Bulldogs Have 72% Hip Dysplasia Rates. Greyhounds Have 3%. We Know This Because of Databases
Twenty years ago, claims about breed-specific disease risk relied heavily on clinical impression and small case series. A veterinary surgeon who had seen three Bernese Mountain Dogs with histiocytic sarcoma could say the breed seemed predisposed, but quantifying the actual risk — how much more common it was compared to other breeds — required population-level data that did not exist.
That has changed. Pet insurance databases, national veterinary hospital record systems, breed club health surveys, and projects like the Dog Aging Project have created datasets with hundreds of thousands to millions of individual animal records. These datasets provide the statistical power to identify genuine breed predispositions, quantify risk ratios, and distinguish real patterns from clinical anecdote.
The Major Data Sources
Pet Insurance Databases
Countries with high pet insurance penetration — particularly Sweden, the United Kingdom, and increasingly the United States — generate massive datasets through insurance claims. The Swedish Agria insurance database, covering over 1.2 million insured dogs, has produced some of the most influential breed health research.
Strengths:
- Very large sample sizes (hundreds of thousands of dogs)
- Longitudinal data: individual animals can be tracked across their insured lifespan
- Standardized diagnostic coding allows systematic analysis
- Mortality data captures both age at death and cause of death
Limitations:
- Insurance populations are not random samples of the general dog population. Insured dogs tend to receive more veterinary care, which means more diagnoses. Breeds with known health problems may be overrepresented (owners insure them because they anticipate costs) or underrepresented (insurance companies exclude high-risk breeds or charge prohibitive premiums).
- Cause-of-death data depends on owner-reported information and may be imprecise for dogs that die at home
- Diagnostic accuracy depends on the attending veterinarian and may not always reflect gold-standard diagnostic workup
Veterinary Hospital Records
University veterinary hospitals and large referral centers maintain diagnostic databases that capture detailed clinical information. Studies using these records can identify breed predispositions for specific conditions with precise diagnostic criteria.
A 2018 BMC Veterinary Research study used the VetCompass database — primary-care veterinary records from over 900 UK practices covering hundreds of thousands of dogs — to quantify breed-specific disease prevalence for dozens of conditions.
Strengths:
- Diagnostic data from veterinary professionals
- Large sample sizes within primary care populations
- Captures the full spectrum of conditions (not just the severe ones that reach referral centers)
Limitations:
- Referral center data overrepresents severe cases. Primary care data captures a broader population but may lack diagnostic precision.
- Breed identification in veterinary records is often owner-reported and may be inaccurate, especially for mixed breeds
Breed Club Health Surveys
Many breed clubs conduct health surveys of their membership, asking owners to report conditions diagnosed in their dogs. The Kennel Club (UK), AKC (US), and breed-specific clubs run periodic surveys.
Strengths:
- Breed-specific focus allows deep analysis within a single breed
- Capture conditions that may not prompt veterinary visits (behavioral problems, mild conditions)
- Engage large numbers of breed enthusiasts who provide detailed histories
Limitations:
- Self-reported data is subject to recall bias, response bias (owners of sick dogs may be more motivated to respond), and diagnostic imprecision
- No veterinary verification of reported diagnoses
- Club membership is not representative of all owners of that breed
Key Findings from Large Database Studies
Lifespan Variation by Breed
The 2016 Veterinary Journal study of over 31,000 deceased dogs in England found:
- Median lifespan ranged from 5.4 years (Dogue de Bordeaux) to 14.2 years (Miniature Poodle)
- Body size was the strongest predictor of lifespan, with a clear inverse relationship between adult body weight and median longevity
- Within size categories, there was still significant breed-to-breed variation, suggesting breed-specific genetic factors beyond size
Cancer as a Breed-Specific Problem
The Swedish insurance database studies (2010) revealed dramatic breed differences in cancer mortality:
- Bernese Mountain Dogs: cancer accounted for approximately 55% of deaths, with histiocytic sarcoma as the predominant type
- Golden Retrievers: cancer accounted for approximately 60% of deaths, with hemangiosarcoma and lymphoma dominating
- Boxers: elevated rates of mast cell tumors and lymphoma
- Small breeds (Chihuahuas, Toy Poodles): cancer was a less common cause of death, with cardiovascular and respiratory disease more prominent
These findings have been replicated across multiple databases and countries, confirming that cancer risk is not uniformly distributed across breeds but is shaped by breed-specific genetics.
Orthopedic Disease
OFA (Orthopedic Foundation for Animals) data, compiled from voluntary radiographic submissions, provides breed-specific prevalence rates for hip dysplasia and elbow dysplasia:
- Bulldog: 71.8% hip dysplasia rate (highest of any breed)
- Pug: 63.8%
- German Shepherd: 20.5%
- Labrador Retriever: 12.3%
- Greyhound: 3.4% (one of the lowest)
OFA data has significant selection bias — owners who suspect their dog has dysplasia may be more or less likely to submit radiographs, depending on their breeding plans. Nevertheless, the breed-specific ranking has been consistent across decades and across independent databases.
How to Use Breed Predisposition Data
For Dog Owners
Knowing your breed’s predispositions allows targeted screening and prevention:
- Golden Retriever owners should discuss cancer screening protocols with their veterinarian, including regular hemangiosarcoma screening via abdominal ultrasound
- Cavalier King Charles Spaniel owners should begin cardiac screening (auscultation, echocardiogram) early, given the breed’s extreme predisposition to mitral valve disease
- Large breed owners should follow breed-specific joint screening protocols
For Breeders
Breed predisposition data should inform breeding decisions:
- Health testing of breeding stock for conditions with documented breed predisposition
- Calculating inbreeding coefficients and selecting for genetic diversity
- Tracking health outcomes in offspring to identify carriers of recessive disease alleles
For Veterinarians
Breed predisposition data supports clinical decision-making:
- Adjusting differential diagnosis lists based on breed
- Recommending breed-specific screening schedules
- Communicating risk information to owners in a way that enables informed decision-making
Limitations Worth Acknowledging
Database studies reveal associations, not causes. A breed with a high rate of hip dysplasia may have that predisposition because of specific gene variants, because of popular sire effects that reduced genetic diversity, because of management practices common among that breed’s owners (such as early spaying/neutering — see the spay/neuter timing article), or because of a combination of all three.
Additionally, breed predisposition data describes population-level risk. An individual Golden Retriever may never develop cancer. An individual Chihuahua may develop it. Breed data informs screening and prevention strategies but does not determine individual outcomes.
For breed-specific longevity data, see the breed longevity guides and the canine longevity genetics research.
Frequently Asked Questions
How reliable are breed predisposition databases for predicting my dog’s disease risk?
Large database studies like the OFA registry and breed-specific health surveys provide population-level risk estimates based on tens of thousands of cases. While they cannot predict whether an individual dog will develop a specific condition, they identify which conditions to screen for proactively. A breed with a 50% hip dysplasia rate warrants different screening than one with a 3% rate.
Do mixed-breed dogs have fewer health problems than purebred dogs?
On average, mixed-breed dogs show lower rates of conditions linked to breed-specific genetic bottlenecks, such as certain cancers and structural disorders. However, they are not immune to genetic disease, and their risk profile is harder to predict because their genetic background is often unknown. Breed predisposition data is most useful when a dog’s ancestry is known.
How are breed predisposition databases built?
Major databases aggregate data from veterinary teaching hospitals, insurance claims, breed health surveys, and registries like the OFA. Each source has limitations — referral hospital data overrepresents severe cases, insurance data reflects covered populations only, and breed surveys depend on owner reporting accuracy. The most reliable conclusions come from findings consistent across multiple data sources.
Should I choose a breed based on health predisposition data?
Health predisposition data should be one factor in breed selection, alongside temperament, exercise needs, and lifestyle compatibility. Within any breed, selecting a breeder who performs recommended health screenings (OFA, PennHIP, genetic panels) and breeds for longevity reduces individual risk substantially compared to breed averages.
Bottom Line
Large-scale breed health databases have replaced clinical anecdote with population-level data on disease risk, revealing dramatic breed-to-breed differences in cancer rates, joint disease prevalence, and lifespan. These data are powerful for informing screening and prevention strategies but describe population-level probabilities, not individual certainties. Knowing your breed’s specific predispositions allows you to screen earlier and more effectively for the conditions most likely to matter.
References
- Breed-specific disease prevalence in companion animals (BMC Veterinary Research, 2018).
- Longevity and mortality of owned dogs in England (Veterinary Journal, 2016).
- Mortality in insured Swedish dogs: breed-specific causes of death (Preventive Veterinary Medicine, 2010).
- Inherited defects in pedigree dogs: disorders that matter (Veterinary Journal, 2010).