Health data on your own offspring are the only evidence that tells you what your breeding actually produced — and they are also the easiest kind of evidence to collect badly.
The most important first
- Decide the fields before you start asking; data gathered inconsistently across litters cannot be compared later, however diligently they were collected.
- Age at onset matters as much as diagnosis, because a condition appearing at two means something different from the same condition at ten.
- Age at death and cause of death are the single most valuable items you will ever record, and the hardest to ask for.
- Your responders are not a random sample of your offspring, so treat every pattern as a question rather than a finding. One breeder’s litters cannot support a rate, a prevalence or a heritability.
- Build the collection on top of your existing records and your contact routine rather than as a separate project.
Decide the fields before you collect anything
The failure mode in breeder health data is not laziness but inconsistency. A breeder who asks open questions and writes down whatever comes back will, after ten years, have rich notes on four dogs and nothing comparable across forty. The fix is unglamorous: fix a short list of fields, use exactly those for every offspring in every litter, and resist improving the list mid-stream. If you must change it, note the date.
A workable core is smaller than most breeders expect. For each offspring: identity and microchip number, date of birth, sex, neuter status and the date it happened, the home’s current contact details, and the date of your last contact. Then, per event, a dated line with what was observed, who observed it, and what the outcome was. For screening, the scheme used, the date, the age at examination and the result exactly as issued. Finally, at the end: date of death, age at death, cause as given by the veterinary surgeon, and whether the dog was euthanized.
Record results in the form the issuing body uses, not your own translation. A hip score, a grade, an eye certificate and a DNA test result each mean something specific within the scheme that produced them, and a note saying “hips fine” is worth almost nothing in five years. Which schemes apply to your breed, and what their results mean, is covered in health screening schemes and interpreting database results.
Age at onset, not just diagnosis
The most common omission in breeder records is the date. A list of diagnoses across a litter tells you what occurred; the same list with ages attached tells you something about severity, about selection pressure and sometimes about whether two cases are even the same thing. Atopic skin disease starting at eleven months and starting at eight years are not interchangeable observations, and a cruciate rupture in a working dog of three is a different data point from one in a sedentary dog of nine.
Ask for the age at which a problem was first noticed, as distinct from the age at which it was diagnosed. Owners often remember the first sign more clearly than the appointment, and the gap between the two is itself informative about how the condition presented. Where a diagnosis is uncertain, record the uncertainty rather than resolving it for the owner: “investigated for, not confirmed” is honest and useful, while a tidy label you invented is neither.
Screening results need the dog’s age too. Schemes have defined minimum ages and reading windows; estimated breeding values for hip and elbow dysplasia published by the Royal Kennel Club, for example, are calculated from scores on dogs examined between one and four years of age. A result read outside the usual window may still be informative, but is not directly comparable with the rest of the breed’s data.
At your later checkpoints, ask specifically about signs that often come before a diagnosis in older dogs: new lumps, weight loss or gain, stiffness or difficulty rising, confusion, changes in sleep, or house soiling. Encourage owners to raise these with their own veterinarian promptly — veterinary bodies recommend that senior dogs are checked at least twice a year, and large and giant breeds reach senior age sooner.
Asking about death without intruding
This is the part breeders avoid, and it is the part that matters most. Age at death and cause of death are what turn a collection of anecdotes into something that can be compared with population figures. Life-table analysis of primary-care veterinary records in the United Kingdom put life expectancy at birth at around 11.2 years overall, with breed group estimates ranging from roughly 9.1 years in the Working group to 12.0 in terriers, and individual breed figures spread far wider than that. Without ages at death from your own dogs, none of that context can be applied to your line.
The way to ask is slowly, once, and with an exit. A short message some weeks after a death, saying that you are sorry, that you do not need details now, and that if the owner is ever willing you would like to record the dog’s age and what happened, gets a better response than a form. Many owners want to tell someone who knew the dog as a puppy. A few do not, and will say so or say nothing; that answer is final.
When an owner does share, record what the veterinary surgeon said rather than your interpretation, and record “unknown” where it genuinely is. Note whether the dog was euthanized and, if the owner volunteers it, why. Euthanasia complicates age-at-death figures, as the published literature acknowledges: dogs are sometimes euthanized for reasons that are not strictly life-limiting, and a dataset that does not distinguish euthanasia from unassisted death will mislead you about both.
In the 2004 UK purebred survey, the most commonly recorded causes of death were cancer (about a quarter), “old age” and heart disease, with a median age at death of a little over 11 years. These make sensible starting categories for your own cause-of-death field, with “other” and “unknown” alongside. Life expectancy also differs widely between breeds — from under five years to nearly thirteen in UK veterinary records — so compare your offspring with your own breed’s figure rather than with dogs in general.
In practice
A breeder with five litters placed over eleven years keeps one row per offspring in a single spreadsheet: thirty-four rows, nineteen columns, no free text except a dated notes field. By year eleven, twenty-six homes are still contactable, four have gone quiet, and four dogs have died. Reading down the columns, the breeder notices that three of the four deaths were in offspring of one dam, at six, seven and eight years, from three unrelated causes. The pattern looks alarming on the page. It is also four deaths in total, from one bitch’s three litters, with no information from the four silent homes — which is not enough to conclude anything. The breeder records the observation, flags the dam’s line for closer attention in the next round of contact, and asks the breed club whether anything similar has been reported elsewhere.
(Illustrative, composite scenario.)
Standardize so that litters can be compared
Comparability is the whole point. If the 2018 litter was asked different questions, at different ages, through a different channel, than the 2023 litter, any difference you observe between them may be a difference in how you asked. Keep the question set stable, send it at the same ages in each litter, and record the date you asked as well as the date you were answered.
Two conventions make a disproportionate difference. First, record non-answers explicitly: a blank cell is ambiguous, while “asked 2026-03-11, no reply” is data. Second, record who reported what. An owner’s report, a veterinary record the owner forwarded, and an official scheme result are three different grades of evidence, and in five years you will not remember which was which unless you noted it.
Keep the structure compatible with your kennel records so that a dog’s puppy notes, its screening results and its adult history can be read in one line. The practical test is whether somebody else could read your file and reconstruct what happened to each offspring.
How these data will mislead you
Three biases are built into breeder-collected health data, and none of them can be removed. The first is non-response. The homes that answer are not a random sample: owners dealing with a serious diagnosis, an early death or a behavior problem they feel responsible for are less likely to reply, which pushes your data toward looking healthier than reality. Even large organized surveys struggle here. The 2004 United Kingdom purebred dog health survey, distributed through breed clubs, returned around 24 percent of questionnaires overall, with breed-level response rates ranging from under 5 percent to about 65 percent, and its authors noted that the breeds in the study may not represent the wider pedigree population.
The second is ascertainment. Owners who are in frequent contact with you, who show or work their dogs, or who are simply more attentive, generate more recorded diagnoses than owners who are not. More data from a home does not mean more disease in that dog. The third is scale. A litter is typically five to nine dogs; forty offspring across six litters is still forty. With numbers like that, chance alone will produce clusters that look like patterns, and a single unusual litter can dominate everything you think you know.
The honest consequence is a restriction on the language you use. Your material can support statements such as “three of eleven offspring from this combination were treated for” and “this condition was associated with offspring from this sire in my records”. It cannot support “this sire produces” or “the incidence in my line is”. Keeping to the first kind of sentence is not false modesty; it is the difference between data and belief.
Where your records meet the breed’s
Individual records become powerful when read against aggregated data, and when contributed to it. Breed-level work is built from exactly this material: the Royal Kennel Club’s breed health and conservation plans were set up to identify each breed’s health concerns through evidence-based criteria drawn from veterinary and scientific literature and health research, working with breed clubs and breed health coordinators, and to build monitoring and review into each plan as evidence changes. A published plan shows what this looks like in practice: the English Toy Terrier plan, for example, brings together breed health surveys, a literature review, DNA test results, health scheme results, reported cesareans and measures of genetic diversity, and ends with the breed’s priorities and an action plan.
Public registries matter for the same reason. Under the Orthopedic Foundation for Animals’ Canine Health Information Center program, a dog qualifies for a CHIC number only when it has completed the screening its parent club recommends, is permanently identified by microchip or tattoo, and has all of those results, normal or abnormal, released publicly; the number reflects completed and published testing rather than normal results. That release requirement is what makes offspring data interpretable to everybody else, and a breeder who also shares results for dogs outside the program extends the same benefit. More on this in interpreting health database results and, for population-level questions about longevity, the evidence on inbreeding and diversity.
When to call your vet
- An owner reports a new lump, a lump that is growing, or a sore that does not heal.
- An offspring loses weight without a diet change, eats less, or drinks noticeably more.
- A dog becomes stiff, limps, or struggles with stairs or rising.
- An older dog seems confused, sleeps differently, or starts soiling indoors.
- A screening result is borderline or abnormal — the owner’s vet or the scheme vet can explain what it means for that dog.
- If in doubt, call — your vet would rather hear from you early.
Work on it
A few minutes here will teach you more than re-reading the article.
Ask yourself
- Could you state, today, the age at death of every dog you have bred that has died?
- Do your records say who reported each piece of information, and on what date?
- For the last pattern you thought you saw in your own offspring, how many dogs was it actually based on?
Find out & check
- Find out which screening schemes apply to your breed in your country, what the minimum examination ages are, and whether results are published openly or only on request.
Do one thing
- Write your field list on one page, then enter your oldest litter into it and see which columns you cannot fill.
Was this article helpful?
Anonymous. We do not store your name, email or IP address.
Sources & further reading
- Teng, Brodbelt, Pegram, Church & O’Neill — Life tables of annual life expectancy and mortality for companion dogs in the United Kingdom. nature.com
- Adams, Evans, Sampson & Wood — Methods and mortality results of a health survey of purebred dogs in the UK. pubmed.ncbi.nlm.nih.gov
- Orthopedic Foundation for Animals — CHIC programs. ofa.org
- The Royal Kennel Club — Kennel Club launching Breed Health and Conservation Plans. royalkennelclub.com
- Wiles, Llewellyn-Zaidi, Evans, O’Neill & Lewis (2017), Canine Genetics and Epidemiology 4:8 — Large-scale survey to estimate the prevalence of disorders for 192 Kennel Club registered breeds. doi.org
- The Kennel Club — English Toy Terrier Breed Health and Conservation Plan, Version 1 (January 2021). english-toy-terrier-club.com
- Royal Kennel Club — Estimated breeding values. royalkennelclub.com
- American Veterinary Medical Association — Senior pet care (FAQ). avma.org
Where evidence is limited, recommendations rest on established practice and expert judgment, and may be updated as knowledge develops. There is no established methodology for health data collection at the scale of a single kennel, so the field list and conventions here are practical convention rather than validated instruments; the population figures quoted describe national datasets with their own stated limitations and cannot be used to judge an individual line.
Next steps
- Read next: Following up temperament and behavior.
- Related: Health databases and interpreting results · Fertility in the offspring: what the evidence shows.
- Return to the area hub: Closing the loop: follow-up and the next generation.