Can Dogs Eat Honey? What Vets and Trials Say

Honey is not poisonous to a dog. It appears nowhere on the standard veterinary list of food hazards — a list that chocolate, grapes and xylitol all earn their places on. Past that, the two things most worth knowing are simple: honey is very nearly pure sugar, which is where the veterinary sources actually put their attention; and the sharpest safety fact in the whole subject concerns a sweetener that is not honey at all. The research record on feeding a dog honey, meanwhile, is almost empty — which is itself a finding, and this page reports it as one.

None of this is veterinary advice, and no amounts or frequencies appear here. Anything concerning an actual dog — emphatically one that is overweight, diabetic, very young, pregnant or unwell — is a question for your own vet or veterinary nurse, who can see the animal. Everything below is the published record, cited as it comes.

Plate I · PaintingRecord

What are the real risks of giving a dog honey?

The sugar, before anything else. Honey runs to 304 kcal and about 82 g of sugars per 100 g on the standard reference analysis, and a tablespoon weighs 21 g — roughly 64 kcal and 17 g of sugars. The composition figures and their source are on the honey evidence page.

Set that against what a dog needs, because the arithmetic is the lesson. The MSD Veterinary Manual gives resting energy requirement as 30 times bodyweight in kilograms plus 70 for dogs of 2 to 45 kg, and maintenance for a moderately active neutered adult as 1.6 times resting: a 10 kg neutered adult sits near 590 kcal a day, and a tablespoon of honey is roughly a tenth of it. The manual cautions that animals of the same weight vary in energy requirement by as much as 30% either way — so that is an order of magnitude, not a ration.

The arithmetic lands in a population already carrying weight. The PDSA’s obesity page states that “Veterinary professionals estimate that nearly half of dogs in the UK are overweight”, and lists what excess weight is associated with: arthritis, diabetes, Cushing’s disease, back and cruciate problems, breathing difficulty, certain cancers and reduced lifespan. It gives no number for extras, directing owners to the practice: “Your vet or vet nurse can help you work out how much you should be giving to your dog.”

Fig. I · The food-hazard shelfDiagram

MSD Veterinary Manual · food-hazards section — seven entries

• Garlic and onion
• Avocado
• Bread dough
• ChocolateMild signs at 20 mg/kg of methylxanthines · seizures at 60 mg/kg or more
• Macadamia nut
• Grape and raisin
• XylitolHypoglycemia above approximately 100 mg/kg · possible severe hepatic insufficiency or failure above 500 mg/kg

Two of the seven carry a dose figure on this page; the other five are named without one.

Standing outside the shelf

Honey — no entry, and no label to draw

Honey is not among the seven. It has no entry because no such toxicity is described.

That is an absence, not a clearance.

Absence from a hazard list is not the same as having been studied and cleared.

What about a diabetic dog?

This is where the sugar question stops being about weight. The PDSA describes canine diabetes as a condition in which a dog “can’t produce enough insulin to control their blood sugar level (causing it to become dangerously high)”, and sets out a feeding regime built on sameness: the same amount and type of food, twice daily, twelve hours apart, and dogs on it “should never have treats in between meals (because the extra food will cause a rise in blood sugar)”. It names the arrangement plainly — “Managing diabetes is a partnership between you and your vet.”

No trial of honey in diabetic dogs was located for this page, and none is needed to settle the point — a diabetic dog’s diet is a clinical prescription, and the vet who wrote it is the only person who can say what belongs in it.

Does sugar rot a dog’s teeth the way it does ours?

The veterinary sources do not say so — and the disease they describe is a different one: plaque-driven, not sugar-driven.

Periodontal disease in dogs is common — the MSD Veterinary Manual reports that up to 80% of dogs have some level of it by the age of two. But the mechanism described is plaque, not sugar: the disease is “caused by plaque bacteria accumulating on the tooth surface and eliciting a host response to the bacterial insult”, with risk raised by toy breeds, immunosuppression, poor nutrition and tooth misalignment. On diet, the manual says texture is what matters. The PDSA’s dental pages likewise name poor brushing, wear, breed and overcrowding, and do not name sugar.

So this is a gap, not a finding. The sticky-sugar-causes-decay story is human dentistry, and the veterinary literature consulted here neither extends it to dogs nor rules it out.

Can puppies, or an unwell dog, have honey?

Here the record thins to almost nothing, and the honest answer is to say so.

The human rule is firm: NHS guidance is that honey should not be given to a child under 12 months, because honey occasionally contains bacteria that can produce toxins in a baby’s intestines, causing infant botulism. That rule was written for human infants and rests on a human clinical literature.

What it does not do is transfer. No veterinary source consulted for this page applies the infant honey rule to puppies or to immunocompromised dogs, and none rejects it either — the question is simply not addressed. That gap is the finding, and the vet who knows the animal is the person to ask.

Is honey toxic to dogs?

Not according to the reference vets use. The MSD Veterinary Manual’s food-hazards section covers seven entries — garlic and onion, avocado, bread dough, chocolate, macadamia nut, grape and raisin, and xylitol — and honey is not among them. The entries earn their places with data: chocolate has a dose-response curve, mild signs at 20 mg/kg of methylxanthines, seizures at 60 mg/kg or more. Honey has no entry because no such toxicity is described.

Worth learning from a hazard list, though, is what it can and cannot say: absence is not a clearance. The standard veterinary reference does not classify honey as a poison; nothing further follows.

Why is xylitol the warning that actually matters?

Because it is the fact here most likely to prevent harm, and it is about labels rather than honey.

Xylitol is severely toxic to dogs. The MSD Veterinary Manual reports that doses “greater than approximately 100 mg/kg (45 mg/lb) have been associated with hypoglycemia in dogs”, and that some dogs ingesting it at doses above 500 mg/kg “may develop severe hepatic insufficiency or failure”. Signs of low blood sugar “can develop within 30 minutes after ingesting xylitol or may be delayed up to 12–18 hours if the xylitol is in a substrate that slows absorption”; signs of liver injury may not appear until 24 to 48 hours or more after ingestion. Prognosis splits sharply — good for uncomplicated hypoglycaemia treated promptly, but “in one study, 62.5% of dogs with clinical signs of liver injury died or were euthanized despite aggressive veterinary intervention”.

The relevance to honey is the packet. The manual lists xylitol as a sweetener in “sugar-free products, such as gums, candies, mints, peanut butter, and baked goods” and in “Sugar-free syrups and elixirs used as medicinal bases”. A honey-style spread or a reduced-sugar syrup is not pure honey, and the ingredients list is the only way to tell. If a dog has swallowed anything containing xylitol, that is a same-hour call to a vet or an out-of-hours service.

Is there any evidence honey does a dog any good?

Very little, and every canine use the veterinary scoping review describes is clinician-applied — honey put on a wound, in an ear, or instilled at surgery — rather than a dog being fed honey.

The most complete survey is a scoping review by Vogt and colleagues (2021), which mapped 436 primary research studies of medicinal honey use in animals — and the imbalance it found is the story. Of those 436, 350 publications were biomedical research, overwhelmingly laboratory rodents, against 47 veterinary publications reporting 49 studies. Of those 49, 23 were case reports or case series, 18 were challenge trials and only 8 were controlled trials — and the hierarchy matters, because a case series has no comparison group and cannot separate the honey from everything else that happened. Dogs featured in 12; the canine indications recorded include wound healing, burns, necrotising fasciitis, otitis externa, bacterial colonisation at an intravenous catheter site, and “the prevention of post-operative peritoneal adhesions in dogs” — that last one instilled at surgery rather than applied to the skin. Across all 49 veterinary studies the review counts 38 that applied honey topically and 4 that gave it orally, and it does not break the oral figure down by species, so whether any of those four involved a dog cannot be recovered from the review. The authors’ conclusion is the honest headline: “there is insufficient literature for synthesis research in companion animals (dogs and cats) and livestock species”. They did not assess the quality of what they found.

Fig. II · The honey-in-animals literatureDiagram

Vogt and colleagues (2021) · a scoping review of 436 primary research studies of honey used in animals

Scale one · 350 runs the full width

Biomedical research350

Publications — overwhelmingly laboratory rodents.

Veterinary49

Studies, reported in 47 veterinary publications.

Scale two · the veterinary 49, redrawn at the full width

23188

Case reports or case series — 23Challenge trials — 18Controlled trials — 8

Studies featuring dogs12 of 49

12

A count on the same scale, not a position: the page does not say how these 12 fall across the three bands above.

Across all 49 · how the honey was given

• 38applied honey topically
• 4gave it orally
• These two counts are all the page gives. It does not say they are exclusive, nor state a route for every one of the 49.

The review does not break the oral figure down by species, so whether any of those four involved a dog cannot be recovered from it. Every canine use the review describes is clinician-applied — honey put on a wound, in an ear, or instilled at surgery — rather than a dog being fed honey.

The review’s own conclusion

“there is insufficient literature for synthesis research in companion animals (dogs and cats) and livestock species”

Where a controlled trial in dogs does exist, it repays close reading. Gouletsou and colleagues (2024) made three four-centimetre sutured incisions per side in 12 purpose-bred laboratory Beagles and 12 cats, randomly assigning one wound per set to manuka honey applied daily to day 15, one to a comparator product and one to no treatment; two surgeons scored cosmetic outcome blind from photographs — blinding being what stops a scorer seeing what they expect to see. Against control, manuka honey produced no statistically significant cosmetic difference, more skin thickening, and no significant difference in the total histological score. Its clearest advantage was a significantly smaller ultrasonographic wound area (p < 0.001), “the difference being more apparent in dogs”. The authors concluded the comparator “was proven to be a better choice in achieving favorable wound healing than Manuka honey in dogs and cats in first-intention healing”, and noted these were laboratory animals under controlled conditions.

The claims people search for fare worse. For a dog’s allergies, no controlled veterinary study was located. For kennel cough, a PubMed search pairing honey with kennel cough, canine infectious respiratory disease or Bordetella bronchiseptica returns no records, and the PDSA’s kennel cough page describes rest, comfort and a steamy room without mentioning honey. The human cough trials sometimes cited here were run in children over a few days and cannot be read across to another species; what they do and do not show is set out on the honey evidence page, and the laboratory work behind graded honeys is read on its own terms on the manuka page.

The grade of honey does not change any of this either. No veterinary source consulted for this page distinguishes raw honey from processed honey for a dog — not the MSD manual, not the PDSA, not the scoping review, not the one controlled canine trial. Raw honey is not recorded as better for a dog, or worse; it is simply not a distinction the veterinary record makes.

How have regulators read it?

For food sold to people in England, Scotland and Wales, the Great Britain nutrition and health claims register is the authoritative record, and its rule is flat: “Only authorised claims in the Great Britain NHC register may be used in Great Britain.” No health claim for honey, or any other bee product, is authorised on it — a useful calibration even when the eater is a dog, because it is how the assessors scored the human evidence.

The record, in one paragraph: honey is not a listed poison; it is nearly pure sugar, arriving in a population of dogs nearly half of which already carry too much weight; almost none of the folklore around it has been tested in a dog, and the one controlled canine trial found the comparator the better choice; the fact most likely to prevent real harm concerns xylitol, not honey; and the dog in front of you is a matter for your vet.