Every breed page on the internet gives you a tidy range. Twelve to fourteen years. Ten to twelve. Seven to ten. Almost none of them tells you where the number came from, how many dogs it was measured on, or whether it was measured at all.
That changed recently. Two of the largest studies ever conducted on canine longevity have now been published — one covering 584,734 dogs, the other building formal life tables from 30,563 deaths — and they let you replace the tidy range with something that has a sample size attached. They also disagree with each other in one spectacular case, and working out why is genuinely more useful than the headline number.
The short answer
- Median survival across 584,734 UK dogs was 12.5 years.
- Life expectancy at birth, measured separately on veterinary records, was 11.23 years.
- Three things move that number more than breed folklore does: body size, skull shape, and body condition. You choose two of them at the point of purchase. The third is yours for the dog’s whole life.
Two studies, two numbers, and why both are correct
The two datasets are worth understanding separately, because you will see both quoted and they are not interchangeable.
The big one. A 2024 paper in Scientific Reports pooled records from 18 sources — rehoming and welfare charities, breed registries, pet insurers, veterinary corporates and university archives — for 584,734 dogs, of which 284,734 had died. It reported median survival for 155 purebreds plus a crossbred group. That headline median: 12.5 years (95% CI 12.5–12.6).
The precise one. A 2022 paper, also in Scientific Reports, used the Royal Veterinary College’s VetCompass database to build proper life tables — the same demographic tool used for human populations — from 30,563 dogs that died between January 2016 and July 2020. Life expectancy at age 0 came out at 11.23 years (11.19–11.27).
The gap is a methods gap, not a contradiction. The first study estimates median survival using the standard survival-analysis approach, which keeps dogs that are still alive in the calculation as incomplete observations rather than discarding them. The second builds a hypothetical cohort from the death rates observed at each age — a more formal instrument, and, as we are about to see, a more fragile one. The two also draw on different populations: one pooled 18 sources including breed registries and insurers, the other used first-opinion veterinary practices only.
The breed table
Here are the breeds the two studies name, side by side. The middle column is life expectancy at birth from the life-table study; the right column is median survival from the large pooled study. Blanks are breeds one study covered and the other did not.
| Breed | Life expectancy at birth (VetCompass life tables) | Median survival (584,734-dog study) |
|---|---|---|
| Lancashire Heeler | — | 15.4 |
| Tibetan Spaniel | — | 15.2 |
| Shiba Inu | — | 14.6 |
| Papillon | — | 14.5 |
| Border Terrier | — | 14.2 |
| Miniature Dachshund | — | 14.0 |
| Border Collie | 12.10 | 13.1 |
| Labrador Retriever | 11.77 | 13.1 |
| Jack Russell Terrier | 12.72 | — |
| Yorkshire Terrier | 12.54 | — |
| Springer Spaniel | 11.92 | — |
| Crossbred | 11.82 | 12.0 |
| Staffordshire Bull Terrier | 11.33 | — |
| Cocker Spaniel | 11.31 | — |
| Shih-tzu | 11.05 | — |
| Pug | 7.65 | 11.6 |
| Cavalier King Charles Spaniel | 10.45 | — |
| German Shepherd Dog | 10.16 | — |
| Boxer | 10.04 | — |
| Beagle | 9.85 | — |
| Husky | 9.53 | — |
| American Bulldog | 7.79 | — |
| Bulldog | 7.39 | 9.8 |
| French Bulldog | 4.53 | 9.8 |
| St Bernard | — | 9.3 |
| Mastiff | — | 9.0 |
| Cane Corso | — | 8.1 |
| Presa Canario | — | 7.7 |
| Caucasian Shepherd Dog | — | 5.4 |
One label is worth a footnote: the life-table study records “Husky” rather than Siberian Husky, so that row may include other husky types. Otherwise the breed names are as each study reported them.
Two things jump out. The first is that the top of the table is small and the bottom is enormous. The second is the French Bulldog row.
The French Bulldog problem — and why it is the most useful part of this article
4.53 years versus 9.8 years. That is not a rounding difference; that is one study saying a breed barely reaches five and another saying it nearly reaches ten.
The life-table authors explain it themselves, and the explanation applies to every fashionable breed. UK Kennel Club registrations for French Bulldogs went from 2,771 in 2011 to 39,266 in 2020. When a breed’s population explodes like that, the dogs alive at any moment are overwhelmingly young. A cohort life table built from recent deaths therefore counts a great many deaths of young dogs and very few of old ones, because the old ones scarcely exist yet. The result systematically underestimates life expectancy for a breed on the way up — and, symmetrically, overestimates it for a breed in decline, whose surviving population skews old.
This is the single most useful thing to carry away from any lifespan article: a lifespan figure is a property of the population that was measured, not a property of the breed. Before you trust one, ask what it was calculated from, and whether that breed’s popularity has been stable.
The same paper is candid about three other limits worth knowing:
- Euthanasia. Euthanasia is common in companion dogs, and the authors flag that it pulls measured life expectancy below biological potential — shifting national figures with cultural attitudes as much as with medicine.
- Primary-care-only data. Dogs that never attend a vet, or that die at home, are under-counted.
- Country. Within the same breed, the numbers move enormously between populations: Labradors in the UK measured 11.77 years against 14.1 in a Japanese dataset; Beagles 9.85 against 14.8. Whatever is driving that gap, it is not the breed.
What actually moves the number
1. Size — and large dogs age faster, they do not start earlier
The size effect is consistent and, by the standards of this field, large:
| Body size | Median survival |
|---|---|
| Small | 12.7 years |
| Medium | 12.5 years |
| Large | 11.9 years |
The interesting question is why. A study fitting mortality models to death data across 74 breeds separated the possibilities: do large dogs die young because more of them die as puppies, because senescence starts earlier, because their baseline risk is higher, or because they age faster once ageing begins? The answer was the last one. Onset of senescence did not track size; baseline hazard was only slightly higher; the strong relationship was between size and the rate of ageing. In the authors’ words, large dogs die young mainly because they age quickly.
There is a cancer component to this too. A 2024 analysis across three independent datasets found that lifetime cancer mortality across breeds fits a standard multi-stage model of carcinogenesis driven by body size — bigger body, more cells, more chances for the handful of driver mutations a tumour needs. Of 85 breeds present in more than one dataset, only the flat-coated retriever was significantly above what its size predicted, with the Scottish terrier, Bernese Mountain Dog and bullmastiff notably elevated. For everyone else, size explained it.
If you are choosing a large breed — and on this site, most people are — this is not a reason to choose differently. It is a reason to build the ownership plan around a shorter, denser life, and to take the joint and weight management seriously from month one. Our guide to how much exercise a dog really needs covers the load side of that.
2. Skull shape — the biggest single-variable gap in the data
| Skull shape | Median survival |
|---|---|
| Mesocephalic (normal muzzle) | 12.8 years |
| Dolichocephalic (long muzzle) | 12.1 years |
| Brachycephalic (flat-faced) | 11.2 years |
A 1.6-year gap between flat-faced and normal-muzzled dogs is already substantial. But skull shape and size interact, and that is where it gets stark. Against the longest-lived group — small, long-muzzled females at a median of 13.3 years — medium-sized brachycephalic breeds died 2.69 times faster, with a median of 9.4 years. Large brachycephalic breeds sat at 10.7.
An independent 2024 analysis of 28,159 deceased UK dogs, using a machine-learning classifier over veterinary records, found a 19% increased risk of premature mortality in brachycephalic breeds — a different dataset, a different method, the same direction.
This matters on a site about active dogs for a second reason. The largest study of heat-related illness in UK dogs, covering 905,543 animals, found brachycephalic dogs had 2.10 times the odds of heat-related illness of normal-muzzled dogs, with the effect showing up at relatively low temperatures and relatively low activity levels. We covered that evidence in full in how hot is too hot to walk your dog. Lifespan and exercise tolerance are, for flat-faced breeds, the same conversation.
3. Sex, and the neutering asterisk
Females outlived males in both studies: 12.7 against 12.4 in the pooled data, 11.41 against 11.07 in the life tables. The gap is real but small.
The life-table authors add a caution that is easy to misread in either direction. Neutered dogs of both sexes showed higher life expectancy than entire dogs — but neutered dogs have, by definition, already survived to the age of neutering, which biases the comparison upward. Neutering may also stand in for a generally more managed life. Their own conclusion was that their neutered-dog life tables “should be interpreted with great caution,” and that is the honest position to repeat. Talk to your vet about the timing decision for your individual dog; it is not a longevity lever you can read off a table.
The one lever that is actually yours
Everything above is settled at purchase. This part is not.
In 50,787 middle-aged neutered dogs attending roughly 900 North American veterinary hospitals, dogs in overweight body condition had a higher instantaneous risk of death than normal-weight dogs of the same breed and sex — in every one of the 12 breeds studied. The hazard ratios ran from 1.35 in German Shepherds to 2.86 in Yorkshire Terriers. Median life span was shorter in all 12 breeds:
- Yorkshire Terrier: 13.7 years overweight vs 16.2 normal — a 2.5-year gap
- German Shepherd Dog: 12.1 years overweight vs 12.5 normal — a 0.4-year gap
That range is worth sitting with. Even at the small end, it is four to five months of life for something that costs nothing.
The experimental evidence points the same way. A lifelong paired-feeding study followed 48 Labrador Retrievers from eight weeks of age to death, with one dog in each littermate pair fed 25% less than its pair-mate. Reviewing the full two decades of data, the researchers reported 1.8 years longer median lifespan in the restricted group, along with delayed onset of late-life disease — osteoarthritis especially.
Two very different study designs — an observational dataset of 50,000 pet dogs and a controlled trial of 48 — agreeing that body condition moves lifespan by months to years. There is no other owner-controlled variable with that weight of evidence behind it.
This is educational information, not veterinary advice. Do not put a dog on a restricted diet on the strength of an article. Body condition scoring is something to do with your vet, who can tell the difference between an overweight dog, a heavily muscled one, and a dog whose weight is a symptom of something else.
The number that actually applies to your dog
Averages at birth are the wrong tool once you own an adult dog, because life expectancy is conditional on the age already reached. From the UK life tables:
| Your dog’s age | Average years remaining |
|---|---|
| 0 (birth) | 11.23 |
| 4 | 7.87 |
| 6 | 6.18 |
| 8 | 4.61 |
| 10 | 3.28 |
| 12 | 2.23 |
| 14 | 1.46 |
| 15 | 1.19 |
| 17 | 0.87 |
An eight-year-old dog is not “three years from the average.” It has, on the population evidence, 4.61 years left. Every year survived pushes the expectation out, because the dog has demonstrably avoided the things that killed its cohort-mates.
This is also the honest frame for the hard conversations. A 12-year-old dog averaging 2.23 more years is not living on borrowed time; it is living in the part of the curve where planning — mobility support, weight, pain management, more frequent check-ups — buys the most.
What this changes about choosing a dog
Five things, in descending order of how much they move the number:
- Skull shape is the biggest single choice. Flat-faced medium breeds sit at a median of 9.4 years against 13.3 for small long-muzzled ones. Nothing else on the list of things you choose comes close.
- Size is the second. 11.9 versus 12.7 years, and the mechanism is a faster rate of ageing, not a weaker start.
- Breed folklore is worth less than the clade. Lifespan showed a very strong phylogenetic signal (Pagel’s λ = 0.808) — breeds within the same evolutionary cluster have similar lifespans, which is why the mastiff-type clades occupy the bottom of the table together rather than at random.
- Where the number came from matters as much as what it says. See: the French Bulldog.
- Then everything you do afterwards. Body condition first, by a distance.
If you are weighing specific breeds, our individual guides carry the health picture in detail — the Labrador and Golden Retriever at the popular end, the Border Collie and Australian Shepherd among the working breeds, and the Siberian Husky and Alaskan Malamute among the northern ones. If you are choosing a first dog rather than a specific breed, start with the best adventure breeds for first-time owners, which deals with the same evidence from the other end.
And if the question underneath yours is really “are mixed breeds healthier?”, that one has its own answer, and it is less obvious than the internet suggests: we went through the evidence here.
The bottom line
The median dog lives about 12.5 years. Your dog’s number is set mostly by three things: how big it is, how long its muzzle is, and whether it stays lean. The first two were decided before you met it. The third is a decision you make every day, and it is worth between four months and two and a half years.
Sources
- McMillan KM, et al. Longevity of companion dog breeds: those at risk from early death. Scientific Reports 14, 531 (2024)
- Teng KT, et al. Life tables of annual life expectancy and mortality for companion dogs in the United Kingdom. Scientific Reports 12, 6415 (2022)
- Kraus C, Pavard S, Promislow DEL. The size–life span trade-off decomposed: why large dogs die young. The American Naturalist 181(4) (2013)
- Salt C, et al. Association between life span and body condition in neutered client-owned dogs. Journal of Veterinary Internal Medicine 33(1) (2019)
- Lawler DF, et al. Diet restriction and ageing in the dog: major observations over two decades. British Journal of Nutrition 99(4) (2008)
- Kealy RD, et al. Effects of diet restriction on life span and age-related changes in dogs. JAVMA 220(9) (2002)
- The effect of body size and inbreeding on cancer mortality in breeds of the domestic dog. Royal Society Open Science (2024)
- Premature mortality analysis of 52,000 deceased cats and dogs exposes socioeconomic disparities. Scientific Reports (2024)
- Hall EJ, Carter AJ, O’Neill DG. Incidence and risk factors for heat-related illness (heatstroke) in UK dogs under primary veterinary care in 2016. Scientific Reports 10, 9128 (2020)
Frequently asked questions
What is the average lifespan of a dog?
Across 584,734 UK dogs, median survival was 12.5 years. A separate UK study using veterinary records from 30,563 dogs that died between 2016 and 2020 put life expectancy at birth at 11.23 years. Both numbers are right; they measure slightly different things, and neither is a prediction about your dog — size and skull shape move the figure by several years.
Which dog breeds live the longest?
In the largest breed comparison to date, the longest median lifespans went to the Lancashire Heeler (15.4 years), Tibetan Spaniel (15.2), Shiba Inu (14.6) and Papillon (14.5), followed by a cluster tied at 14.2 — the Lakeland Terrier, Schipperke and Border Terrier. The pattern behind the list matters more than the list: small dogs with normal-length muzzles live longest, and the longest-lived group of all — small, long-muzzled females — reached a median of 13.3 years.
Do big dogs really have shorter lives?
Yes, and the mechanism is specific. Large breeds had a median of 11.9 years against 12.7 for small breeds. A study of mortality curves across 74 breeds found that large dogs do not start ageing earlier — they age faster once ageing begins. The authors' conclusion was blunt: large dogs die young mainly because they age quickly.
Does being overweight actually shorten a dog’s life?
It is the best-evidenced thing an owner controls. In 50,787 middle-aged neutered dogs across roughly 900 North American veterinary hospitals, overweight body condition raised the risk of death in every one of the 12 breeds studied, and median life span was shorter in all 12 — from a 0.4-year gap in German Shepherds to a 2.5-year gap in Yorkshire Terriers. A separate lifelong feeding trial in Labradors found 25% food restriction added 1.8 years to median lifespan.
Why do different sources give completely different lifespans for French Bulldogs?
Because one method is thrown off by a breed's popularity curve. A life table built from recent deaths gave French Bulldogs 4.53 years; a survival analysis of a much larger, mixed-source dataset gave 9.8. UK Kennel Club registrations for the breed rose from 2,771 in 2011 to 39,266 in 2020, so the population is overwhelmingly young — and a life table built on that population counts a great many young deaths and very few old ones. The study's own authors flag this.
How long does my 8-year-old dog have left?
More than subtracting from an average suggests. Life expectancy is conditional on the age already reached: in the UK life tables, a dog that has made it to 8 has an average of 4.61 years remaining, a 12-year-old has 2.23, and a 15-year-old still has 1.19. Every year survived moves the finish line.
Comments
Got a question, a tip, or a correction? Leave a comment — no account, no sign-in. Just your name and your thoughts.
Loading comments…