Medical disclaimer. This article is general information and does not replace medical advice. Caffeine guidance can change with your health history and your pregnancy, so check with your midwife or obstetrician about what applies to you.
Most guideline bodies land on the same ceiling for caffeine during pregnancy: under 200 mg a day. The NHS states it plainly — "you should not have more than 200mg of caffeine a day" — and it counts a cup of instant coffee at 100 mg, so on its own numbers that ceiling is about two cups.
The number is the easy part. The hard part is that 200 mg does not arrive in tidy units — it arrives as a large cold brew on the way to work, a cup of tea at your desk, a cola with lunch and two squares of dark chocolate at 9 pm. Here is the arithmetic, the drinks that quietly eat the budget, and an honest read of what the research does and doesn't show.
Where the 200 mg rule comes from
Three bodies get you to the same figure by different routes.
The NHS is the bluntest: no more than 200 mg a day, and it says that having too much caffeine "has been linked to a higher chance of miscarriage and stillbirth."
EFSA, in its 2015 Scientific Opinion on the Safety of Caffeine, puts it as a safety threshold rather than a warning: intakes up to 200 mg a day, spread through the day, "do not raise safety concerns for the foetus." For other healthy adults, EFSA's equivalent figure is 400 mg.
ACOG approaches it from the other end. Its Committee Opinion 462 concludes that "moderate caffeine consumption (less than 200 mg per day) does not appear to be a major contributing factor in miscarriage or preterm birth." It is careful about what it does not claim: "the relationship of caffeine to growth restriction remains undetermined," and no final conclusion can be drawn about whether high intake and miscarriage are correlated.
The FDA is the outlier, and worth understanding rather than misreading. It cites 400 mg a day as an amount "not generally associated with negative effects" for most adults, and then tells anyone pregnant, trying to become pregnant or breastfeeding to talk to their own clinician about whether they need to limit caffeine. It does not publish a pregnancy number of its own.
So 200 mg is not a cliff edge with harm on one side and safety on the other. It is the level below which the reassuring evidence is strongest — and, as the evidence section below shows, one of the outcomes ACOG deliberately left open does show a signal below 200 mg too. Treat it as a sensible cautious ceiling, not a guarantee.
What 200 mg looks like in real cups
This is the table the ProbaBaby iOS app's caffeine calculator uses — 14 sources, grouped as coffee, tea, cold drinks, and food and medicine.
| Source | Serving | Caffeine |
|---|---|---|
| Cold brew | 350 mL | 155 mg |
| Latte / cappuccino | 350 mL | 128 mg |
| Brewed coffee | 240 mL | 95 mg |
| Matcha | 240 mL | 70 mg |
| Espresso | 30 mL | 64 mg |
| Instant coffee | 240 mL | 62 mg |
| Black tea | 240 mL | 47 mg |
| Green tea | 240 mL | 28 mg |
| Decaf coffee | 240 mL | 5 mg |
| Energy shot | 60 mL | 200 mg |
| Energy drink | 250 mL | 80 mg |
| Cola | 350 mL | 34 mg |
| Pain reliever with caffeine | 1 tablet | 65 mg |
| Dark chocolate | 28 g | 12 mg |
Now notice how far apart the published figures are. The NHS counts a cup of instant coffee at 100 mg; the table above uses 62 mg. The NHS puts a cup of tea at 75 mg against 47 mg here, and a can of cola at 40 mg against 34 mg. The FDA, meanwhile, puts a 12 fl oz brewed coffee anywhere between 113 mg and 247 mg.
None of these is wrong. Real caffeine content varies that much with brand, brew, bean and cup size, and every organisation picks a different typical value. Treat any single number as an estimate, and if you are near the ceiling, assume the higher one.
The drinks that quietly blow the budget
Three patterns account for most accidental overshoots.
Cup size, not cup count. "Two coffees a day is fine" is true at 95 mg a mug and false at 155 mg. One large cold brew plus a latte is 283 mg — over the limit before lunch, on two drinks.
Energy shots and energy drinks. One 60 mL shot is 200 mg on the table above: the entire daily allowance in something the size of a perfume sample. Energy drinks are the widest range the FDA lists — 41 mg to 246 mg for a 12 fl oz can — so the label is worth reading rather than guessing.
Non-drinks. Some over-the-counter pain relievers include caffeine; the app counts one at 65 mg a tablet, about a third of the day's budget. Dark chocolate is modest per square but adds up across a bar.
How a caffeine calculator counts your day
The math is simple enough to do in your head while standing in a queue.
- Total = each drink's caffeine multiplied by how many you had, summed.
- Percentage = total divided by your ceiling. Against 200 mg, a 300 mg day is 150% — not "maxed out."
- Over the limit means strictly above the ceiling. Landing on exactly 200 mg is within it.
Two details matter. A percentage that stops at 100% hides the difference between a 210 mg day and a 600 mg one, so the app's meter keeps rescaling past the ceiling instead. And the ceiling itself is not a constant: in the app it moves with your life stage — 200 mg pregnant, 300 mg breastfeeding, 400 mg otherwise. To run your own day through it, use the pregnancy caffeine calculator.
What the evidence actually says
Here is where honesty matters more than a clean answer.
The observational evidence. A 2014 dose-response meta-analysis in BMC Medicine included 13 prospective studies, 9 of them with low birth weight as a yes-or-no outcome — 90,747 participants and 6,303 cases. Low-birth-weight risk rose with intake: relative risk 1.13 (95% CI 1.06 to 1.21) at 50–149 mg a day, 1.38 (1.18 to 1.62) at 150–349 mg, and 1.60 (1.24 to 2.08) at 350 mg or more. Each extra 100 mg a day — about one cup of coffee — carried a 13% higher risk. In grams rather than ratios, moderate intake tracked with a birth weight 33 g lower (95% CI 4 to 63) than the low-intake reference. The authors concluded that their findings "support recommendations to restrict caffeine intake during pregnancy to low levels."
Note where that 1.13 sits: the lowest exposure band, 50–149 mg a day, is below the 200 mg ceiling. If this association is causal, the ceiling is not a safe harbour.
The experimental evidence. A 2015 Cochrane review looked for trials that actually randomised women to more or less caffeine. It found essentially one usable study, run in Denmark: 1,197 women enrolled before 20 weeks, given caffeinated or decaffeinated instant coffee. Cutting about 182 mg a day in the second and third trimesters produced a mean birth-weight difference of 20 g (95% CI −48.68 to 88.68) — nothing — and no effect on preterm birth (RR 0.81, 95% CI 0.48 to 1.37), on evidence the reviewers graded low quality. Cochrane's verdict: there is insufficient evidence to confirm or refute the effectiveness of caffeine avoidance.
Why both can be true. The meta-analysis authors name the problem themselves. Nausea, vomiting and aversions to smells and tastes are more common in healthy pregnancies; women with healthy pregnancies are therefore more likely to go off coffee without deciding to. So "drank less caffeine" and "had a better outcome" can travel together without one causing the other — which is exactly the kind of link a randomised trial exists to break. The trial that exists is small and its confidence interval is wide, which is why Cochrane rated it low quality rather than reassuring.
The reasonable read: 200 mg is a sensible cautious ceiling, the evidence below it is imperfect rather than clean, and the cup you drank before you knew you were pregnant is not worth losing sleep over. If you are still working out how far along you are, our guide to how due date calculators work covers the dating side.
Why the same latte hits harder later on
Caffeine's half-life in a healthy adult is about 4 to 5 hours. In pregnancy, CYP1A2 — the liver enzyme responsible for roughly 95% of caffeine's breakdown — becomes less active, and the effect grows as pregnancy advances. A 2025 review in Nutrients puts the half-life at up to about 15 hours in the late gestational trimester.
So the 3 pm coffee that used to be gone by bedtime is still substantially in your system at 3 am in week 34. Caffeine also readily crosses the placenta, and the enzyme that breaks it down is absent in both the placenta and the fetus.
Practically: the total still matters, but so does timing. If afternoon coffee started wrecking your sleep in the third trimester and you blamed pregnancy insomnia, it may have been the coffee. Swapping that cup for water helps twice over — our pregnancy water intake calculator shows what your daily target looks like.
Breastfeeding, and twins
Two follow-up questions come up constantly.
Breastfeeding. The ceiling goes up, not away. The CDC describes a low-to-moderate amount as about 300 mg a day or less — two to three cups of coffee. It links very high intakes, around 10 cups of coffee a day or more, to irritability, poor sleeping patterns, fussiness and jitteriness in the infant, and notes that preterm and younger newborns break caffeine down more slowly, so their mothers might consider consuming even less.
Twins. The limit does not change. No guideline body sets a separate caffeine ceiling for twins or higher-order multiples, so it is the same 200 mg. Where carrying multiples genuinely changes the target is elsewhere — weight gain, for instance, where the IOM publishes provisional twin totals — not caffeine.
Cutting down without the headache
Going from 400 mg to nothing overnight can buy you a withdrawal headache in your first trimester, which is a miserable trade.
- Cut the size before the count. A 240 mL brewed coffee instead of a 350 mL cold brew saves 60 mg on the table above, with no ritual lost.
- Blend your cup. Half regular, half decaf keeps the taste and the habit at roughly half the caffeine.
- Move the last cup earlier. With a half-life that can reach about 15 hours late on, a 10 am coffee is a very different drink from a 4 pm one.
- Read energy drink labels. They are the widest range on the FDA's list — 41 mg to 246 mg for the same 12 fl oz.
How ProbaBaby helps
The ProbaBaby iOS app's caffeine tool works from a fixed catalogue of 14 sources across coffee, tea, cold drinks, and food and medicine — you tap a tile per serving instead of typing milligrams, and servings render in your own measurement system rather than in a unit baked into an English label.
A life-stage picker sets the ceiling: 200 mg pregnant, 300 mg breastfeeding, 400 mg otherwise. The result is a meter segmented by source, so you can see which drink used the most of the day, and it keeps rescaling past the ceiling instead of stopping at 100%.
Set the stage to pregnant and a multiples toggle appears — and it deliberately does not move the number. It shows a note saying that no guideline body sets a separate caffeine limit for twins or higher multiples, because that question deserves a direct answer rather than an invented one. The app also names the guidelines behind its ceilings on screen — ACOG's Committee Opinion 462, EFSA's 2015 opinion, the CDC and the FDA — so you can check the number rather than take it on trust.
And it carries the caveat this whole article has been circling: these are typical averages, and real caffeine content varies a lot with brand, brew and cup size.