English (UK) 
Italiano 
Français 
简体中文 
Honey and Cough: What the Evidence Shows, and What We Are Not Allowed to Say
NICE lists honey as a self-care option for acute cough. The claims regulator has refused every health claim ever submitted for honey. Both of those things are true, and the gap between them is the whole story.
Never give honey to a baby under 12 months
This is not a precaution. It is an absolute rule. Honey can contain spores of Clostridium botulinum. An adult gut handles them without difficulty. An infant gut cannot, and the spores can germinate and produce toxin, causing infant botulism, a rare but serious illness.
The age threshold is twelve months, not "when they start solids" and not "a little on a dummy". It applies to raw honey, to pasteurised honey, to honey baked into food, and to honey in any product. If your child is under one, none of the rest of this page applies to them.
There is a genuine paradox at the centre of this subject, and almost nobody writes about it, because doing so requires a honey company to argue against its own interests.
On one side, the National Institute for Health and Care Excellence lists honey among the self-care options a person might try for an acute cough. On the other, the Great Britain nutrition and health claims register contains eight submitted claims for honey and has authorised none of them. One of the refused claims is, word for word, "soothing for the throat".
So a doctor may suggest it and a honey seller may not claim it. That sounds absurd until you look at what the evidence actually says, at which point it starts to look quite reasonable. This page is an attempt to set out both sides properly, which is a change from what used to be here.
What NICE actually says
The relevant document is NICE guideline NG120, on acute cough and antimicrobial prescribing. Its self-care section is short, and the exact wording matters a great deal more than the summaries of it that circulate online.
Be aware that some people may wish to try the following self-care treatments, which have limited evidence of some benefit for the relief of cough symptoms: honey (in people aged over 1 year).
Cough (acute): antimicrobial prescribing, NICE guideline NG120Read the construction of that sentence. It is not a recommendation to use honey. It is an instruction to clinicians to be aware that patients may wish to try it, qualified by "limited evidence of some benefit". Honey appears alongside pelargonium and over-the-counter guaifenesin in the same list, on the same footing.
NICE also frames the condition itself in a way that gets lost. Acute cough is usually viral and self-limiting, with a typical course of up to three or four weeks. Anything taken during that period is competing against a symptom that was going to resolve anyway.
What the trials actually measured
NICE's position rests on a Cochrane systematic review, Oduwole and colleagues, which pooled three randomised controlled trials involving 568 children and young people aged one to seventeen with acute cough caused by an upper respiratory tract infection.
Three trials. Of those, the review recorded two at high risk of bias and one at low risk of bias. That is a thin and uneven evidence base by the standards of clinical medicine, and it is the reason NICE graded the finding low to moderate quality rather than strong.
The outcome measure was a seven-point Likert scale completed by carers, running from "extremely" at six points down to "not at all" at zero. NICE's own evidence summary notes that it was not clear how those responses were defined. A parent's rating of their child's cough is a legitimate outcome, since it is what the family experiences, but it is not an objective count of coughs.
Against that scale, honey reduced cough frequency and severity at one day follow-up by roughly 0.5 to 2 points compared with placebo, no treatment, or the antihistamine diphenhydramine. That is the finding. It is real, and it is modest.
Outcomes as summarised in the NICE NG120 evidence review, drawn from the Cochrane systematic review by Oduwole and colleagues.
The bottom row is the one that never appears in marketing copy. Compared with dextromethorphan, an ordinary over-the-counter cough suppressant, honey did not reduce cough frequency or severity, on very low quality evidence. And there was no significant difference in sleepiness when honey was compared with diphenhydramine, which is worth knowing given how often honey is promoted for night-time coughing.
Chronic cough: there is no evidence at all
A separate Cochrane review looked specifically at honey and lozenges for children with chronic non-specific cough, meaning a dry cough with no known cause lasting more than four consecutive weeks. It found no randomised controlled trials that met its inclusion criteria.
Not weak evidence. No applicable trials. If your cough has lasted more than three or four weeks, you are outside the situation any of this research examined, and that is a reason to see a doctor rather than to reach for a jar.
Why we cannot say any of this as a claim
Here is where a honey business has to be careful, and where this page previously was not.
In Great Britain, only health claims listed as authorised on the GB nutrition and health claims register may be used in marketing, a requirement reflected in Section 15 of the CAP Code. The register contains eight submitted entries for honey. Every one carries the status "non-authorised".
Every honey claim ever submitted, and its status
Source: Great Britain nutrition and health claims register, last updated 19 May 2026. The list above is abbreviated; there are eight honey entries in total and none is authorised.
The consequences are strict. Advertising guidance is clear that where a product contains no nutrient or ingredient about which an authorised claim can be made, advertising for that product should not contain health claims at all. The rules also cover implied claims, not only explicit ones, and advertising is judged on how a consumer is likely to read it rather than on what the advertiser intended.
Quoting NICE does not solve the problem either. A claim does not become permissible because a third party said it first. The Advertising Standards Authority has previously upheld a complaint against a honey product making cough claims that echoed public health advice.
So: we can tell you what NICE published, in an article, with the qualifications attached. We cannot put "soothes coughs" on a jar, in an advert, or in a product description. That distinction is not a loophole we are exploiting. It is the reason this page reports rather than promises.
The safety boundaries that actually matter
Setting claims aside, there are practical limits worth knowing, and the previous version of this page got some of them wrong.
| Situation | Position |
|---|---|
| Babies under 12 months | Never, in any form. Risk of infant botulism. This is absolute. |
| Children over 1 year | NICE applies the same evidence base as for adults, with the same caveats about quality. |
| Teeth | Honey is a sugar. The NICE committee specifically discussed concerns about tooth decay, which matters most with repeated night-time use in children. |
| Diabetes | Honey raises blood glucose much like other sugars. Anyone managing diabetes should treat it as sugar and take clinical advice. |
| Bee product or pollen allergy | Avoid, and take medical advice. |
| Heating | Never above 40C, and never in a microwave. Let a hot drink cool for a minute or two first. This is about protecting flavour and enzymes, not about efficacy. |
One correction we owe you. An earlier version of this article stated that honey may increase the sedative effect of some medicines or interfere with antibiotic absorption. We could not substantiate that from any reliable source and it has been removed. If you take regular medication and are unsure about anything you eat alongside it, ask a pharmacist, who will give you a better answer than a honey website.
When to stop reading this and see a doctor
NICE sets out clear safety-netting advice for acute cough. Seek medical help if there is:
- Worsening breathlessness or difficulty breathing
- Chest pain
- Coughing up blood
- New confusion
- A persistent fever, or symptoms that get rapidly or significantly worse
- A cough lasting beyond three to four weeks
HoneyBee & Co. sells food, not medicine, and makes no medicinal claims. Nothing here is medical advice or a substitute for it.
So what is honey actually good for here?
An honest answer: it is a pleasant thing to put in a hot drink when you feel rotten, it is cheap, it is well tolerated, and a large number of people find it comforting. NICE thought that combination was worth mentioning to clinicians. We think that is a perfectly good reason to keep a jar in the cupboard, and a bad reason to make promises about it.
If you want the buyer's side of this, which honeys hold up in hot water and how the varieties differ in a mug, we have written that separately in our guide to honey for a sore throat. This page is about the evidence. That one is about the choosing.
Honeys that hold their character in a hot drink
Chosen for flavour and how they behave in hot water. Nothing on this page is a claim about what they will do for a cough.
Linden Honey
Delicate and faintly herbal, from lime tree blossom. The one most people reach for in a bedtime mug.
British Heather Honey
Robust and dark, with a character strong enough to come through hot water and lemon. A single annual harvest.
Acacia Honey
The mildest of the six, pale and clean, for anyone who wants sweetness without a strong floral note.
Also available: British Wildflower and British Soft Set. Browse the full shop or subscribe and save 20%. Free UK delivery on orders of three jars or more.
The hives and the flowers
Where the honey on this page comes from, before it reaches a jar.
Our honeys
The six varieties, photographed across several harvests.
On the table
How people actually use it, which is the part the evidence has least to say about.
Questions people actually ask
Does honey help a cough?
NICE lists honey among self-care options that have limited evidence of some benefit for relief of cough symptoms in people aged over one year. The underlying evidence is three randomised trials in 568 children, two of which were at high risk of bias, graded low to moderate quality. Honey performed better than placebo, no treatment and an antihistamine, but showed no reduction compared with the cough suppressant dextromethorphan.
We are not permitted to turn that into a claim about our honey, and we are not going to. It is reported here so you can weigh it yourself.
At what age can a child have honey?
Twelve months. Not before, in any form, including cooked into food or on a dummy, because of the risk of infant botulism. From one year onwards NICE applies the same evidence base as for adults.
Which honey is best for a cough?
No UK guidance recommends one variety over another. NICE refers to honey generally. Manuka is often marketed for this and there is no head-to-head clinical evidence in UK guidance placing it above other honeys for cough.
Choose on flavour and how it behaves in a hot drink. Heather stands up to lemon, Linden is gentler, Acacia is the mildest.
Should I put honey in hot tea?
Let the drink cool for a minute or two first. Raw honey should not be heated above 40C, and it should never be microwaved, because heat drives off the volatile aromatics and degrades the enzymes that distinguish raw honey from processed. This is a point about quality, not about effectiveness.
How much honey should I take for a cough?
We are not going to give you a dose. Recommending a quantity of a food for the relief of a symptom moves it towards being presented as a medicine, which honey is not and which we are not licensed to sell. A previous version of this page did exactly that, and it was wrong to.
Honey is food. Use it as you would use any sweetener, and talk to a pharmacist or GP if you want guidance on treating a cough.
Why does NICE mention honey if the claim is not authorised?
They are different systems doing different jobs. NICE writes clinical guidance for healthcare professionals, weighing evidence of any quality and saying so plainly when it is limited. The GB nutrition and health claims register governs what a business may say to sell a product, and sets a much higher evidential bar before a claim can be attached to a food.
A clinician mentioning something with limited evidence is giving balanced advice. A company printing it on a label is making a commercial promise. The second requires authorisation, and for honey there is none.
Sources
- NICE guideline NG120, Cough (acute): antimicrobial prescribing, including section 1.2.1 and the summary of the evidence. nice.org.uk
- Oduwole O, Meremikwu MM, Oyo-Ita A, Udoh EE. Honey for acute cough in children. Cochrane Database of Systematic Reviews, CD007094.
- Mulholland S, Chang AB. Honey and lozenges for children with non-specific cough. Cochrane Database of Systematic Reviews, CD007523.
- Department of Health and Social Care, Great Britain nutrition and health claims register, last updated 19 May 2026.
- Committee of Advertising Practice, AdviceOnline guidance on food health claims, and Section 15 of the CAP Code.
- NHS guidance on foods to avoid giving babies and young children.