Do Working-Line German Shepherds Have Better Hips Than Show Lines? We Went Looking for the Study

We looked for the study and could not find it. On 4 November 2026 we searched PubMed, scanned 144 full-text papers and ran a German-language search, and we found no paper reporting hip scores or dysplasia rates by line. The closest we found was one abstract on police and companion dogs, not lines, reporting a similar incidence in both. OFA's own statistics table and the peer-reviewed analysis of its database both report the German Shepherd as one breed. Working lines may have better hips. We could not find it shown, and the percentages going round are unsourced or about something else. So choose the line for the dog you want to live with, and choose the litter on paper: both parents' results, what happened to the siblings, and how many dogs that kennel actually x-rays.

Two tabs at 11pm#

One kennel breeds for work. The other breeds for the show ring. You like both, and you have read that this choice decides his hips. Forums say working lines are sounder. A few sites rank the lines from best to worst. Someone quotes a percentage. It feels like the one thing about this dog's future you can settle tonight.

I understand the pull of that. If the line settled it, you could stop worrying before he came home.

On what we could find published, it doesn't. The line tells you what his ancestors were asked to do. Nobody has shown us that it tells you what is in his hip sockets.

We looked for the study#

This is what we did, so you can repeat the database searches and prove us wrong.

This was a keyword scan, not a formal systematic review, and we did not read all 144 papers. A sentence that never uses a line term, or a table split by line, could slip past the filter. Our terms did not include "performance line", "breeding line" or "Schutzhund", and the German search was an ordinary web search with no fixed string. It cannot see inside paywalled papers beyond the abstract, or inside a club's own files. One specific gap: a 2009 Hannover thesis on body traits and hip dysplasia in German Shepherds sat behind a security page we could not open. The abstract of the 2012 paper from the same group, using 14,416 male and 21,612 female dogs, does not mention lines.^[15] A conference abstract, a book or that thesis could hold something we missed. If you know of one, send it, and we will print the correction.

We are not the first to come up empty. Southern Wind K9, a breeder's blog, went looking for the same thing and wrote: "I could not locate a strong, representative, peer-reviewed study establishing those percentages as a general working-line-versus-show-line comparison."^[2]

Where do the numbers going round come from?#

Every percentage we could find comparing the two lines was either unsourced or about something else. That does not make the belief false. It means that, as far as we can find, the belief is running on repetition.

Can OFA answer this?#

People search for an OFA survey of working and show lines. We could not find one, and it is not in OFA's published numbers. On 4 November 2026 we rendered OFA's live statistics page in a browser and read all of it. Its hips table has a single German Shepherd Dog row, one result for the whole breed. The word "working" and the term "show line" appear nowhere on the page.^[22] The peer-reviewed analysis of OFA records, covering 1970 to April 2015, is the same: one figure per breed, 18.9% dysplastic for 107,048 German Shepherds, with nothing separating working dogs from show dogs.^[3] OFA may hold more than it publishes. But if a page credits a working-line or show-line percentage to OFA, ask where it came from, because it did not come from this table.

A split would also carry a second problem. OFA submission is voluntary. In a prospective study, only 53% of films taken for OFA certification were sent in, and a film with normal-looking hips was 8.2 times as likely to be sent as one without.^[4] That study was in Golden Retrievers and Rottweilers, not Shepherds. But the lesson carries over: if show kennels and sport kennels submit at different rates, a gap between their numbers could be a gap in paperwork rather than hips. That last step is our reasoning. We found no measurement of it.

dogorthopedic.com

Large datasets exist. We found no published split.#

Some schemes are less exposed to selective reporting than a voluntary registry. Sweden, for one, records the result of every official screening, though dogs with early clinical signs may never be screened.^[5] The maddening part is that files like these could probably answer the question, if someone coded every pedigree by line and published the rates.

Do not rank those rows against each other. They use different grades, ages and rules, and our German Shepherd numbers page explains why national schemes disagree. Even Germany's scheme, where x-rays are mandatory for dogs entering the club's breeding programme,^[19] is not clean. The club's own 2014 article says films with poor results are often not sent in, and that this pre-selection is probably carried out as consistently in this breed as in hardly any other. That is why the official figures and the 25% estimate sit so far apart.^[8]

What do the studies with both lines actually show?#

Three kinds of research have had both lines in the room. None of them answers the hip question, and each is easy to misquote.

The hip genetics study. A Finnish team recruited 531 Shepherds, 247 cases and 284 controls, picked "according to their hip scores and pedigrees, creating a balanced study population of working, mixed and show line dogs."^[6] The lines form separate genetic clusters, with mixed-line dogs between them. So hip score and line have sat in one dataset. But the sample was hand-built, so it can give no rate for either line, and the paper's text reports no comparison of hips between them.

The 120-year genome study. A 2025 paper compared museum specimens with modern dogs and found inbreeding far higher in contemporary Shepherds. It compared the timing of population bottlenecks between working and American conformation dogs. It did not compare hips.^[9]

The body and gait studies. The two types look different, but the measured evidence is thin. A book study of 10 working and 14 show Shepherds reported more forward and backward limb swing and a shorter femur in the show dogs, and later authors note that it did not report statistical significance. A study that recruited 60 Shepherds, grouped by the slope of the back rather than by line, reported that slope did not significantly affect limb-segment lengths or protraction and retraction angles, and said this disagreed with the book. Treat that as a weak null. Full 3D movement data could not be captured for all 60, the paper gives no separate count for that result, and it counted a measure as affected only if it passed both a correlation test and a comparison between groups. Slope did affect posture and movement: standing and trot forces across the 60, greater movement asymmetry in the 30 dogs with the best data, and an irregular motion pattern on video in more sloped dogs. Every dog in that study had been cleared of hip dysplasia, so it could not test the disease at all, and the authors say sloped dogs were harder to recruit. It left any link to disease for further research, though its discussion does suggest that lower force through the hind legs may be related to the development of hip dysplasia. That is speculation from other studies, not a result, but it is the one place the paper leans towards the show-line worry.^[10] We cover that study on the main German Shepherd hip page. A different shape is a reasonable thing to suspect. It is not a measured difference in disease.

"Dogs with bad hips can't work." Can't they?#

This is the best argument for working lines, so it deserves a look. If generations of dogs had to do hard physical work to breed, surely bad hips were weeded out.

The limit comes first. The studies below follow service dogs. None follows breeding stock, and none looks at the sport titles working-line breeders point to, a term our search did not cover. So they can only show how dysplasia and work sit together in a dog, not what a line has been selected for.

Five studies bear on it, and none reports line. They do not all point the same way. In the first, forty male police dogs, 35 German Shepherds and 5 Rottweilers, were x-rayed when they were bought and again 60 months later. At purchase, 37 had no dysplasia and 3 had mild dysplasia. Five years on, 17 of the 40 had dysplasia, all of them Shepherds. Secondary joint changes rose from 3 dogs to 14. Yet only one dog showed clinical signs, and no dog was lost from service because of hip dysplasia.^[11]

We read the abstract, not the full paper. It is small and retrospective and it covers one service. All 17 dysplastic dogs were Shepherds, so for this breed the figure is 17 of 35, about half. Those films came after five years of police work and the abstract gives no ages, so this is not a rate at screening age, and it should not be set beside the registry figures above. Three more limits matter. Thirty-seven of the 40 started clear, which may reflect selection at purchase, though the abstract does not say. The second films were taken after 60 months of service, so a dog who left early may never have been counted. And the abstract gives no breakdown of final grades, so we cannot say how severe the 17 were, though it does say that "the increase in degree of hip dysplasia was generally more than one degree".^[11] What it does show is narrow: among dogs still in service at 60 months, dysplasia on the film had not ended the job.

The second is larger, and it found no difference either. In 116 military working dogs, total months worked did not differ significantly between normal and dysplastic dogs.^[24] The abstract, all we could read, names no breed and gives no group sizes, and failing to find a difference does not prove there was none. The first part of the same study shows the joints did change on film: dysplastic dogs had an odds ratio of 70.2 for developing degenerative joint disease compared with normal dogs. That is a radiographic result, not a measure of pain or lost work.^[25]

The third is the most direct, and it points the other way. A handler survey covering 182 New Zealand police German Shepherds asked why dogs left duty. Mean age at loss was 6.6 years, and only 40% reached the planned retirement age of 8. The largest cause of retirement was not coping with the physical demands of the job, and degenerative musculoskeletal disease was the primary factor in 42 of those 61 dogs. Leaving out planned retirement and accidents, spinal disorders accounted for 34% of losses and hip arthritis for 23%.^[26] It rests on handlers' memories rather than veterinary records, and it does not report line. But it is a plain answer to anyone who says hips do not stop working Shepherds.

The fourth is the closest thing to a comparison of groups we found. A study of pelvic x-rays in police working and companion German Shepherds reported a "similar incidence of hip dysplasia in both the groups."^[18] It was not about lines: a police dog and a working-line dog are not the same thing, and a companion dog is not a show-line dog. The abstract, all we could read, gives no percentages for hips. The same abstract reports transitional vertebrae in 12% of working and 33% of companion dogs, more severe dysplasia in dogs with those vertebrae, and more sacroiliac degeneration in working dogs.^[18] "Similar incidence" says nothing about severity, and with no hip numbers to check, read it as a hint and no more.

The fifth also leans that way. Fifty active police dogs in Portugal, 17 of them German Shepherds, were seen for hip osteoarthritis at a mean age of 6.5 years, and their trainers' complaints included difficulty rising and jumping. All had been screened before training and work, and the authors suggest that the demands of the job may explain the earlier diagnosis: "it is possible that the disease actually develops or is simply detected earlier than in other dogs." They also note that these dogs' high drive may mask some complaints.^[12] The paper has no comparison group of its own for "earlier", it does not report what the early screens showed, and it gives no rate, because the dogs were selected for having the disease. All 50 were active dogs, so it shows diagnosis, not dogs removed from work or breeding.

So the record is mixed: some dysplastic dogs worked on, and in New Zealand hip arthritis ended careers early. None of it shows a dog dropped from breeding because work exposed his hips. Our reasoning from there: selection for work may help a line, but we found nothing that measures it. Within the German club there is also a shared gate: x-rays are mandatory for dogs entering the breeding programme,^[19] and the club material we read draws no distinction by line. That does not reach American show lines.

None of this proves working lines are no better. Selection for work may still help. What it removes is the certainty, in both directions.

“Because your dog can't tell you how much pain he or she is in.”

The danger is not only in the buying#

Here is why I think this matters more than one puppy purchase. A label that answers the question also ends the looking.

"He's working-line, so it won't be his hips." Then he takes longer to get up. He stops at the bottom of the stairs. Those are the quiet signs that get read as something else, and a confident story about his pedigree makes them easier to read that way. The same trap has a show-line version: "All show dogs have bad backs, what did I expect?" Both sentences stop the same enquiry, and it takes a vet's hands and a film to restart it. If you are watching a change right now, our guide to limp versus weakness is the place to begin.

Working line is not a free pass in the other direction either. Our search was built for hips, not for this, but one disease described mainly in working-line dogs surfaced by name, and it is not in the hip at all. Fibrotic myopathy of the hamstring muscles is uncommon, but the majority of reported cases are young male working-line German Shepherds.^[1] The 10 affected dogs measured in that paper averaged 6.7 years old, and two were sport dogs while the rest were neither working nor sporting dogs. Their hamstring stretch angle averaged 102° in the worse leg, against 147° in unaffected Shepherds, though most affected dogs were measured under sedation, which the authors expected would overestimate the angle if anything.^[1] That counts reported cases, not risk, but if a young working-line dog develops a stiff, odd hind-leg gait, it is worth asking your vet about by name.

What should you look at instead?#

Family records. Nothing here predicts one puppy with certainty, and a buyer reading certificates is not running a breeding programme. But selection on records has been shown to work at the level of a colony, and we found no test of selection by label. The Seeing Eye's guide dog colony took hip dysplasia in its German Shepherds from 55% to 24% in under five generations by choosing parents on estimated breeding values, which combine a dog's own record with its relatives'.^[13] That was neither a show kennel nor a sport kennel. It was a closed colony using numbers, and 24% is still a lot of dogs. In the club's magazine, the article on hips says the club's breeding values, in use since 1999, say more about a dog's genetic disposition than the parents' own results, because they include relatives and correct for environment, while warning that with mostly favourable data, genetic differences are hard to detect with the reliability needed, and admitting the improvement was less clear than hoped.^[8]

The litter matters too. In a Norwegian cohort of four other large breeds, dysplasia clustered significantly within litters.^[14] Littermates share a rearing as well as genes, so that is not proof of inheritance. Either way, two excellent parents are a good start and not the whole family.

So this is what I would ask, in either kennel:

None of this guarantees his hips. Nothing does. It replaces a label with records, which is about the most a buyer can do. If you want an early reading on the puppy himself, PennHIP against OFA covers the timing.

Breeders describe the lines as different in drive and energy. We did not review that evidence, but it is a fair thing to ask each kennel about and to meet in person. Choose the dog you can live with. Just do not let the word on the pedigree stand in for the x-ray.

dogorthopedic.com

Once he is home#

For the choice of kennel, buy nothing. Spend the effort on the paperwork.

After that, the question I would turn to is feeding, though our sources disagree on how much it counts. In a study of 48 Labrador puppies, one group of 24 ate 25% less than the other from eight weeks of age. At two years, 7 of the 24 limit-fed dogs were dysplastic by OFA score against 16 of the 24 fed freely, and 5 against 18 by the Swedish score.^[20] Those were Labradors, the work came from Ralston Purina's research department, and it compared free feeding against 25% less, not one way of measuring against another. Dr. Tellhelm says in the club's article that the influence of feeding is often overrated and that genetic disposition decides most cases, while adding that faulty feeding, and too much or too little load, can worsen the disease in a dog already predisposed.^[8] Those two do not sit easily together: a fall from 16 dogs to 7 is not what "overrated" suggests, and we cannot settle it, because neither source is a feeding trial in Shepherds. Seven of 24 also means feeding lowered dysplasia and did not prevent it. I would keep a Shepherd puppy lean anyway, because it costs little and the trial we read points that way.

The free version comes first, and for most homes it is enough. Learn to feel his ribs, and ask your vet to score his condition at each puppy visit.

Nothing here needs buying. If you feed by the cup and want one aid, a gram scale makes a portion repeatable, though no study shows that weighing food lowers dysplasia. Any kitchen scale with 1 g steps will do, including the one you may already own. Our German Shepherd numbers page has the measurement of how far a cup strays, if you want the case for it.

One free precaution for the first weeks: stairs. In the Norwegian cohort, puppies walking on stairs up to 3 months had a higher risk of radiographic dysplasia, and off-lead exercise in the same weeks was linked to a lower risk.^[14] Our sources disagree on how much this matters. Asked about puppies in the club's article, Dr. Tellhelm calls measured, controlled stair use acceptable and says genetics has more influence than normal housing.^[8] Neither source is a trial in Shepherds. A closed door or carrying him costs nothing, so I would do that and not worry further.

What to do this week#

If you are choosing a puppy: send both kennels the same five questions from the table above, in the same words, and compare the answers.

If you already have the dog and your confidence rests on his line: take that off the table. Ask your vet what a hip film would cost and what it would change. If anything has quietly shifted, the way he rises, the stairs, the car, book the look now.

We cannot give you the percentage for his line. You can still look at your own dog: a vet's exam and a film tell you about his hips, which a label cannot, though a film does not show what he feels. A label will not shorten the delay. Looking might.

Sources#

1. Birdwhistell KE, Miscioscia E, Montalbano C, et al. "The hamstring stretch angle: a screening and monitoring tool for canine fibrotic myopathy." Frontiers in Veterinary Science. 2025;12:1600602. doi:10.3389/fvets.2025.1600602 — full text read 4 November 2026. Ten affected Shepherds, 10 unaffected Shepherds and 10 unaffected retrievers; affected measurements collected retrospectively.

2. Southern Wind K9. "German Shepherd Back Debate: What Science Really Says About Slope, Structure, and Hip Dysplasia." southernwindk9.com — a breeder's blog, cited for what it looked for and did not find. Read 4 November 2026.

3. Oberbauer AM, Keller GG, Famula TR. "Long-term genetic selection reduced prevalence of hip and elbow dysplasia in 60 dog breeds." PLoS ONE. 2017;12(2):e0172918. doi:10.1371/journal.pone.0172918 — full text searched for line terms 4 November 2026.

4. Paster ER, LaFond E, Biery DN, et al. "Estimates of prevalence of hip dysplasia in Golden Retrievers and Rottweilers and the influence of bias on published prevalence figures." JAVMA. 2005;226(3):387-392. doi:10.2460/javma.2005.226.387 — abstract read 4 November 2026.

5. Engdahl K, Malm Persson S, Andersson L, Ringmark S, Bjørnvad CR, Hedhammar Å, Höglund K. "Prevalence of hip and elbow dysplasia in young adult dogs in Sweden." Veterinary Record. 2026;198(12):e560-e573. doi:10.1002/vetr.70224 — full text read 4 November 2026. German Shepherd hip row: 3,154 of 12,335 graded C or worse; elbow row: 2,077 of 12,309.

6. Mikkola LI, Holopainen S, Lappalainen AK, et al. "Novel protective and risk loci in hip dysplasia in German Shepherds." PLoS Genetics. 2019;15(7):e1008197. doi:10.1371/journal.pgen.1008197 — full text read 4 November 2026. Cases were FCI C or worse in both hips, controls A in both hips.

7. Stock KF, Klein S, Tellhelm B, Distl O. "Genetic analyses of elbow and hip dysplasia in the German shepherd dog." Journal of Animal Breeding and Genetics. 2011;128(3):219-229. doi:10.1111/j.1439-0388.2010.00901.x — abstract only, read 4 November 2026.

8. SV-Aktuell. "Gesundheit des Deutschen Schäferhundes: Vorurteile und Fakten." SV-Zeitung, October 2014. schaeferhunde.de — the breed club's own magazine, reporting Dr. Bernd Tellhelm's work at Giessen. German; our translation. Figures are twelve years old. The 70%, 15% and "nearly 10%" describe the club's reading centre; the 8% is attributed to the GRSK official statistics. They appear to be different statistics. Read 4 November 2026.

9. Scarsbrook L, Spatola GJ, Dreger DL, et al. "A 120-y time series of genomes reveals the consequences of closed breeding in German Shepherd Dogs." Proceedings of the National Academy of Sciences. 2025;122(48):e2421755122. doi:10.1073/pnas.2421755122 — full text read 4 November 2026.

10. Humphries A, Shaheen AF, Gómez Álvarez CB. "Different conformations of the German shepherd dog breed affect its posture and movement." Scientific Reports. 2020;10(1):16924. doi:10.1038/s41598-020-73550-x — full text read 4 November 2026. The 10-and-14-dog comparison is Fischer MS, Lilje KE, Dogs in Motion, VDH Service GmbH, 2014, which we know only through this paper and a companion paper; we have not read the book.

11. Zorko B, Ivanusa T, Pelc R. "Progression of hip dysplasia in 40 police working dogs: a retrospective study." Slovenian Veterinary Research. 2005;42:71-76. OSTI record — abstract only; the full PDF was not reachable. Read 4 November 2026.

12. Alves JC, Santos A, Jorge P, Lavrador C, Carreira LM. "Clinical and diagnostic imaging findings in police working dogs referred for hip osteoarthritis." BMC Veterinary Research. 2020;16(1):425. doi:10.1186/s12917-020-02647-2 — full text read 4 November 2026.

13. Leighton EA. "Genetics of canine hip dysplasia." JAVMA. 1997;210(10):1474-1479. PubMed 9154200 — abstract read 4 November 2026.

14. Krontveit RI, Nødtvedt A, Sævik BK, Ropstad E, Trangerud C. "Housing- and exercise-related risk factors associated with the development of hip dysplasia as determined by radiographic evaluation in a prospective cohort of Newfoundlands, Labrador Retrievers, Leonbergers, and Irish Wolfhounds in Norway." American Journal of Veterinary Research. 2012;73(6):838-846. doi:10.2460/ajvr.73.6.838 — abstract read 4 November 2026.

15. Stock KF, Dammann M, Distl O. "Selection for conformation and conformational homogeneity of litters in the German shepherd dog." Journal of Animal Science. 2012;90(4):1088-1096. doi:10.2527/jas.2011-3855 — abstract read 4 November 2026. The related Dammann thesis (TiHo Hannover, 2009) could not be opened.

16. Boss Paws. "German Shepherd Hip Dysplasia: Lines, Genetics and Prevention." bosspaws.ai — cited as an example of the claim, not as evidence. Read 4 November 2026.

17. Chen M. "Breed Susceptibility: Which Shepherds Are Most at Risk." 18 January 2026. hip-dysplasia-shepherds-com.pages.dev — cited as an example of the claim, not as evidence. Read 4 November 2026.

18. Komsta R, Łojszczyk-Szczepaniak A, Dębiak P. "Lumbosacral transitional vertebrae, canine hip dysplasia, and sacroiliac joint degenerative changes on ventrodorsal radiographs of the pelvis in police working German shepherd dogs." Topics in Companion Animal Medicine. 2015;30(1):10-15. doi:10.1053/j.tcam.2015.02.005 — abstract only, read 4 November 2026. It gives no hip percentages.

19. Verein für Deutsche Schäferhunde (SV). "The German Shepherd Dog Breed: Health Promotion through Research." 8 January 2019. schaeferhunde.de — read 4 November 2026.

20. Kealy RD, Olsson SE, Monti KL, Lawler DF, Biery DN, Helms RW, Lust G, Smith GK. "Effects of limited food consumption on the incidence of hip dysplasia in growing dogs." JAVMA. 1992;201(6):857-863. doi:10.2460/javma.1992.201.06.857 — abstract read 4 November 2026.

21. McGreevy PD, Mellor DJ, O'Neill D, et al. "Teleonomes in Tatters: The Welfare Costs of Selecting Dogs for Extreme Phenotypes, and the Case for Change." Animals. 2026;16(18):2867. doi:10.3390/ani16182867 — review; table entry read 4 November 2026, with its reference 28 (Asher et al., Veterinary Journal 2009) read as an abstract and reference 20 (McGreevy and Nicholas, Animal Welfare 1999) seen by title only.

22. Orthopedic Foundation for Animals. Disease Statistics by Breed. ofa.org — rendered in headless Chrome and searched 4 November 2026. 13,244 table rows on the page; one German Shepherd Dog row in the hips table; no line terms on the page. OFA restricts reproduction of its database, so we report what the page contains and not its figures. The table is undated on the page.

23. Quite Puppy. "German Shepherd Breeds Characteristics." quitepuppy.com — cited as an example of the claim, not as evidence. Read 4 November 2026.

24. Banfield CM, Bartels JE, Hudson JA, Wright JC, Montgomery RD, Hathcock JT. "A retrospective study of canine hip dysplasia in 116 military working dogs. Part II: Clinical signs and performance data." Journal of the American Animal Hospital Association. 1996;32(5):423-430. doi:10.5326/15473317-32-5-423 — abstract only, read 4 November 2026. Surfaced by an editorial check, not by our search strings: the abstract names no breed.

25. Banfield CM, Bartels JE, Hudson JA, Wright JC, Montgomery RD, Hathcock JT. "A retrospective study of canine hip dysplasia in 116 military working dogs. Part I: Angle measurements and orthopedic foundation for animals (OFA) grading." Journal of the American Animal Hospital Association. 1996;32(5):413-422. doi:10.5326/15473317-32-5-413 — abstract only, read 4 November 2026.

26. Worth AJ, Sandford M, Gibson B, et al. "Causes of loss or retirement from active duty for New Zealand police German shepherd dogs." Animal Welfare. 2013;22(2):167-174. doi:10.7120/09627286.22.2.167 — full text read 4 November 2026. Surfaced by an editorial check, not by our search strings. Based on handler recall; the authors state diagnoses were not confirmed from veterinary records.

Search strings, run 4 November 2026. PubMed: ("German shepherd"[tiab] OR "German shepherds"[tiab] OR "Alsatian"[tiab]) AND ("hip dysplasia"[tiab] OR "hip"[tiab]) AND ("working line"[tiab] OR "working lines"[tiab] OR "show line"[tiab] OR "show lines"[tiab] OR "show dogs"[tiab] OR "show-line"[tiab] OR "working-line"[tiab] OR "conformation line"[tiab]) (1 result) and ("German shepherd"[tiab]) AND ("hip dysplasia"[tiab]) AND (working[tiab] OR police[tiab] OR military[tiab]) (5 results). Europe PMC: ("German shepherd" OR "German shepherds") AND ("working line" OR "working lines" OR "working-line" OR "show line" OR "show lines" OR "show-line") (44 results) and ("German shepherd") AND ("hip dysplasia") AND ("working" AND ("show" OR "conformation" OR "showing")) AND (line OR lines OR type OR population) (117 results). 153 unique records, 152 with a PMC record, 144 returning full text; a script pulled every sentence pairing "working", "show", "conformation", "sport" or "police" with "line" or "type" and mentioning Shepherds, hips, dysplasia or arthritis. Author check, Europe PMC: AUTH:"Kennedy P" AND "German shepherd" AND hip (0 results).