Monday 10 October 2011

Lifespan: the long and the short of it

A comment on a Bulldog blog caught my attention this evening. It claims that the Kennel Club has "updated" the average lifespan on its Bulldog breed info page from six to 10.

I was surprised because the 2004 KC health survey found that the median age of death in Bulldogs (based on a good-sized sample of 180 dogs that had died) was 6 yrs and 3 months - not great. As far as I'm aware, there has been no decent survey done since that supercedes this figure,  so I hot-footed it over to the KC website to see what it says.

As you can, it lists Bulldog lifespan as "Under 10 years". Now this is true  - but not terribly helpful for anyone seeking accurate information.

In fact, the KC breed info pages gloss over the longevity issue by listing all breeds' lifespans as either "under 10 years" or "over 10 years".  I guess the truth - that some breeds die, on average, a lot younger than 10 - is just too unpalatable, but it is doing puppy buyers a disservice.

After all, the KC health survey found that the median age of death for the Neapolitan Mastiff is 2.33 years. It would be a bit of a shock, wouldn't it, to lose a dog at that age if you hadn't been pre-warned? Particularly when you've forked out £1000 or more for a Mastino pup.

So here, then, is a list of the Top 21 shortest living breeds according to the KC's own 2004 health survey - and,as you see, the median age of death in some breeds is a real cause for concern.

NB: I have excluded the Bracco Italiano as the survey only included one dog that had died, and the stats for the first six breeds are based on very small sample sizes and should be read with caution. A larger survey for Dogue de Bordeaux (based on 79 dead dogs rather than the KC's five) puts longevity at 5.3. Usually, the bigger the sample size, the better - although not if the survey methodology is flawed.

John Armstrong, in an internet survey of Standard Poodle mortality collected data on nearly 1000 dogs but found that means changed very little after 150-200 dogs.)  

Russian Black Terrier - 1.79
Neapolitan Mastiff - 2.33

Dogue de Bordeaux - 3.83
Kooikerhondje - 3.92
Grand Bleu de Gascoigne - 4.54
Pyrenean Sheepdog - 5.79

Miniature Bull Terrier - 6.08
Shar-pei - 6.29

Bulldog - 6.29
Great Dane - 6.50 

Bloodhound - 6.79
Mastiff - 6.83

Shiba Inu - 7.00
St Bernard - 7.00
Irish Wolfhound - 7.04
Leonburger - 7.08
Finnish Lapphund - 7.33
Bullmastiff - 7.46
Bernese Mountain Dog  - 8.00
Nova Scotia Tolling Retriever - 8.00
Cesky Terrier - 8.42

However, on its breed info pages, the KC lists 11 of these 21 breeds as living for "Over 10 years" - including six that are among the eight very shortest-living breeds, including the Dogue de Bordeaux, Kooikerhundje, Grand Bleu de Gascoigne, Miniature Bull Terrier and the Shar Pei.  (The KC also insists that both the Toller and Cesky live longer than 10 despite its survey findings.)

In fact, there are many other discrepancies on the KC's breed info pages as regards lifespan - with many listed as enjoying 10 years or more when the KC's own health survey found a median age of death under 10. This needs to be rectified - unless of course, there are more recent surveys of sufficient quality that shows a longer lifespan.  I'd be happy to include them here if anyone wants to send me the them.

Here, meanwhile, are the 37 breeds who live at least twice as long as the average Bulldog.

Siberian Husky - 12.58
Welsh Springer Spaniel - 12.58
Collie - 12.67
Dachshund - 12.67
Norwegian Buhund - 12.67
Welsh Terrier - 12.67
Beagle - 12.67
Yorkshire Terrier - 12.67
Staffordshire Bull Terrier - 12.75
Whippet - 12.79
Australian Terrier - 12.80
Manchester Terrier - 12.83
Briard - 12.88
Bichon Frise - 12.92
Hovawart - 12.92
Hungarian Viszla - 12.92
West Highland White Terrier - 13.00
Miniature Pinscher - 13.00
Parson Russell Terrier - 13.00 
Papillon  13.08
Fox Terrier - 13.13
Norwegian Elkhound - 13.17
Bedlington Terrier - 13.38
Norwich Terrier - 13.38
Bearded Collie - 13.50
Italian Greyhound - 13.50
Basenji - 13.54
Miniature Poodle - 13.92
Border Terrier - 14.00
Cairn Terrier - 14.00
Swedish Vallhund - 14.42
Canaan Dog - 14.63
Australian Silky Terrier - 14.25
Lhasa Apso - 14.33
Toy Poodle - 14.63
Irish Terrier - 14.83
Lakeland Terrier - 15.46

And guess what? No mistakes there. The KC lists them all accurately as living for more than 10 years.

Individual breed results from the KC's 2004 health survey are available here.

More breed longevity data can be found on Kelly Cassidy's Canine Longevity website (in need of an update but containing much interesting information).


  1. hang on a minute - have you looked at the number of dogs actually used to get those figure ! - in Russian Black Terriers the 'average lifespan ' was calculated from only 4 deaths !! - in Koikers it was from only 7 and in Grand Bleu's only 6 - how can any meaningful lifespan average be deduced form such small smaples ? ..... in fact the site itself says

    " Warning: The results of this survey and particularly the breed-specific analyses
    should be interpreted with caution. The overall response rate was
    only 24% with breed-specific response rates from 4.5% to 64.7%. "

  2. Yes, you're right, Bijou - as I address above (published slightly prematurely last night and have added more this morning).

    It is indeed hard to deduce much from such small sample sizes - other than that larger surveys need to be done to produce more meaningful longevity data. Unfortunately, in many of the short-lifespan breeds, this is not being done and until such time as it is, this is all that is available.

    1. Hard to deduce from a small sample? It's impossible to deduce ANYTHING from a sample that is not representative. Unless of course writing a tabloid article

  3. Jemima, many of the breeds in your first list are newly established in the UK. This gives a biased sample for longevity assessment because the population is young.

    The proportion of aged dogs is smaller than would be in an established breed, so premature deaths, that occur in all breeds, are over represented, giving a shortened average lifespan.

    The change in the Bulldog profile, yep really misleading. Not a good move by the KC.

    Julie Vaughan

  4. To me this points to the largest single flaw in the Pedigree dog system: recording births but ignoring deaths. It's sad that lots of attention goes to this or that measurable health defect, but the summary data on mortality is not systematically collected. I tried to report the death of my first pedigree dog to the KC and they looked at me like I was crazy. Apparently many breeders have hundreds upon hundreds of dead dogs registered in their names.

  5. "The shorter-lived a breed, the greater the reluctance among breed fanciers to accept how short-lived the breed really is. Hence, the shorter-lived the breed, the more 'padding' is added to the longevity estimate. Breed fanciers may like to believe they have a healthy breed and might be inclined to feel the longer-lived individuals are more representative of the breed than dogs that die young."

    I think that pretty much describes it.

    At one time, it was said that golden retrievers lived 13 to 15 years. It's in the earliest edition of Bruce Fogle's The Encyclopedia of the Dog.

    The reality is that 10 to 13 is a more realistic age, and if you're in North America, 10 is closer to reality than 13. The GRCA did a very good longevity study a few years back ( The lifespan was closer to 10 years.

    Fogle has commented on my blog, and that his statistics of goldens at his practice are closer to the 13 mark. I lost one at 13 and the other 14.

    But I'd be lying if I said that 13-14 was the average life expectancy of the breed.

    They were outliers. The oldest golden retriever on record was 19.5.

  6. Vicky Collins-Nattrass10 October 2011 at 17:41

    In over 20 years of bulldog owning I have had 7 bulldogs live beyond 10 and at present have a 13 and a half rescue dog ,the oldest one personally known to me but not mine was 14 when she died and a multi CC winner at that.The KC survey was only given to 1 of the 17 Breed Clubs and had a very poor response .FACT

  7. so after your 'indepth' research does this now mean that you will apologise to the owner of the 12 year old bulldog you tore into on your youtube channel shortly after the first 'programme'?

    as a footnote Wendy, the bulldog in question died peacefully at home aged 13 (THIRTEEN).
    she was, sadly, one of seven bulldog deaths reported this year, all in their early double figured age groups (11-13).

    so it would appear that the bulldog has, if your updated postis as accurate as you lead your readers to believe, the normal living age of many dog breeds

  8. Also, breeds that were already in numerical decline by 2003, the year of data collection, may have a biased data sample because they are over represented with old dogs. This would skew the lifespan average by lengthening it. The Neapolitan Mastiff and the Bullmastiff for example.

    This is based on registration stats from 2001. I can't find any for before that year.
    Julie Vaughan

  9. I am shocked by the terribly short lifespan of the Shiba Inu in the UK. The Shiba Club of America conducted a longevity study which had an average age of death well into the teens (or so I thought). Do the UK breeds clubs have any longevity studies for their breeds?

  10. I would just like to say you are an incredibly narrow minded, uneducated woman. I bet this response doesnt get posted, but you genuinely have your blinkers on when it comes to so many breeds. Statistics mean sweetF.A. As they dont actually delve into what i call 'the real dogs' IE: the bulldogs that live quite happily in peoples homes, not the showring until they are 12 or 13 years old. Maybe you should come along to one of the many national bulldog walks that are organised, to see what a fantastic HEALTHY breed this dog REALLY is... after 3 or 4 hours of walking, running and swimming, it is more likely the owners are more exhausted than the dogs.

    But noooo you wouldnt do that would you as you wouldnt want to be proven wrong.

  11. Julie Vaughan's observations about using
    survey data from finite populations/participants for "life expectancy" are right on! Life expectancy calculations are extremely complicated - I urge everyone to read up on them a little. They can not be determined by simply surveying a group of dog owners who may or may not have experienced the deaths of their dogs! Please stop making sweeping conclusions based on these silly data sets.

  12. Here we go again - another blinkered article from a vile and uneducated woman! For goodness sake, give it a rest will you! What exactly have you got against the bulldog? Why do you think it is your god given right to slag this breed to high heaven? Why don't you concentrate on brushing up your reporting skills, because it's quite obvious that the info in this article is very one sided and inaccurate.

  13. Finnish Kennel Club collects the cause of deaths and dates also. Owners report by themselves. Koiranetti (unfortunately only in Finnish) shows that:

    English Bulldogs dies at the age 5 years 3 months (113 dogs)
    Russian Black Terrier: 4 y 6 m (62)
    Neapolitan Mastiff : 4 y 11 m (17)
    Dogue de Bordeaux: 6 y (54)
    Kooikerhondje: 7 y 4 m (29)
    Pyrenean Sheepdog: 10 y 10 m (39)
    Miniature Bull Terrier: 3 y 8 m (6)
    Shar-pei: 5 y 1 m (37)
    Great Dane: 6 y 4 m (306) (yellows only)
    Bloodhound: 7 y (50)
    Mastiff: 6 y 1 m (38)
    Shiba Inu: 5 y 4 m (13)
    St Bernard: 5 y 9 m (55) (short haired)
    Irish Wolfhound: 5 y 9 m (218)
    Leonburger: 7 y 1 kk (318)
    Finnish Lapphund: 10 y (505)
    Bullmastiff: 6 y 9 m (216)
    Bernese Mountain Dog: 6 y 11 m (491)
    Nova Scotia Tolling Retriever: 7 y 11 m (216)
    Cesky Terrier: 8 y 9 m (14)

    Intresting isn't it? How sad it is that in many breeds the estimated lifespan is so short?

  14. Why do you not just "go do one". You are clearly a very small minded lady with far too much time on your hands.
    Go get a job or do some work in the local PDSA shop with all this spare time on your hands.
    Leave the British Bulldog alone.

    You silly small minded twit.

  15. Vicky Collins-Natress wrote: "The KC survey was only given to 1 of the 17 Breed Clubs and had a very poor response .FACT"

    841 questionnaires were sent out and they got back 143. That's only a 17 per cent response rate - but they did get back data on 180 bulldog deaths and that is more than enough to be able to draw some conclusions (always subject to review give more data, of course). As I have mentioned above, the figure is unlikely to change much once you get to 150 dogs or more.

    That Bulldogs can potentially live longer (the oldest one in the KC survey was 14.42yrs) makes the average of 6.29 found in the KC survey all the sadder in my view.

    Really, the ad hominen attacks aimed at me and others who argue that things need to improve are fooling no-one. This is NOT beating up the breed - it is intended to urge good and committed breeders to recognise that there is considerable room for improvement. Some do of course, as my private email bears testament.

    As ever, anecdotal evidence about longer-living bulldogs is interesting but to go forward we need good data based on a decent-sized, representative cohort of dogs. The KC survey does indeed suffer from low sample sizes in many breeds - but not really in the Bulldog.

    You can't prove to me or any other thinking person that the bulldog is healthy by calling me a twit, uneducated or caring. It is just stupid.


  16. Dear god Ms Harrison! Shut up and go away. You really have nothing important to say. All your data is an absolute crock and anyone with half an ounce of common sense will see it and you for what it is. Go seek attention somewhere else.

  17. It's the KC's data, not mine Anon.


  18. Can I ask?? How many of these dogs surveyed died of natural causes? because if they havent then the figures are ludicrous IMO

  19. Alistair wrote: "so after your 'indepth' research does this now mean that you will apologise to the owner of the 12 year old bulldog you tore into on your youtube channel shortly after the first 'programme'?"

    What bulldog would that be?

    This is the only Youtube video on bulldogs we've posted - an extended cut of the sequence in PDE.


  20. Retrieverman said the average lifespan for Golden Retrievers was 10-13 years. I am afraid that here in Austria it is lower....rather like 7 to 8 years, for the same reasons as the Flatcoats: cancer.

  21. Anonymous said to Jemim:
    I would just like to say you are an incredibly narrow minded, uneducated woman.

    It is very easy to be abusive when you hide behind an Anonymous.
    However, I believe that it was clearly stated that the KC statistics might be questionable.
    Nonetheless, the only Bulldogs I have seen which can run and breathe even in hote weather are the Continental Bulldogs.

  22. Anonymous said...
    Dear god Ms Harrison! Shut up and go away. You really have nothing important to say. All your data is an absolute crock and anyone with half an ounce of common sense will see it and you for what it is. Go seek attention somewhere else.

    You don't have much to say either, it seems to me....and again, you are one of those who hides behind the anonymous. that is not only cowardly, it makes your messages worthless. If you don't like what you read, why don't you direct your attention somewhere else?

  23. as Disraeli said "Lies, damned lies, and statistics" perhaps that shouldn now be "Lies, damned lies,statistics ..........and Jemima Harrison's logic"

  24. Any reason why the German shepherd dog is not included in the stats? It's not as if they are rare.

  25. re youtube: so 'dorothysdog' isnt another alias for you on the internet

  26. The University of Georgia has bulldog mascots. They are all named Uga.

    Every couple of years, the university's bulldog dies of something.

    The most recent one was only 17 months old!

    Here's the list of all the Ugas:

    Since the 1960's, not a one of them has made it to ten. And the last two didn't make it to their second birthdays.

    You can't tell me there's nothing wrong with bulldogs.

    It's ironic, too, because the state of Georgia has native working bulldogs, the Alapaha and the "White English," that are nothing like these dogs.

  27. Dorothea,

    There are some lines of golden in the United States that are notorious for developing bone cancers before they are two years old.

  28. What's the deal with Black Russian terriers?

    I know they were created under a Lysenkoist breeding program, but I haven't seen how this has affected their population structure.

    And if you're doing Lysenkoist breeding schemes, wouldn't you be doing less inbreeding?

  29. @Anon 23:53

    You gotta be delusional. Just because someone took a random footage of a dog show and is critical of it, it doesn't necessarily correlates with a public figure like Jemima Harrison. Has it ever occurred to you, it might be an advocate who has no ties with Pedigree Dogs Exposed?

    I can tell you within my own breed, people are always deriding the Corgis' conformation; and they are quite open about it. There is no anonymity. One would think because the breeders are upholding the same registries, they wouldn't make potshots at other breeds.

    Similarly, some of the show-ring breeders who keep in contact are constantly ridiculing "ruined" breeds like German Shepherds, English Bulldogs and so on.

  30. Alistair asked: "re youtube: so 'dorothysdog' isnt another alias for you on the internet"

    No. I have no aliases on the internet. I only ever post, anywhere, using my own name.


  31. "Can I ask?? How many of these dogs surveyed died of natural causes? because if they havent then the figures are ludicrous IMO"

    8.9%(16 individuals) of the bulldogs in the survey died of "old age".( Across all breeds, 17.8% die of old age.( The average lifespan of bulldogs that die of old age is about 10 to 11.25

    "Any reason why the German shepherd dog is not included in the stats? It's not as if they are rare. "

    Those breeds with response rate less than 15% did not get their survey results published.(

  32. That's really funny.

    As if Jemima invented the notion that there's something not kosher about deformed, disabled, and incest-bred show dogs. No one ever thought of this before. She invented it whole cloth. And is the only human with a video camera. If we could just make her go away, everything would be perfect again!

    Boy, there sure are a lot of brave Anonymice out there calling names from under their rocks and leaving spittle spray on the inside of our computer monitors.

  33. Heather asked: "Any reason why the German shepherd dog is not included in the stats? It's not as if they are rare."

    Here's the reply I got from the KC re this in 2008:

    "The GSD is interesting and disappointing. Essentially the return from the original survey was 0 and we don’t know why. We did receive a number of boxes of unopened questionnaires back at the Animal Health Trust, but no indication where they came from. One possibility was that the GSD Club we chose decided not to participate and returned them. We did chase this up with the club and, of course, they denied any knowledge. So, I had another 650 questionnaires and had them sent out by a second club; unfortunately, the return rate from this second survey was less than 5%, well below our lower threshold."

    I believe this might be a reflecton of the long-standing hostility between the KC and the gSD breed clubs more than a lack of GSD breeders' commitment to health which is among the best - at least in doing availale health tests (for what they're worth).


  34. The Finnish KC pedigree database lists the average age of death for the German Shepher as 7,5years (2207 dogs - and no typo, that's over 2000 dogs). The average age of German Shepherds that die of old age is 11,5 years (424 dogs).

  35. Goodness. That is low.

    The Finnish database, available free to the public, is an amazing resource. Are owners/breeders actively encouraged to report dog deaths? Such a no-brainer - absolutely something that needs to be done here in the UK, too.

    Clearly there isn't 100 per cent take-up, though. What is the culture there in Finlnd re reporting deaths, Maija?

  36. Finns are painfully honest people ;). The Kennel Club has almost 145 000 members who all have access to their own dogs info in the database so they can report the deaths and reasons by themselves via internet. And most of the active dog people does it.

  37. Jeesh - one would think that you want dogs to live to a 100. Remember that these deaths are not all HEALTH related. So also not an indication of longevity IMO. What about the accidental deaths also reported here or does that not score any points? Surely it should be split?

  38. Like I posted earlier there are 2207 German Shepherds with cause of death in Finnish KC database. If you leave out the
    - 130 dogs listed as accidental deaths (averarge age of death 4yrs 3months)
    - 9 dogs reported missing
    - 1 dog killed by a predator (bear, wolf ect)
    - 67 dogs put down for behaviour related causes (average age of death 4yrs)
    - 398 dogs with no cause of death given (average age of death for those 7yrs 4 months)
    - 424 dogs who are reported to have died of old age (mentioned in my earlier post, average age of death 11yrs 6months.)

    You get 1178 German Shepherds dying of health related causes, that's more than half (53% ) of the total. The most common ones are:
    - Cancers and tumours (257 dogs, 8yrs 9 months)
    - Bone or joint diseases including athritis in hips & elbows (255 dogs, 4yrs 9months)
    - Back diseases including spondylosis and joint deformities (100 dogs, 7yrs)
    - Liver and intestinal diseases including bloat, megaesofagus and liver hypofunction (69 dogs, 6yrs 3months)
    - Skin and ear diseases including allergies and atopies (56 dogs, 4yrs 11months )

    Also likely some of those with no cause of death listed have also died of health related reasons. Granted that is only one breed in one country but the sample size is large enough.

  39. Huh. I question the diagnosis of "old age" at 11.5 years for a GSD. (Both mine lived to be 13.5, and died of discrete causes without being generally "elderly" in physical or mental condition.) Especially since that's given as an average, so clearly dogs of nine or ten are included in that group.

    I also question orthopedic and such things as allergies as causes of death at a young age, in that I think it is unlikely that these health problems actually killed the dogs, but moreso that the owners either could not afford to treat, or chose not to treat, chronic conditions.

    Which I point out not to object that GSDs are, in fact, really and truly a long-lived and healthy population, but to highlight some of the difficulties in compiling mortality statistics for animals that can be euthanized, often long before they would have died naturally.

    Nevertheless, absent some evidence that owners of different breeds of dog are uniformly more or less likely to throw in the towel prematurely on a dog (or more or less likely to kindly offer a way out, depending on one's point of view), the relative figures within one study should be able to provide a basis for comparison.

    IOW, if a statistically significant sample of Shelties is living a bunch longer than a statistically significant sample of Great Danes, we can likely conclude that Shelties generally live longer.

  40. Thank you Maija, stats set out like this is much more useful than throwing all into the same pot to brew up a stew! I also have to agree with Heather that some owners give up on their animals way too soon and euthenize them at a early age. What is worrying though is the number of dog (67) put to sleep because of their personalities. Maybe not enough research about what they need to function normally. Not raised properly etc. More education to people on how to raise and care for animals is needed. 67 killed for this reason is 67 too many IMO.

  41. Re Finnish KC database record of deaths:

    Some of those dogs that are listed as having died of illness X are certainly put down. Then again if the rest of the life of the dog would be misery then in my opinion it is better to put that dog down. In Finland people treat their dogs usually quite well, take them to vet if necessary and lots of them have insurance for the veterinary costs. So if a dog is put down because of hips etc. that usually means that the quality of life of the dog (even with veterinary treatment) would be so low as to be not worth it (and I don't mean in financial terms).

    So it is true that many breeds have the potential to live until way over 10-12-14yrs and some individuals in all breeds do (even if they have medical condition). However I think it is important to know how large percentage of the breed live as long as possible with no faults or illnesses that cause the quality of life to lower.

    Below are couple of comparisons from the Finnish KC database. The dogs that are reported to have died of old age are included in the average. While it is true that this method of data gathering is not ideal but it is the same for all breeds.

    Labrador Retriever:
    - 9 yrs 6 months on average (847 dogs)
    - 12yrs 8 months - old age (271 dogs)

    (English) Cocker Spaniel
    - 9yrs 3 months on average (613 dogs)
    - 12 yrs 4 months - old age (133 dogs)

    - 8yrs 11months on average (308 dogs)
    - 12yrs 3 months - old age (74 dogs)

    - English Bulldog
    - 5yrs 6 months on average (113 dogs)
    - 10yrs 2 months - old age (16 dogs)

    West Highland White Terrier
    - 10yrs 5 months on average (139 dogs)
    - 13yrs 3 months - old age (51 dogs)

    Fox Terrier, Smooth
    - 8yrs 5 months on average (115 dogs)
    - 11 yrs 7 months - old age (26 dogs)

    Dachshund, normal, rough hair
    - 9yrs 7 months on average (353 dogs)
    - 13 yrs 8 months - old age (93 dogs)

    - 8yrs 6 months on average (268 dogs)
    - 12yrs old age (52 dogs)

    - 9 yrs on average (152 dogs)
    - 12 yrs 3 months - old age (41 dogs)

    Poodle, dwarf, all colours
    - 10yrs 8 months on average (180 dogs)
    - 14 yrs 3 months - old age (47 dogs)

    Irish Wolfhound
    - 5 yrs 6 months on average (218 dogs)
    - 9 yrs 1 month - old age (14 dogs)

    - Whippet
    - 9yrs 7 months on average (262 dogs)
    - 13 yrs 5 months - old age (85 dogs)

    And if you look at the three of the Finnish national breeds you see that there's not much sugarcoating going on:

    Finnish Hound (a Finnish hunting hound breed, rare outside Finland, height around 55-60cm)
    - 7 yrs 1 month on average (1730 dogs)
    - 10yrs 7 months - old age (362)

    Finnish Lapphund (also in Jemima's original post, height around 45-50cm)
    - 10 yrs on average (505 dogs)
    - 13 yrs 10 months - old age (133 dogs)

    Karelian Bear Dog (a Finnish hunting spitz breed, height around 55-60cm)
    - 6yrs 6 months on average (707 dogs)
    - 10yrs 4 months - old age (140 dogs)

  42. Ah - Maija you are a saint! I am sure that globally the same type of trends would be seen, although I would think that maybe there is factors such as the type of weather etc. that is better for some than for others? Also just as an observation - the pug also being on the list of breathing difficulty concern, seems to be quite healthy? your input is invaluable. Thank you.

  43. I also meant to include the much discussed Cavalier King Charles Spaniel:
    - 8 yrs 11 months on average (674 dogs)
    - 11 yrs 6 months - old age (117 dogs)
    - 9 yrs 1 month - heart disease (162 dogs)
    - 6yrs 5 months - disease of the nervous system including syringomyelia (22 dogs)

  44. I must admit I would never have thought of reporting my dogs death to the KC! I think the way data was collected for these surveys could be flawed as certain groups tend to be more likely to reply; perhaps those who are upset at the short lives of their dogs, or who are proud of their exceptionaly long lived examples. Too many of either in a breed ould skew the data.
    Maybe if we had compulsory registration (so reporting death or sale would be compulsory too) we would have more accurate data across all dog breeds, crosses and types?

  45. However if the dog is chipped , you can update with Pet Log when a dog has passed .They reply with a sympathetic message and presumably hold the data , so a good source of records is already there.

  46. A lot of owners never report to the chip companies that their dogs have died. After all, there is little incentive. But I agree Pet Log/Avid etc could help in gathering mortality data.


  47. Jemima, Thanks so much for this excellent blog. You must have thick skin to publish and overcome the terrible comments from your uneducated detractors. Good for you for sticking with this.

    I've been showing and competing in dog sports since 1977 in the USA, UK and now Turkey. I don't breed, but owe thanks to the ethical breeders of my dogs, each of which does OFA-certified genetic testing on their breeding dogs, and who generally breed from older dogs to verify no early-onset health problems.

    I love my dogs and invest hundreds or thousands of hours (and $$$) into their competition careers. I've looked everywhere for longevity data, and thank you for making it available.

    In my 34 years of competiion experience and giving classes to hundreds of dogs of many breeds, I concur with your observations about the cruelty of breeding unnatural dogs, such as bulldogs that cannot naturally whelp, breathe, participate in sports, or even walk without collapsing. One of these brachyencephalic dogs collapsed and died at the inaugural Turkish FCI conformation show earlier this year. (Turkey is the newest FCI member).

    I'm totally PRO-bulldog, which is why I find it sad that breeders are taking a dog with a great, active mind, and forcing that mind to be housed in a substandard body. The same goes for all brachyencephalic dogs, dogs that are achondroplastic dwarves, dogs with entropion or with bulging eyes or too-narrow heads, etc.

    The only purebreds I own are "natural" breeds, such as Border Collies, Australian Cattle dogs, and my current addition, a Papillon. Paps do have unnaturally large upright ears, but this doesn't affect their physical abilities in the least. The most successful agility dog in America is a Papillon with twice as many agility titles as the closest competitor (a Border Collie with the same owner).

    Breeders should feel free to breed dogs with interesting colors, personality traits that help them herd, hunt, etc.; different coat textures, different ear shapes, different markings, slightly different leg length, etc. But please leave dogs the ability to breathe, see, eat and reproduce without needing surgery!

    People who are breeding these dogs are practicing cruelty. I don't want to eliminate purebreds by any means, but want to change the breed standards to quickly fix the "broken" breeds and eliminate these serious structural and genetic issues.

  48. Jemima said ‘GSD breeders' commitment to health which is among the best - at least in doing available health tests (for what they're worth)’.

    If GSD breeders’ commitment to health is among the best, I fear for pedigree dogs, J. Yes, the breed pioneered the Hip Scoring scheme, they worked with the kennel Club on the DNA test for haemophilia A, they took steps to eliminate epilepsy by moving towards the German bloodlines but those were the breeders of yesteryear, and when the GSD Breed Council had teeth. Apart from a few consciousness GSD breeders still around, the breed as a whole did not embrace elbow scoring when it first came into affect in the late 90’s and the attitude towards health in latter years is generally disappointing. I remember when it wasn’t common to find GSD living to 16 years of age, now we’re lucky if they get past 10. This website was set up by a dedicated person in the USA a number of years back to record health issues and mortality rates in GSD worldwide - not enough use it.

  49. That was meant to say it wasn't uncommon to find GSD living to 16. Sorry...

  50. Could anyone please tell me where you have found the data on the average lifespan of the neapolitan mastiff? And how many dogs where included in this survey? I found nothing on the KCs website, nor on the breed longivety website. I am considering doing a research project on this, and would really like to know.

  51. MK, the Neo data can be found in this paper:

    It shows that 80 survey forms were sent out, nine were returned and it included death data on 7 dogs (so a very low sample size).

    The youngest Neapolitan died at 0.65yrs; the oldest at 16.08 (amazing!). Most, however, must have died young to get the median age of death of 2.33yrs.

    The Finnish database - Koiranet - shows that, based on death data for 17 dogs (so a bit better but still not many) the average age of death for the Neapolitan is 4yrs 11 months - higher but still extremely low.


  52. 2.33yrs - that dog is barely out of adolescence! You've only just finishe It's not just the considerable cost in buying, feeding, raising and vaccinating the puppy. It's also the hard work training and socialising it, not to mention being woken up in the night by a young pup needing to toilet.

    The whole point of getting a puppy is so that you can put the effort in whilst the dog is young, so you reap the rewards for the rest of the dog's (considerable) life.

  53. just a thought.. Some dogs may die prematurely due to radiation exposure from several sets of x- rays taken within a short space of time. can be a cause of lymphoma for example.

  54. I am quite sceptical about the reported ages at which dogs reputedly die. There is very little in the way of reliable data. In the UK, the KC appears to base its figures on the health surveys which were done through breed clubs. Breed club membership is heavily weighted towards breeders and owners of show bred dogs, while pet owners and owners of working dogs are far less likely to be members of breed clubs. In my experience dog breeders tend to mention their long lived dogs with pride but conveniently forget about the ones who died at an earlier age. If one could collect more data from pet owners, I suspect it would be rather more reliable than the data from breeders
    The only way to get reliable data would be to follow a very large cohort of dogs, from a carefully stratified sample ( different breeds, different kinds of owners, show, working, pet, registered and unregistered) , followed through from birth to death

  55. Absolutely agree. We need a Register of deaths as well as births of dogs - and for all dogs. And vets could play a big role here in encouraging the reporting.


  56. Without a standard of recording deaths, it's hard to get an unbiased picture. I suspect we're biased to base our expectations on the dogs we know who live to a ripe old age and to ignore the pup who gets run over at 18 months or has to be put to sleep at three due to bilateral HD and owners who aren't willing or able to pay for the surgery. . . . not to mention the many many puppies who don't make it to 8 weeks. We want to forget puppy tragedies, we are proud of the old survivors.

  57. I agree in 2004 the data relating to the Bracco Italiano would not be reliable but I wonder just how reliable it would be now too..given that a lot of dogs have died from torsion and, most recently, kidney failure. This is still a developing breed and it pains me to see these dogs dying so young and such terrible deaths. I have always thought there should be a "life register" for dogs ie birth, matings, health tests, health problems and death reporting. All the stats can be easily manipulated by those that are not keen to see their bloodlines tarnished with low mortality. I also agree that breed clubs are very weighted towards showing and breeders; the real truth is out there in the form of pet or working owners and that is where the effort needs to be applied. Pet owners will IME always tell the truth about their dogs......

  58. FAO Anon @23:00
    I have heard that some people have radiographs made of pregnant bitches for various reasons but as is well known x-rays can be harmful, especially to rapidly dividing cells, such as growth cells and gonadal cells. It's the primary reason why lead shields are used over reproductive organs and people under 18 as well as pre......gnant women should not be allowed in a radiographic suit when radiation is in use. Protection against scatter radiation, not necessarily exposure to the primary beam. X-Ray Debates
    (as seen in the AKC Gazette – June 2001 issue)
    As breeders we have certain things we "know," either gleaned by experience or from our mentors. Today, modern technology makes it possible to use X-rays on a pregnant bitch to determine the number of puppies and their size especially important information for Pugs, as the pups' large heads sometimes make Caesarian sections necessary.

    But what does the procedure do to the puppies? The general rule of thumb is that you do not X-ray early in a pregnancy. Wait 54 days, when the pups are fully developed and the fetal bones are calcified. By waiting until this time, it is believed that the radiation will not damage the unborn litter.

    Yet over the past 15 years, studies have been performed on dogs X-rayed at different growth stages. One study followed the lives of more than 1,600 dogs exposed only once at different stages of growth. The stages were eight, 28 and 55 days after conception, and two, 70 and 365 days after birth. There were two control groups of pups that were not X-rayed, and genders were represented equally in all groups. The study of those puppies radiated during the perinatal period (just prior to birth at 55 days and just two days after birth) is surprising.

    Four perinatally exposed dogs died of cancer prior to age 2. This is significantly higher than the normal canine population. Within the study group, including a non-exposed control group, 71 percent of all cancers and 56 percent of all tumorous growths that were reported during the first four years of life were found in 29 percent of the dogs exposed to radiation. This strongly suggests an increased risk of cancer or tumors in dogs that are X-rayed early in their lives.

    The overall numbers are rather frightening: 40 percent of these dogs died due to neoplasia or cancerous growths. There is an increase in the number of both benign and malignant tumors in dogs under 4 years of age. The majority of tumors reported in these studies were of three types: lymphomas,
    hemangiosarcomas and mammary carcinomas, which together comprised 51 percent of all the fatal tumors. Dogs radiated as fetuses also showed increases in the lifetime occurrences of early-onset lymphomas.

    A side effect of these studies is that puppies which were X-rayed within the uterus showed a significant reduction in immune system antibody responses at 12 to 16 weeks of age, and also had higher numbers of dogs with defects in thymus development.

    All of this is presented so that we can learn and grow as breeders. We all start out with certain information that is "known" to us and then make educated decisions within our breeding programs based on what we know at the time. It is only by continuing our own education and keeping up with the new information that breeders can learn more about the risks we take with our dogs and how to minimize them. Very few people could individually produce and study this large of a group of animals and be able to chart and report the results.

    So no matter what everyone else does, no matter what all the research tells us, we still have to make our own decisions based on the information we get from all sources. It's up to us as breeders to be as educated as we can, and to know the consequences of what we do to our dogs.

    Published sources for this article are available from me or via MedLine. The primary study was performed at Colorado State University Veterinary School.

    Marcy Heathman (Pugs Columnist)

  59. I for one think the work Jemima is doing is invaluable to the future of dog breeding in the UK. Naysayers aside, this information both shocks me and as an average (although better informed than most) pet owner, I am appalled that breeders have allowed this to happen to "man/woman's best friend!

  60. Another reason to track dog mortality and morbidity -- not just in purebreds -- is to identify risk factors, including, but not limited to, the kinds of things that concern Ms. Heathman.

    If the public health agencies in all countries were looking systematically (through veterinary reporting, say) at age of death and cause of death in domestic dogs, and gathering some additional data with those reports. they have a chance at catching serious problems for human health -- zoonoses, carcinogens, food contamination, and other environmental risk factors.

    Dogs are particularly sensitive miners' canaries for carcinogens and possibly teratogens.

    If we can include breed and bloodline in that epidemiology, then the community who cares about the genetic health of purpose-bred dogs will have much more reliable information with which to start working.

  61. The Colorado State study with the beagles was 'Radiation Carcinogenesis in Dogs Irradiated During Prenatal and Postnatal
    Development.' There are also a bunch of spin off studies, you can find them on Pubmed.

    From the study:

    "To evaluate the lifetime hazards of ionizing radiation exposure, 1680 beagles received whole-body, 60 Cobalt gamma exposures or sham-exposures during development. Eight groups of 120 dogs each received -mean doses of 16 or 83 cGy at 8 (preimplantation), 28 (embryonic), or 55 (late fetal) days postcoitus (dpc), or 2 (neonatal) days postpartum (dpp). One group of 120 dogs received 83 cGy at 70 dpp (juvenile), and one group of 240 dogs received 83 cGy at 365 dpp (young adult). Sham-irradiations were delivered to 360 controls. Sexes were equally represented. Young dogs, up to 4 years of age, had an increase in benign and malignant neoplasms after irradiation in the perinatal period at 55 dpc or 2 dpp. Among these, 4 fatal cancers were observed. No malignancies occurred in comparably-aged controls. The increase in both fatal neoplasms and all neoplasms in the perinatally-exposed groups were statistically significant. Over the full lifetime, dogs irradiated in the perinatal period also had the strongest evidence for an increased risk for fatal malignancies of all types. Though not as strong, there was a trend for increased risk for fatal cancer in dogs irradiated at all other ages. The risk of fatal malignancy after irradiation was greater in females than in males. Dogs exposed at 55 dpc had a significant increase in lymphoid neoplasia and dogs exposed at 8 and 55 dpc had increased risk for hemangiosarcoma. There was no evidence for an increased risk for mammary carcinoma in irradiated females. Dogs exposed as juveniles at 70 dpp had a significant increase in all benign and malignant thyroid neoplasms, including fatal thyroid carcinoma."

    Now, this sounds very frightening. Until you realize that your typical fetal exposure during an abdominal x-ray (in humans) is from 3 mGy (milligrays) to 15 mGy. The dogs in the study were exposed to a minimum of 18 cGy (centigrays), which is 180 mGy. To put it simply, the dogs in the study were exposed to considerably more radiation than your typical near the end of gestation bitch x-ray. Let's try to keep some perspective on that risk factor, okay?

    IMO, using this study to induce fear and make excuses for high rates of cancer in some breeds is intellectually dishonest. Our dogs deserve better from us.

    Resources (all regarding humans, sorry):

  62. To Jess
    but as is well known x-rays can be harmful, especially to rapidly dividing cells, such as growth cells and gonadal cells. It's the primary reason why lead shields are used over reproductive organs and people under 18 as well as pre......gnant women should not be allowed in a radiographic suit when radiation is in use. Protection against scatter radiation, not necessarily exposure to the primary beams.

    Funny though even at 3Miliigrays exposure that you state humans receive that the above have to be protected so imo even a lower dose would cause animals harm

  63. Dear Anonymous at 16 October 2011 21:26:

    It is a given that x-rays can be harmful. It is a matter of degree and dose, and if you had bothered to read the study OR the links, most of which pertain explicitly to x-rays ON PREGNANT WOMEN, you wouldn't have made such an asinine, moronic comment.

    It's just easier to let someone else tell you what to think, isn't it?

    A little knowledge can be a dangerous thing.

  64. At Jess Lordy lordy lol It amazes me how silly people like yourself think they can get away with calling posters morons and then I remember where you are posting. Read the part of statement I posted Doh!!! It was about non pregnant woman

  65. I never called anyone a moron. I said your statement was moronic. If you really feel that makes you a moron, that's your problem, not mine. Took the bait, though, didn't you?

    "Better to remain silent and be thought a fool than to speak out and remove all doubt." Abraham Lincoln.

    The reason x-ray technicians wear protective gear is because they are subject to REPEATED EXPOSURE. Nobody (I hope) takes a pregnant bitch in for an x-ray every other day for weeks, do they?

    In Germany, owners are allowed to hold their dogs during x-rays. The authorities passed a law that stated that the radiation exposure of the owners had to be measured. A study was done, and it was determined that the amount of radiation that these owners were exposed to while they held their dogs was so negligible that it was actually a waste of time to bother measuring it. I'd post a link to the abstract, but I know you won't go look at it.

    That article is nothing but scaremongering and excuse making. The fact that the author does not even name the study or it's authors is very telling and typical of the sort of idiocy that some dog breeders engage in when talking about 'science.' :::cue scary music.:::

  66. Well Jess Its strange how nurses still have to wear protective aprons even although they may only assist with 1 Xrayed patient a year Any of your so called google knowledge cover that one

  67. Anonymous 17 October 2011 20:08

    And your comment(s) has nothing to do with the original article, or my criticism of it. Anyone who can work a keyboard can look up the facts about x-ray safety and exposure and see what I said is true, and that Ms. Heathman's article is a misleading, scaremongering, intellectually dishonest piece of 'journalism.'

    An assertion which, for some reason, you are unable to refute, so you keep twisting about, wandering down side roads. Fancy that.

    Keep trying.

  68. Actually, now that I've considered it, Anonymous, let's turn this around. Why don't you toddle off and read the study, then explain to the rest of us *exactly* how this data is relevant to the common practice of x-raying pregnant bitches, once, at the end of term.

    Then, once you've done that, let's have a nice analysis of the article written by Ms. Heathman. It is truthful? Does it convey the study results in a manner that gives the reader enough information to weigh the risks and form their own opinion based on available evidence? Or is it slanted, with just enough information withheld, to produce fear and doubt?

  69. LOL Jess You imo could actually be describing Jemima's ways with your post Mmm so its not okay for Ms Heathman to take what she wants from a study!!! Pot Kettle Black

  70. This is the result of a study I did on Tollers.

    Country Longevity # of Dogs
    Australia 8.7 7
    Belgium 9.0 8
    Canada 11.0 208
    Denmark 10.7 3
    Finland 11.1 5
    Great Britain 9.5 9
    Netherlands 11.6 28
    Norway 11.9 2
    Sweden 8.4 73
    Switzerland 15.9 1
    United States 10.6 71
    Total 10.4 415

    While I don't claim all the fancy analysis or presentations, clearly this raises questions about one breed and the UK study. I also had 9 dogs but the longevity was 9.5 years rather than 8! If one more dog had been reported in the UK and had lived to 16, the longevity would have been 10+ years.

    Since both the total and the country by country breakdown seems to hover around 10-11 years, I'd guess that 10-11 years is a far more accurate figure.

  71. Thanks for this, Eric. I think what it proves most of all is how important it is to keep gathering the data in as sound a way as possible. Low sample sizes are clearly a problem. Note how your Finnish data based on very few dogs makes the breed look much longer-lived than the Finnish KC data based on a much larger sample which found age of death at 7 yrs 11 months. Interesting to see the Canadian and US data though - good-sized samples so shd be fairly accurate (subject to knowing more about the survey design - can u tell us?). In which case I wonder why the north american dogs are living two years longer than the finnish Tollers? Any ideas?


  72. The data comes from a database of 25,000+ Tollers, world-wide (

    This is a public database that involves self-recording of data by owners and breeders. We started with the studbooks of Canada, the US, and I think GB. As a result, we started with a fairly accurate picture. My impression is that since start-up, breeders worldwide have been conscientious about data entry though owners may be less so. That really impacts the longevity issue because I threw out any record for which there was not a death date.

    This breed really isn't a good candidate for longevity analysis because it is only recently popular. As the popularity in the US has grown, many of the dogs that are a part of that popularity are still alive! This also is true in the UK. The first 2-3 breeding programs there are just 20 years old.

  73. Eric, are you willing to let the rest of us look at your research data? It would be interesting to see how you raised an average of 7 years to 11.

    Because KoiraNet is actually contradictory to your finding:

    Nerve disease 5 years 5 months 6
    Skin and ear disorders 4 years 2 months 1
    Immunological diseases 5 years 9 months 5
    Lost 8 years 9 months 1
    Of tumor diseases, cancer 9 years 7 months 27
    Died without the diagnosis of disease 7 years 2 months 3
    End without the disease diagnosis 5 years 11 months 9
    End-behavioral or behavioral reason 4 years 0 months 10
    Bone and joint disease 5 years 8 months 2
    Liver and gastrointestinal disease 5 years 10 months 10
    Other medical conditions, which is not in the list 5 years 0 months 14
    Puppy born with a defect or malformation 0 years 2 months 2
    Spinal Disease 10 years 6 months 6
    Eye Disease 7 years 0 months 1
    Sis├Ąeriterauhasten disease 4 years 9 months 1
    Heart disease 6 years 6 months 4
    The difficulty in childbirth 2 years 8 months 1
    Injury or accident 3 years 5 months 12
    Old age (or natural finish) 13 years 4 months 30
    Urinary and reproductive tract disease 8 years 11 months 3
    Cause of death has not been reported 7 years 8 months 69
    Grand Total 7 years 9 months 217

    That's 217 dogs out of 312 registered dogs in the system. That means about 70% of all the registered FKC dogs are already deceased!

    Even if you remove "Injuries" and "End-behaviour" from the stat, it still doesn't even average out to be 11.

  74. Average lifespan data is problematic without other sorts of analysis. This is because certain forms of death can greatly alter the average.

    Anyone who has ever rated a restaurant or movie online will be familiar with this problem with averages. Let's say a recent movie "deserves" an 8. People who rate it a 10 are 2 points too generous. People who rate it a 0 are 8 points too stingy.

    And we know that certain votes are hater trolls or fanatic trolls. But one hater troll has 4 times the power of a fanatic troll.

    Well, what if we had a breed that had a particular disease, like TNS (Trapped Neutrophil Syndrome) that kills puppies, and this disease was very rampant in the breed. This would drive down the average lifespan a great deal because many dogs would be counted as 0-1s instead of 10-14s.

    This, of course, is a problematic disease... but if this disease killed very fast or not at all (carriers, clears), with no other effects on carriers, then this low average would never really apply to a puppy buyer because the affected dogs would literally be dead before anyone bought them.

    From the buyer's point of view, their dogs would live a lot longer than the "average."

    A study that doesn't track whole litters of dogs from birth to death would likely miss all the dogs that are dead as puppies. Many of these puppies would likely never be registered.

    And it's really about what you want to know. If I were an owner of young dogs, say 2 years old... I'd want to know the average life expectancy of my breed for dogs who lived at least 2 years old. That would be a lot higher than the breed average that included puppyhood diseases.

    If I were a breeder, I'd want to know about full litter studies that tracked dogs from birth to death. Same if I were a puppy buyer.

    As it is, we're mostly left with reporting errors all over the place.

  75. Dave-

    It's a public database.

    What I did was simply toss all the records for which I didn't have a death date and then compute the lifespan. I found I had 415 dogs of one breed which certainly is more representative than the 9 in the UK study and which put paid to the claims of 8 as the life expectancy.

    If people fill in the death dates of their dogs over the next couple years, I'll do it again.

    BTW, in the database we do look at lifespan of whole litters or at least the tool is there. It is just as limited by the failure of folks to enter their dog's death dates. Nonetheless, we can look at a whole litter at a time.

    I've raised Tollers for 21 years. I've had 3 dogs die in all that time. My senior dog now is from my foundation bitch and is still very much alive. She's deaf as a post (or at least pretends to be when she wants to be) from years of gunfire but otherwise is in excellent health at age 12. The 3 that we've lost are a female that died from an autoimmune disorder at age 6, a male that died at age 3 from some form of poisoning, suspect antifreeze, and the foundation bitch that died at 12 of transitional CA, a cancer of the bladder. All in all, nothing particularly notable.

    What I'll end with I meant to say in my first post, "The dogs that really prove the 8 year figure to be wrong are still alive."

  76. It's called right censoring( When a Irish Wolfhound study was controlled for right censored data, the average lifespan went from 6.47 to 7.37.

    Of course, depending on how the data is collected, we can also expect data to be left censored. The KC survey asked for information on dogs that died in the past 10 years. If you exclude from this data set dogs that were born less than 15 years ago, you would have missing info on dogs that died between 0-5 years. Whether right or left censoring has a bigger effect would also depend on whether the breed has gotten more or less popular over time, as one of the earlier commenter pointed out.

  77. I realise that I'm coming in late on this discussion, but I was interested in the results for Finnish Lapphunds. Only 5 deaths are recorded in the 2004 survey, two of which were for PRA (non-lethal, presumably the dogs were euthanised) and one for Parvo in an eight month old puppy. I think it is rather misleading and scaremongering to characterise a breed which is rare outside its home country and has a relatively young population in the UK as among "the 20 shortest-lived breeds" based on that data! The parvo pup dragged the mean age at death down considerably, and parvo is an infectious disease which has nothing to do with breed characteristics. I also noticed that in the full heath survey results for Lappies that one of the most common "conditions" recorded for Lappies was kennel cough! Hardly a systematic breed fault.

    That is not to say that there are no problems with Lappies, and the breeders and clubs that I know of are very proactive in reducing eye problems in particular. The DNA test for prcd-PRA did not become available until the year after this survey was done, and Lappie breeders are in most cases rigourously testing for the disease and planning their breeding accordingly in order to eliminate the chance of producing affected dogs. Once the (rare) PRA affected dogs born before the test was made available have died, there will never need to be another Lappie euthanised for PRA blindness unless an unethical person breeds without testing. There is also work being done on developing a test for retinal dysplasia.

    It just goes to show that reading the headline might get people agitated, but sometimes the detail reveals a different picture. It might have been good to mention in the article that this health survey included deaths from all causes, not just those which can be attributed to genetics or breed characteristics.