Alcohol, Caffeine and Smoking When Trying to Conceive

The honest picture, with numbers

These three get bundled together in every leaflet, which is unhelpful, because the evidence behind them is completely different in strength. One has a large measured effect. One has a small measured effect. One has, for the outcome you are asking about, essentially none.

Smoking has the strongest evidence. ASRM's committee opinion puts the odds ratio for infertility in smokers at 1.60 (95% CI 1.34 to 1.91), and the odds of taking longer than 12 months to conceive at 1.42 (95% CI 1.27 to 1.58) — roughly 54% more likely to experience a conception delay of over a year. Smokers need nearly twice the number of IVF cycles to conceive as non-smokers, the odds ratio for live birth in IVF for smokers is 0.59 to 0.66, and menopause arrives 1 to 4 years earlier in a dose-dependent way. NG257 states that smoking is likely to reduce fertility and that passive smoking is likely to affect the chance of conceiving; ASRM notes the effects of passive exposure were only slightly smaller than for active smoking by either partner.

Alcohol has a smaller, real effect. A dose-response meta-analysis of 19 studies and 98,657 women found a pooled fecundability estimate of 0.87 (95% CI 0.78 to 0.95) for drinkers versus non-drinkers — 0.89 (95% CI 0.82 to 0.97) for light drinkers at 12.5 g of ethanol a day or less, and 0.77 (95% CI 0.61 to 0.94) above that. The dose-response relationship was linear, with a risk ratio of 0.98 (95% CI 0.97 to 0.99) per additional 12.5 g a day.

Caffeine has essentially no effect on conceiving. NICE states there is no consistent evidence of an association between caffeinated drinks — tea, coffee, energy drinks and colas — and fertility problems. A dose-response meta-analysis found no clear association between coffee or caffeine and fecundability or time to pregnancy, and little relation between fecundability and intake of 300 mg a day or more compared with under 100 mg.

What actually affects it

Smoking, and why it is the priority

Of everything in this article, stopping smoking is the intervention with the largest effect and the clearest mechanism. ASRM notes that chemicals in cigarette smoke including nicotine, cyanide and carbon monoxide speed up the rate of egg loss, and that even IVF may not fully overcome the effect. Male smokers show lower sperm counts, lower motility and more abnormally shaped sperm. NG257 recommends that clinicians encourage and support people to stop smoking in line with NICE's tobacco guideline — support, not instruction. On e-cigarettes, ASRM's position is that human data remain limited and that use may be detrimental to fetuses, with concerns about detected metals and harmful carbonyl compounds; it is not established as a safe substitute in pregnancy.

Alcohol: the guideline numbers

NG257 advises that drinking no more than 1 or 2 units of alcohol once or twice per week and avoiding episodes of intoxication reduces the risk of harming a developing fetus, while noting that under the Chief Medical Officer's guidelines the safest approach is to avoid alcohol altogether. For male partners, NICE states that excessive alcohol intake is detrimental to semen quality, but that drinking within the UK weekly guidance of 14 units, spread across several days — for example up to 2 units a day — is unlikely to affect semen quality.

The meta-analysis is worth reading carefully because of what it did not find. Split by beverage type, the estimates were 0.98 (95% CI 0.85 to 1.11) for wine, 1.02 (95% CI 0.99 to 1.05) for beer, and 0.92 (95% CI 0.83 to 1.01) for spirits — none of them clearly showing an effect on their own. The signal is in total ethanol, not in which drink.

Caffeine: conception versus pregnancy

The distinction that gets lost is between trying and being pregnant. For conceiving, the evidence does not support cutting caffeine. For an established pregnancy, a dose-response meta-analysis of observational studies found coffee or caffeine consumption associated with increased risk of spontaneous abortion — relative risk 1.37 (95% CI 1.19 to 1.57) at 300 mg a day and 2.32 (95% CI 1.62 to 3.31) at 600 mg a day. These are observational data with well-known confounding, but they are the reason pregnancy advice and preconception advice differ. A reasonable reading: coffee is not the reason you have not conceived, and moderating it once pregnant is a separate decision.

What the guidance conspicuously does not include

There is no NICE recommendation about stress or relaxing, because there is no guidance-grade evidence to support one. NICE does state that the effectiveness of complementary therapies for fertility problems has not been properly evaluated. And on marijuana, ASRM's 2024 position is that use has not been consistently associated with male or female fecundity, time to pregnancy, reproductive hormone levels, semen parameters or ART outcomes — which is a statement about the weakness of the evidence, not a recommendation to use it.

What to do next

1. If either of you smokes, put everything else second. Ask for structured stop-smoking support rather than attempting it unaided; NG257 points to NICE's tobacco guideline for exactly this.
2. Address passive exposure too. The measured effect of passive smoking is only slightly smaller than active smoking by either partner.
3. Set an alcohol level you can actually keep to. The CMO position is that not drinking is safest; NICE's operational figure is no more than 1 or 2 units once or twice a week with no episodes of intoxication.
4. Stop rationing coffee for fertility reasons. Reconsider intake if and when you are pregnant, on different evidence.
5. Check what else is in your day — NG257 asks about GLP-1 agonists, testosterone-replacement therapy, finasteride, NSAIDs, vaginal lubricants, anabolic steroids and cannabis.
6. Do not spend a year optimising before being investigated. Lifestyle change and referral run in parallel, not in sequence.

When to seek help

NG257 recommends offering both partners further clinical assessment and investigation after 1 year of unprotected vaginal intercourse without conception, and referral at first presentation if the woman trying to conceive is 36 or over, or if either partner has a suspected or known clinical cause of infertility or a history of predisposing factors. ACOG advises evaluation after 12 months, or 6 months over the age of 35.

Two practical points. First, ask for stop-smoking support from a GP or pharmacy rather than treating it as a willpower problem; it is the one item here with an effect size large enough to be worth clinical help. Second, if you have already stopped smoking, reduced alcohol and cut caffeine and nothing has changed, that is not evidence you did it wrong. Population effects of this size do not determine what happens to one person in one year, and the more useful next step is investigation.