Honey and coughs: the short answer
In children over one year old, a spoonful of honey before bed slightly reduces nighttime coughing from a cold, about as much as a cough suppressant syrup. In adults, the effect has not been demonstrated. And before age 1, it’s a no, without exception.
It’s a modest result, but in a field where almost nothing works. The evidence for cough syrups is so thin that the British NICE guideline lists honey among the few self-care options with “some limited evidence” of benefit, while finding that antihistamines, decongestants and codeine do not help.
Honey has not been shown to shorten the illness. It may make one night a little less miserable, for children and their parents. For something that costs a few cents a spoonful, that’s a fair deal.
Why a sweet spoonful might soothe a cough
The mechanism is not established, and that needs saying before we look at the trials. Three hypotheses are circulating:
- the soothing effect : a thick, sweet liquid coats the throat and soothes the irritation that triggers coughing, like any syrup
- the sweet taste : British researcher Ronald Eccles proposed in 2006 that sweetness modulates the cough reflex in the brainstem, perhaps through opioids produced by the body. He himself describes this as speculation
- antimicrobial activity : real in the laboratory for some honeys, but with no demonstrated link to coughing, which is almost always caused by a virus
There is a fourth player to add, more powerful than the other three: the placebo. In a 2002 review of 8 trials of syrups in adults with colds, Eccles estimated that 85% of the reduction in coughing came from the placebo effect, and only 15% from the active ingredient. A 2005 trial measured this directly in 54 adults: cough frequency fell six times more with a placebo than with nothing, with a median reduction of 18 versus 3.
Keep that idea in mind. A trial comparing honey with “nothing” measures the honey, the sweetness and the spoonful ritual all at once. That’s not a flaw if your aim is a calmer night. It is if someone is selling you a benefit supposedly unique to a particular honey.
What the trials in children show
The two trials that started the idea
The first, published in 2007 by Paul and colleagues, included 130 children aged 2 to 18 who had had a cold for no more than 7 days, of whom 105 were analyzed. A single dose was given 30 minutes before bedtime: buckwheat honey, honey-flavored dextromethorphan, or nothing. Parents rated coughing from 0 to 6.
Cough frequency improved by 1.89 points with honey, 1.39 with dextromethorphan and 0.92 with nothing. Only the difference between honey and no treatment was significant. For cough severity and sleep, honey did not perform significantly better than nothing, and dextromethorphan did not outperform nothing on any outcome.
Two limitations weigh on this trial. The “nothing” group knew they were getting nothing, which automatically inflates the difference. And the study was funded by an unrestricted grant from the National Honey Board, the US honey promotion organization.
The second, published in 2012 by Cohen and colleagues, addressed the first flaw. In six Israeli clinics, 300 children aged 1 to 5 received, under double-blind conditions, 10 g of eucalyptus, citrus or Lamiaceae honey, or 10 g of date extract as a placebo, once before bedtime. The three honeys improved cough frequency by 1.8–2 points, compared with 1 point for the placebo.
Here is the most revealing detail: the placebo, which was also sweet, already substantially improved coughing. Honey did better, but against an already strong response. This trial was co-funded by the Honey Board of Israel, among others.
What the Cochrane review concludes
The 2018 Cochrane review brought together 6 trials and 899 children aged 1 to 18, using 7-point cough scales where a lower score means less coughing:
| Comparison | Difference favoring honey | Certainty |
|---|---|---|
| Versus nothing | 1.05 points, 154 children | Moderate |
| Versus placebo | 1.62 points, 402 children | Moderate |
| Versus diphenhydramine | 0.57 points, 80 children | Low |
| Versus dextromethorphan | No difference, 149 children | Low |
In parent language: about one point out of seven, meaning a shift from “frequent” coughing to “slightly less frequent” coughing. Most children were treated for just one night. Over longer periods, honey’s advantage fades: beyond 3 days, it probably performs no better than a placebo.
The authors conclude that there is no strong evidence for or against honey. Two trials had a high risk of bias because they lacked blinding.
Two more recent reviews point in the same direction. In 2023, Kuitunen and Renko identified 10 trials and an effect of 0–1.1 points versus placebo or no treatment, with low- to very-low-quality evidence. In 2026, a meta-analysis of honey versus medicines, covering 576 children followed for no more than 3 days, found no difference, apart from a small advantage for children’s sleep.
What about adults?
The evidence in adults is much weaker.
The meta-analysis by Abuelgasim, published in 2021, brought together 14 trials on upper respiratory tract infections. Compared with usual care, honey reduced cough frequency and severity. Compared with a placebo, the difference was no longer significant. And in the subgroup of 3 adult trials, there was no significant effect on cough frequency.
This meta-analysis also counted combination syrups containing other plants, honey mixed with coffee or milk, as “honey.” That is one reason its overall result cannot be applied to an adult’s spoonful of honey.
The only recent adult trial was negative. In Catalonia, Llor and colleagues followed 194 adults with acute bronchitis. Honey, at one tablespoon 3 times a day for up to 14 days, did not shorten coughing: a median of 6 days of moderate to severe coughing, compared with 5 with usual care. The trial was open-label and did not reach its planned sample size because of Covid-19, which limits its scope.
You may come across a 2013 Iranian trial reporting that coughing almost disappeared within a week with a honey-and-coffee paste. It has internal inconsistencies, with 85 participants detailed out of 97 announced and an implausible drug dose in the text, and it cannot separate the effect of honey from that of coffee. I wouldn’t rely on it for anything.
For adults, honey in a warm drink therefore remains a comfort measure. It is not a proven treatment.
Honey versus syrup: a clash of lightweights
Honey beats or matches cough syrups mainly because cough syrups don’t do much. In Paul’s trial, dextromethorphan did not outperform nothing on any outcome. The NICE guideline finds that antihistamines, decongestants and codeine do not help with acute cough.
In France, the list of cough syrups advised against has grown:
- mucolytics and mucus-thinning medicines have been contraindicated in infants since April 29, 2010, and that same year Afssaps considered that cough suppressants should not be prescribed before age 2
- codeine has been contraindicated for coughs before age 12 since a 2015 European decision
- pholcodine was suspended and all batches recalled on September 8, 2022, then its authorizations were withdrawn on April 5, 2023, because of a risk of severe allergy to neuromuscular blocking agents during anesthesia
Diphenhydramine, the antihistamine compared with honey in the Cochrane review, improves sleep because it is sedating. The article on antihistamines used for sleep · FR explains why this drowsiness is not harmless.
Honey is not free of side effects either. In the Cochrane review, 9.3% of children given honey became agitated, nervous or slept poorly, compared with 2.7% given dextromethorphan. The difference is not significant, but it is a reminder that a spoonful of sugar at bedtime is not a sleeping pill.
The studied dose, and how to use it
The most widely used protocol comes from Paul’s trial: a single dose, 30 minutes before bedtime. The volumes matched those of the comparison syrup:
- ages 2 to 5 : half a teaspoon
- ages 6 to 11 : one teaspoon
- ages 12 to 18 : two teaspoons
Between ages 1 and 2, the only reference is Cohen’s trial, which gave 10 g to children aged 1 to 5, or about one and a half teaspoons.
For adults, no dose has proven effective. Llor’s trial used 30 g three times a day: with honey containing 82 g of sugars per 100 g according to the USDA database, that adds up to nearly 75 g of sugars a day. The World Health Organization counts the sugar in honey as free sugars, to be kept below 10% of energy intake, or 50 g for 2,000 kcal. One or two teaspoons in the evening remain well below that threshold.
Three practical rules follow from the trials:
- Try it for one to three nights. That is the duration studied. If nighttime coughing doesn’t change, there’s no point continuing.
- Brush teeth afterward. The authors of Cohen’s trial note that frequent honey consumption promotes tooth decay, and the dose comes just before bedtime.
- Choose ordinary honey. No trial shows that one variety works better than another for coughs, and Cohen’s three honeys produced almost identical results.
If you have diabetes, count the spoonful as sugar: about 5–6 g of sugars in a 7 g teaspoon.
Manuka: pay more, cough just as much
Manuka honey, produced mainly in New Zealand, is sold at a much higher price than ordinary honey, with UMF or MGO ratings on the label. These ratings mainly relate to its methylglyoxal, the molecule responsible for most of its antibacterial activity in the laboratory : most honeys no longer inhibit certain bacteria when diluted to 80%, whereas manuka remains active when diluted to 15–30%.
That is a real asset for honey wound dressings, not for coughs. No clinical trial has tested manuka for coughs or compared it with ordinary honey. The only study on respiratory pathogens I found, published in 2026, was conducted in the laboratory and co-funded by a manuka seller.
It’s a textbook case of Mitochondria Marketing : using a real biological mechanism to sell a product whose benefit has not been demonstrated. The antibacterial activity exists, but a cold-related cough is caused by a virus, and the rating on the jar has never been linked to a calmer night.
For other winter products sold for colds, the same questions produce mixed answers: zinc lozenges, echinacea and elderberry syrup each have their own evidence base.
Does your jar really contain honey?
The question isn’t just provocative. In 2021 and 2022, the European Commission had 320 batches of imported honey tested as they entered the Union. Of these, 147, or 46%, were suspected of being adulterated with sugar syrups, compared with 14% in the previous campaign. The rates were 74% for China, 93% for Turkey and 100% for batches from the United Kingdom, probably honey produced elsewhere and then processed there.
Suspected does not mean proven, and adulterated honey does not make people ill. But the irony is complete: if honey’s effect on coughing comes mainly from sweetness, a jar diluted with syrup probably wouldn’t cost you much, except your money. It’s only a hypothesis, but it fits the trials where the sweet placebo already improved coughing. With “functional” honeys, fraud sometimes takes a much more serious form, as shown by aphrodisiac honeys containing a hidden drug.
Before age 1: never, and here’s why
This is the only absolute rule in this article. Anses “strongly advises against” giving honey to a child under one year old, and Santé publique France makes the same recommendation.
Honey can contain spores of Clostridium botulinum, the bacterium that causes botulism. In adults, gut flora generally prevents them from germinating. In infants, whose gut flora is still immature, they can germinate and produce a toxin in the intestine that paralyzes muscles. According to Anses, honey is the only identified dietary source of exposure.
Infant botulism remains rare in France. Santé publique France recorded 7 cases between 1991 and 2009, including 3 in infants who had consumed honey, then 11 cases between 2018 and 2024, all under one year old, with no identified honey consumption. Spores also come from dust and soil. Rarity does not change the rule: the expected benefit of honey for a cough is tiny; the risk is paralysis.
For a coughing infant, the 2010 Afssaps guidance does not recommend any cough suppressant and lists the signs that require medical attention: unusual irritability or drowsiness, fever that the child is coping with poorly, feeling unwell, blue lips, a floppy child, coughing that causes vomiting, bottle intake reduced by more than half, or difficulty breathing with chest retractions, pauses or grunting.
When a cough needs a doctor
An acute cough from a cold gets better on its own within 3 to 4 weeks, according to NICE. Honey does not change that timeline. The following situations are beyond self-care:
- call 15 in France or 112 in the EU, 911 in the United States, 999 in the UK, or your local emergency number in other countries if coughing comes with difficulty breathing or swallowing, according to the Santé.fr
- seek urgent medical attention if the cough rapidly worsens, or if you have chest pain, feel very unwell or cough up blood, according to the NHS
- see your doctor if the cough lasts more than 3 weeks, if you lose weight for no reason, or if your immune system is weakened
- for an infant, at any of the warning signs listed above





