sexually transmitted infection screen is no longer required before fitting one. People still delay fits for a screen result. And second, the counselling, which is where most coils fail. Irregular bleeding is common, and it's the main reason for removal, so you counsel that upfront.

Ben: And that upfront word is doing the work. A woman warned in advance that she may bleed unpredictably for months usually rides it out. The same woman, unwarned, comes back at eight weeks convinced something is wrong and has it taken out. Same device, same bleeding, completely different outcome, decided entirely by a sentence at the fitting.

Sarah: And one safety-net that has to be automatic with any coil user. If she has pain or abnormal bleeding and a positive pregnancy test, that's a same-day assessment to exclude an ectopic. Not a routine appointment. That's the one to have hard-wired.

Ben: Warn about the bleeding before you fit it, not after.

Sarah: On effectiveness, there's a gap worth being honest about, because it drives method choice more than anything we say about hormones. Our article puts it plainly: the pill is more than ninety-nine per cent effective in theory, but around ninety-one per cent in real life. That difference is entirely about remembering to take it, and it's the strongest argument for a long-acting method there is.

Ben: And where risk actually sits with the combined pill is not where people assume. When it's appropriate, the progestogen drives the VTE risk more than the dose of oestrogen does. So switching someone to a lower-oestrogen pill because you're worried about clots may not be doing what you think it's doing.

Sarah: And the pill holiday, which patients propose and we sometimes agree to. The risk is highest in the first few months of use, and after restarting following a break of four weeks or more. So a break isn't a rest from risk, it's a reset back to the riskiest phase. That's genuinely counter-intuitive and worth saying kindly, because it's usually suggested by someone trying to be careful.

Ben: A break is not a rest. It is a reset to the riskiest few months.

Sarah: One group where the answer is simple and absolute. Current breast cancer makes every hormonal method a UKMEC four. All of them. And the copper coil is category one, so that's the safe choice. It's a rare conversation but it's one where the reflex to reach for a progestogen-only option, because it feels like the gentle one, is wrong.

Ben: And the general principle our article gives, which is the one to hold when you're uncertain in the room: when in doubt about oestrogen, a progestogen-only or intrauterine method is almost always safe. That's the fallback. If you can't remember which category her condition sits in, you don't have to guess, because there's a direction to move in.

Sarah: When oestrogen is doubtful, progestogen-only or intrauterine is almost always safe.

Sarah: Let me go back to Chloe, because I want to finish where we started. She came in for a repeat and she's expecting two minutes. And what's actually happened is that one question about her headaches has taken the combined pill off the table entirely, and now I have to tell her that, in a way that doesn't sound like I've decided she's ill.

Sarah: And the way I do it is to be honest that this is about stroke risk and oestrogen, not about her migraines being bad. Because otherwise what she hears is that her headaches are worse than she thought, and she goes away frightened about the wrong thing. The aura isn't a sign her migraines are dangerous. It's a sign that adding oestrogen would be.

Sarah: And then the good news, which is genuine. Almost everything else is open to her. A progestogen-only pill, the implant, the injection, either coil. Our article's own position is that when oestrogen is doubtful, progestogen-only or intrauterine is almost always safe. So it's a narrowing of one option, not a closing of the field, and if you frame it that way the consultation lands completely differently.

Ben: And that's worth generalising. Most of what we've talked about today is a set of no's: no combined pill with aura, no adding up of categories, no stopping at fifty, no assuming antibiotics interact. But the practical shape of it is that there's nearly always a safe method available, and the skill is knowing which question rules which one out.

Sarah: Narrowing one option is not closing the field. Say that out loud.

Ben: One more group worth naming, because the answer is easy to get wrong in the direction of caution. Postpartum and breastfeeding women. Our article's category framework still applies, and the same principle holds: when oestrogen is doubtful, progestogen-only or intrauterine is almost always safe.

Ben: And the general shape of UKMEC is worth ending on, because it's designed to be used rather than memorised. It grades safety, not effectiveness. The categories are not additive. And a three means expert judgement or specialist input, use only if better options are unacceptable, which is a real option rather than a soft no.

Sarah: So the practical version, when you're uncertain in the room and don't have the tables in front of you: ask about aura, take a blood pressure, check the drug list for enzyme- inducers, and if any of that is doubtful, move away from oestrogen rather than away from contraception. That's four things and it covers most of the risk.

Ben: Four questions. Aura, blood pressure, drug list, then move off oestrogen.

Ben: Exam corner. A thirty-one year old woman on carbamazepine for epilepsy asks for contraception. She has no other medical problems and would prefer not to have a coil. Which is the most appropriate option? A, the combined pill. B, the progestogen-only pill. C, the injectable. Or D, the implant.

Ben: It's C, the injectable. Carbamazepine is an enzyme-inducer, and enzyme-inducers reduce the effectiveness of combined contraception, the progestogen-only pill and the implant, but not the injectable or the intrauterine methods. So A, B and D are all undermined by her carbamazepine. The coil would be equally reliable, but she's declined it, and the injectable is the remaining method that the interaction does not touch. And worth saying: this is exactly why the question about her other medicines has to come before the question about her preference.

Sarah: So. Three things for Monday. One. Before any combined method, ask about aura specifically, and remember it's a yes-or-no question, not a how-bad question. Two. Over seventy kilos or a BMI over twenty-six means you double the levonorgestrel. And three. Nobody stops contraception at fifty. Two years if she's under fifty, one year if she's over, and everything stops at fifty-five.

Sarah: One thing to reflect on, if you're logging this. Think about how you ask the migraine question. Do you ask if she gets migraines, or do you ask what happens just before one starts? Those get different answers. The full transcript and the references are on the episode page. Ben and I are synthetic voices. The medicine isn't. See you next week.

More from Clinical Rounds

Every episode of Clinical Rounds, free to listen with a full transcript.