Does Fertility Acupuncture Work? What the Trials Found

No, on the evidence available. The trial that tested this most directly randomised 848 women having a fresh IVF cycle at 16 centres in Australia and New Zealand to real acupuncture or to sham needling placed away from the acupuncture points. Live births occurred in 74 of 405 women given acupuncture and 72 of 404 given sham — 18.3% against 17.8%, a difference of 0.5% with a confidence interval running from -4.9% to 5.8% (Smith 2018, JAMA). The interval includes zero, and it is narrow enough to rule out a large effect in either direction.

The 2013 Cochrane review reached the same place from 20 randomised trials: "There is no evidence that acupuncture improves live birth or pregnancy rates in assisted conception."

That is the answer to the question most people are actually asking. The rest of this page is about why the internet says otherwise, what acupuncture might still be worth to you, and what to do with the money and the weeks if you decide against it.

Because of what they compared it against. This is the single most useful thing to understand before you pay for anything, and it is not complicated.

A 2025 meta-analysis of 11 randomised trials and 3,561 women ran both comparisons side by side. Against a control group given no treatment, acupuncture around embryo transfer raised the clinical pregnancy rate: risk ratio 1.25, 95% confidence interval 1.05 to 1.50, p=0.013. Against a sham — needling that looks and feels like the real thing but is placed wrongly or does not penetrate — the same analysis found risk ratio 1.01, 95% CI 0.87 to 1.17, p=0.907. Nothing. And on live birth, the outcome that matters, it found risk ratio 1.01, 95% CI 0.88 to 1.15, p=0.930 (Wang 2025, Frontiers in Reproductive Health).

A trial against no treatment cannot separate the needles from everything else in the appointment: lying still for half an hour, being asked how you are, expecting it to help, and knowing you were the one who got the real thing. When those are held constant by a sham, the difference disappears. That is an interpretation of the contrast between the two figures rather than a separately measured result, but the two figures are the ones above and they come from the same 11 trials.

There is also a dated-source problem. Claims that acupuncture on the day of embryo transfer works usually trace back to the 2008 version of the Cochrane review, which did report a benefit for live birth (odds ratio 1.89, 95% CI 1.29 to 2.77) — while saying in the same sentence that it "could be attributed to placebo effect and the small number of women included in the trials". That version was replaced in 2013 by the one quoted above, using more trials. PubMed flags the 2008 record as superseded. Pages still quoting it are quoting a withdrawn conclusion.

It is worth seeing the numbers rather than the headline, because the pooled estimates are not dramatic negatives — they are wide and they straddle no effect.

• Around egg collection: odds ratio 0.87 for live birth, 95% CI 0.59 to 1.29, from 2 studies and 464 women, no heterogeneity, rated low-quality evidence.
• Around embryo transfer: odds ratio 1.22 for live birth, 95% CI 0.87 to 1.70, from 8 studies and 2,505 women, with substantial inconsistency between them (I-squared 69%), also rated low-quality evidence.
• Pregnancy and miscarriage: no evidence of any effect.
• Side effects: no evidence of significant side effects across the 20 included trials.

Low-quality evidence means the true effect could differ from these estimates. It does not mean a benefit is hiding in there: the JAMA trial published five years later randomised 848 women across 16 centres, was properly sham-controlled, and landed on no difference.

Yes, and it is not the one clinics advertise.

Most people who book acupuncture during treatment are not doing a cost-benefit calculation on live birth rates. They are in the gap between appointments, where nothing is being done and nothing can be hurried, and booking something is a way of not just waiting. That is a real motive and it deserves to be said out loud rather than dressed up as pharmacology.

If that is the reason, the evidence does not stand in your way. The 2013 Cochrane review found no significant side effects across 20 trials. A licensed practitioner using single-use sterile needles is doing something low-risk. Paying for an hour a week in which someone is unhurried and attentive is a legitimate thing to want.

What the evidence will not let you say is that it is raising your chance of a baby, or that skipping it is costing you one. Hold those two things apart and the decision gets easier, not harder.

If the underlying thing you are trying to treat is the strain itself, that is worth naming as its own problem rather than a fertility intervention. Distress during fertility treatment is common and measurable, and what stress does and does not do to conception has been looked at directly.

Acupuncture is bought in courses, not single visits, and the money leaves slowly. Before the first appointment, get these answered.

1. How many sessions are you proposing, over how many weeks, and what does the whole course cost including any herbs or supplements? Total it up before you start, not after.
2. What outcome are you treating? If the answer is a higher chance of pregnancy or live birth, ask which sham-controlled trial that comes from. If the answer is that you will feel better able to cope, that is an answer you can assess yourself in three or four visits.
3. Are the needles single-use and sterile, and are they disposed of in front of you?
4. Will you tell my fertility team what you are doing, and will you avoid prescribing any herb or supplement without telling them? This matters because your IVF drugs are timed to the day and nobody should be adding unlisted compounds into that.
5. At what point would you tell me to stop?

And one question for your fertility team rather than the acupuncturist, phrased so it does not put anyone on the defensive: is there anything about my treatment plan that acupuncture or herbal medicine could interfere with, and would you rather I told you before I start? Most will say they have no objection to the needling and a firm objection to unlabelled herbal preparations. That is useful to know in advance.

This is the part of the page worth reading twice. Each of the following is commonly claimed, including in the previous version of this page, and none of it is supported by the sources cited here.

• That acupuncture improves live birth rates. The opposite is what was found. Live birth 18.3% versus 17.8% against sham in 848 women (Smith 2018); pooled risk ratio 1.01, 95% CI 0.88 to 1.15 across 11 trials (Wang 2025); no evidence of benefit across 20 trials (Cheong 2013). Three independent analyses, same answer.
• That it reduces miscarriage. The 2013 Cochrane review looked at miscarriage as a stated outcome and found no evidence of an effect. No source here supports the claim.
• That it improves sperm count, motility or erectile function. No source retrieved for this page measures acupuncture against any male fertility parameter. The claim is removed rather than softened, and it is not replaced with a hedge.
• That it thickens the endometrium or improves egg quality. These are mechanism stories. Even if a mechanism were real, the outcome trials above would still show it, and they do not. A mechanism with no outcome behind it is not evidence of benefit.
• That treating stress improves fertility outcomes. This one is genuinely open, and it is worth being precise. A Cochrane review of psychological and educational interventions included 39 studies and 4,925 people undergoing ART and declined to pool the results at all, judging that pooling would not give a clinically meaningful estimate. Only 2 of the 39 studies reported live birth and 1 reported ongoing pregnancy, all with substantial drop-out, and every study was at high risk of bias in at least one domain. Its conclusion is that the effect on live birth or ongoing pregnancy is uncertain on very low quality evidence (Verkuijlen 2016, Cochrane CD011034). That is too poor to say, not shown not to work — and it is about counselling and education, not about acupuncture.
• That the 2025 meta-analysis endorses acupuncture. It is sometimes cited that way because of its 1.25 figure against untreated controls. Its own live birth result is 1.01, and its own sham comparison is 1.01. Read the results, not the framing.
• How much a course costs in India. No pricing data was retrieved for this page, so no figure is given. Ask for the full course price in writing.

On the male-factor point, the measured ground is elsewhere: what lifestyle changes have actually been measured in men and what a semen analysis tells you.

Two neighbouring pages take the same approach to other popular add-ons: Ayurveda and fertility, what has actually been tested, and fertility yoga and meditation. A longer list of claims checked against evidence is in fertility myths versus facts, and the same sham-versus-no-treatment problem runs through most of the paid IVF add-ons offered alongside a cycle.

If you are deciding this week, three things are worth more than the acupuncture question.

• Ask your clinic what is actually limiting your chance in this cycle, and whether anything on that list is modifiable. A specific answer beats a general intervention.
• If either of you has a semen analysis in the file, ask when it was done and whether it should be repeated. Sperm in any given sample was made over the preceding weeks, which is why a repeat is usually left about three months rather than done straight away.
• Decide the acupuncture question on its own terms — cost, time, how it makes you feel — and not as a bet on the outcome. The trials have taken that part of the decision off the table.

Sometimes the answer is testing, timing or a simpler treatment rather than another thing added on top. That is not a lesser answer.