Is My German Shepherd's Elbow Score Good? Reading One OFA Grade, One IEWG Grade or One BVA Number Against the Breed, and Why There Is No Elbow 'Fair'

An elbow grade is read differently from a hip grade, and the habit most owners bring from the hip certificate gets it wrong in both directions. On a hip certificate there are passing grades below the top: OFA Fair is one of three normal grades, and FCI B is "near normal" and not counted as dysplastic. On an elbow certificate there is no second rung that passes. Every scheme grades elbows on the same main scale, 0 to 3, and anything above 0 is dysplasia, because the grade records arthritis already present in the joint or a visible lesion that causes it. Some countries add a borderline grade just above 0; that is the only soft rung there is. Among Swedish German Shepherds, where every screening film taken is counted, a 0 puts the dog with roughly 83 in 100; a 1 with the next 11; a 2 or 3 with the worst 6.^[8] For breeding, a 1 is worse than it looks. For the dog in your house, a 1 is neither a forecast of lameness nor nothing: in a Swedish follow-up, owners reported lameness in the month before the interview about as often for grade 1 dogs as for grade 0 dogs, yet the grade 1 dogs had more often been given painkillers for their elbows. And a 0 promises less than it sounds, because the film mostly grades the bone's reaction to a lesion, not the lesion, and radiographs miss a share of early lesions.

The sentence that sounded like good news#

"Elbows normal." "ED 0/0." "Elbows: 0." It is often the shortest line on the certificate, and the hip line tends to take the whole conversation. Or it is the line that says 1, and the breeder's email says "just a grade 1, basically normal," and that sounds like the elbow version of a Fair hip.

Both readings carry over the grammar of the hip certificate, where most of the breed sits in a band of normal-but-not-perfect grades. The elbow certificate is written in a different grammar. We covered the hip half of the document, and the four hip crosswalks that disagree with one another, on the hip-score page. This page is for the other line.

What each scheme actually hands you#

All the major schemes descend from the International Elbow Working Group (IEWG) protocol, which is why an elbow grade converts across borders far better than a hip grade does. Here is what each document is.

The IEWG's boundaries are set mainly by the height of the new bone at the edge of the joint. Its 2010 table: grade 1 is osteophytes less than 2 mm high, or sclerosis at the base of the coronoid process; grade 2 is osteophytes of 2 to 5 mm, obvious sclerosis, a 3 to 5 mm step between the radius and ulna, or "indirect signs for a primary lesion"; grade 3 is osteophytes over 5 mm, a step over 5 mm, or an "obvious presence of a primary lesion (UAP, FCP, OCD)".^[3] The OFA's grades use the same three millimetre bands, which the OFA measures along the anconeal process, with trochlear notch sclerosis added at Grade II,^[1] and its FAQ says the protocol "is based on the IEWG recommendations".^[2] We found no document from any of these bodies that formally equates one scheme's grades with another's, so treat a cross-border comparison as close, not official. But it is far closer than the hip comparison, where four published crosswalks disagree about the second-best grade.

Where your Shepherd's grade sits among screened Shepherds#

The striking thing about this table is how little it moves. The breed's hip figures swing widely between national registries, and most of that spread is who was allowed to count. The elbow figures from a Swedish scheme where every film taken is recorded, a voluntary British one and an OFA analysis that included the withheld records land within about two points of each other: 16.9, 17.8 and 19. All three count screened dogs, not the whole breed, but three differently biased registries agreeing is worth something. Roughly one screened German Shepherd in six has a dysplastic elbow grade.

Two refinements. Males do worse: in the OFA analysis the German Shepherd's sex effect on elbow score was 0.364 ± 0.051, where positive values mean lower risk for females.^[9] And hips and elbows travel together. Among the Swedish Shepherds, 87.0% of dogs with A hips had grade 0 elbows, against 74.5% of dogs with D or E hips, and grade 2 or 3 elbows were more than twice as common behind the worst hips (9.6%) as behind the best (4.1%).^[8] A poor hip line is a reason to read the elbow line more carefully, not less.

dogorthopedic.com

Why there is no elbow "Fair"#

On the hip certificate, three OFA grades are normal and the lowest of them, Fair, is a hip that is not perfect but passes. Breeders who know hips often assume the elbow scale works the same way, with a grade 1 as the elbow's Fair. The OFA's FAQ takes up that exact belief, that "Grade I elbows are within the 'normal' range, much like Fair hips are considered to be within the normal range", and answers it: "The OFA's distinction between normal and abnormal elbows is actually more clearly defined than are the differences between fair and borderline hips. Elbows are diagnosed as dysplastic when evidence of Degenerative Joint Disease (DJD) is present as evidenced by osteophytes or sclerosis. It is not a gradual continuum from normal to abnormal, in which minor differences might be interpreted as normal by one reader and abnormal by another."^[2]

The reason is what the grade measures. A hip grade reads the shape and fit of the joint, which is why it can describe a range of normal. An elbow grade reads the joint's response to something going wrong inside it. The IEWG describes the reading as two stages, "a) to assess the degree of secondary joint disease (arthrosis) and b) to check for signs of a primary lesion."^[3] The OFA's page is blunter: "The only grades involved are for abnormal elbows with radiographic changes associated with secondary degenerative joint disease."^[1] The primary lesions underneath are the fragmented medial coronoid process, osteochondrosis of the humeral condyle, and the ununited anconeal process. The grade is mostly a measurement of the bone the joint has laid down in reaction to them.

So a grade 1 elbow is not a slightly imperfect normal elbow. It is an elbow with osteophytes under 2 mm high or sclerosis at the base of the coronoid process, which is arthritis, early and small. On the hip certificate there are passing grades below the top. On the elbow certificate every grade after 0 is dysplasia: arthritis already present, or at grades 2 and 3 possibly a primary lesion the film can see.

That is the breeding reading, and the numbers behind it are stark. The OFA's FAQ, undated, gives mating results for 13,151 breeding pairs in its database, breeds not stated: two normal-elbowed parents produced affected offspring 12.2% of the time; a normal parent with a dysplastic one, 26.1 to 31.3%; two dysplastic parents, 41.5%.^[2] In a large German study of 48,367 German Shepherds, 28,011 of them with elbow scores and 47,730 with hip scores, the authors "argued for both diseases against interpretation of borderline findings as healthy." Borderline there is the soft rung below grade 1, the Grenzfall some countries add just above 0, so their argument reaches further than ours: not only is a 1 not normal, even the grade short of a 1 should not be read as healthy.^[11] The German breed club has gone part of that way, and it is worth being exact about which part. The SV's own magazine maps its stamps as "ED 1/noch zugelassen, ED 2/mittlere ED, ED 3/schwere ED"^[12], and for dogs born from 1 January 2025 breeding approval requires "normal" or "fast normal" elbows and hips: "Die Befundung mit 'noch zugelassen' genügt nicht mehr."^[13] The notice says the rule applies once the new breeding regulations are entered in the club's official register.^[13] So the SV has stopped accepting a grade 1 for breeding, while "fast normal", which sits between normal and noch zugelassen in the order the German-line clubs list the results, still passes. Whatever the rule on a given date, a "still permitted" elbow is a grade 1.

What a grade 1 or 2 means for the dog in your house#

Now the other direction, which matters more to most readers, because most German Shepherds are not being bred. A dysplastic elbow grade is a diagnosis on paper. What it means for the dog is less dramatic than the word and more than nothing, and the largest follow-up we found shows both at once.

Researchers in Sweden took dogs of five breeds, 139 of them German Shepherds, graded 0, 1 or 2 by the kennel club, and interviewed the owners two to six years later.^[18] In the month before the interview, owners reported lameness in 16% of grade 0 dogs, 17% of grade 1 dogs and 30% of grade 2 dogs. The authors' summary of the day-to-day picture: "ED1-graded dogs had a lower risk than might be expected to develop visible clinical symptoms and showed a similar quality of life as dogs with ED0. ED2-graded dogs were more likely than ED0-graded dogs to have their lives impaired by lameness, according to the owners' perception."

The longer picture is less comfortable, in the same paper.^[18] By the interview, 25% of grade 0 dogs had been taken to a vet for orthopedic reasons at some point in their lives, against 38% of grade 1 and 43% of grade 2; that figure is not limited to the elbow. Grade 1 dogs had seven times the odds of having been given anti-inflammatory painkillers for the elbow, grade 2 dogs ten times. Seventeen grade 1 dogs had elbow surgery, against four grade 0 dogs, though in the authors' analysis grade was not significantly associated with surgery. And 61 dogs had already been euthanized for orthopedic reasons before the interviews; the authors found that outcome significantly associated with grades 1 and 2, and those dogs were left out of the analyses, so every figure in this paragraph describes the survivors. The OFA's position fits the day-to-day half: "it is incorrect to assume that a dog with Grade I ED will eventually develop Grade II or III ED."^[2] A grade 1 does not predict a limp. It does predict a dog more likely to be treated for his elbows, though the owners knew the grade, and the authors speculate that owners who paid closer attention were "more likely to seek veterinary care and get treatment, irrespective of the ED grading."

Then there is a sentence in the same paper that is worth sitting with: "Dogs with ED2 had moderate OA at the time of radiography, yet only 43% received veterinary care for orthopedic reasons at any point during the following 2- to 6-year period." The authors offer three explanations: owners saw no lameness or stiffness they judged worth a visit, some dogs do not show signs of arthritis until later in life, and some owners may expect a little lameness in a grade 2 dog and not seek care for it.

That is the site's whole argument in one sentence of someone else's paper. Among the surviving dogs whose films showed moderate arthritis, more than half were never taken to a vet for orthopedic reasons in the years after. Some showed no lameness at all, and arthritis on film and a visible limp line up loosely; the OFA cites a study of Rottweilers in which "68% of the non-lame dogs had DJD of the elbow."^[2] But the problem sits somewhere that is easy to under-read: dogs carry about 60% of their standing weight on the front legs, on average across 552 dogs of 123 breeds^[20], so a sore elbow is one of the two joints carrying the larger share of him. An owner watching for a limp is watching for a leg held up. A dog who is merely slower off the mark, or less keen on the stairs down, reads as a dog getting older.

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

What a 0 does and does not promise#

This is the part that makes an elbow 0 weaker reassurance than a hip A, and it is also the part where the evidence is more mixed than the claim usually made about it. Both halves belong here.

What supports caution. The OFA's own text: "If fragmentation of the medial coronoid only involves the cartilage, it may not be seen radiographically."^[1] In a 2025 study that put 424 elbow joints from 212 dogs of a screening population (median age 21 months, mostly retrievers, 13 German Shepherds) through both radiographs and CT, radiographs "correctly identified 65% of the cases (17/26)" of medial coronoid disease that CT found, and the authors conclude that "subtle or early-stage MCD changes are often missed."^[15] The same study's example of a missed case is a 24-month-old German Shepherd.

What cuts the other way. In the same screening study, because most elbows were healthy, a normal radiograph was right 98% of the time (369 of 378 joints). And in lame dogs, where the disease has had time to leave its mark, radiographs read by two experienced observers in consensus detected medial coronoid disease in 98% of affected joints, against CT, in a study with only two German Shepherds among its 90 dogs.^[14] Read together, the two studies suggest the film is good at finding elbow disease once it is established and weaker at finding it early, though they differ in both the dogs and the readers. One caution about the screening study's 98% for a clean film: it was measured where only 6% of joints had coronoid disease. Where disease is commoner among the dogs being screened, a clean film rules out less, and we have not found this accuracy measured in German Shepherds.

Time. The OFA's FAQ, which is undated, says it had performed "more than 19,000 GSD elbow evaluations"; its German Shepherd tables run to dogs born in 2001, so these are old figures. Of the resubmissions, 1,326 dogs sent in a second time and another 40 a third, 87% kept the same result, 3% went from dysplastic to normal, and "in 10% of the resubmissions, the change was from normal to dysplastic. The majority of the initial evaluations in these cases were preliminary reports, with the change most likely due to the onset of DJD as the dog matured."^[2] Dogs that get re-filmed are not a random sample, so this is not a rate; it shows only that a young normal can become a dysplastic grade later, and that it is not rare among the dogs who were checked again. And the OFA says of its own preliminary reads that "there are no long-term studies for preliminary elbow examinations like there are for hips."^[1] A European grade given at 12 months is a reading of the elbow a full year before the OFA would agree to certify it.

Age. A lifelong study of elbows followed 48 Labradors at Nestlé Purina, which funded it, until each died or was euthanized for severe disease. No dog had a fragmented coronoid, an ununited anconeal process or osteochondrosis. Even so, radiographic elbow arthritis was present at end of life in 12 of 23 normally fed dogs and 5 of 22 fed 25% less, and under the microscope in 12 of 23 and 18 of 22. The difference in how many dogs were affected was not significant on film or under the microscope; the restricted dogs lived a median 1.8 years longer.^[16] In that colony, elbows developed arthritis with age with no fragmented coronoid, ununited anconeal process or osteochondrosis in any of them (seven dogs did show an un-united medial epicondyle, a rarer lesion), and in the restricted group the microscope found far more than the film did. It was one breed in one colony, but a clean certificate at one year old was never designed to speak to that.

Read together: a grade 0 means no bone had reacted yet on the day of the film. It does not mean nothing was there to react to, and it says nothing about the elbow at ten.

dogorthopedic.com

The elbow lesion one study found only in German Shepherds#

One primary lesion is tied to this breed in the study below. The ununited anconeal process is a piece of bone at the back of the elbow that should fuse to the ulna and does not. In a 2026 Polish study of 108 dogs aged 3 to 15 months, referred for lameness, joint pain or routine breed screening, 69 of them German Shepherds, "UAP was identified in 9 cases and only in German Shepherds", and in 8 of those 9 it coexisted with another elbow lesion. Radiographs and CT agreed perfectly on it across the study's dogs (κ = 1.0), with only nine positive cases, and the authors say that "should not be interpreted as strong evidence of equivalent diagnostic performance"; they also ask for caution on the breed finding, since their breed mix was unequal.^[17] The paper's background section adds that fusion in large breeds "may not occur until between 20 and 24 weeks of age," and that the radiolucent line of an ununited process becomes diagnostic after about 20 weeks.

Two practical consequences. A German Shepherd puppy with a front-leg lameness at five or six months does not need to wait for a screening age to be filmed; this lesion can be seen before any certificate would be issued. And under the IEWG guidelines Sweden uses, "dogs with obvious presence of a primary lesion (UAP, MCPD or OCD) will be graded as ED 3", whatever the arthritis.^[8] What a 0 is weakest against is early coronoid disease, which the screening study above found the film missing in 9 of 26 affected joints.

If you are reading a parent's elbow grade before buying a puppy#

Ask for both elbows, not the overall grade, and ask which scheme graded them. Anything above 0 on a parent is dysplasia, whatever the breeder's email calls it. The OFA's pairing figures above, undated and not breed-specific, show the direction of the arithmetic you are buying into, and the German study estimated a genetic correlation of 0.3 between hip and elbow dysplasia in German Shepherds born 2001 to 2007, so a litter from a poor-hipped parent is not insulated on the elbow side.^[11] An OFA "Normal" on a dog filmed at 12 or 18 months is a preliminary read by one radiologist; ask whether there is a final one. And a German pedigree stamped noch zugelassen is a grade 1, whatever the breeding rules allowed in the year the dog was born.

What there is to buy for a grade, and the one load you can take off the elbow#

Nothing changes a grade, and the reader holding a 0 needs nothing. For a 1 or worse, there is nothing to buy for the grade itself either. There is one daily load you can take off the elbows, and it applies to the dog who gets out of a high car by jumping.

The front legs take the landing. In a study of fifteen healthy dogs jumping from platforms at 0.55, 0.65 and 0.75 m, heights chosen to match common car boot sills, the peak vertical force through the front limbs rose from 27.35 to 30.84 to 34.12 newtons per kilogram, every step significant.^[19] Divided by gravity, that is roughly 2.8 to 3.5 times the dog's body weight going through the front legs in one landing. That study measured force, not injury: its authors state that it "did not attempt to investigate the consequences of vGRF on joints and soft tissues," and the dogs were healthy. We found no study of car jumps in dogs with dysplastic elbows. But it puts a number on an ordinary daily event in which the front legs together take several times the dog's weight in one landing, and it is a part of your dog's day you can control easily.

The free fixes come first. Park so that he lands on a curb, not the road; a six-inch curb takes six inches off the drop. If the back-seat door opening is lower than the cargo floor, let him out that way. Lift his front end down if you can do it safely; for a full-grown Shepherd, at every stop, that is a lot to ask of a back, which is where a ramp comes in, with the honest cost that this one weighs about 25 lb, per its product page, and has to be carried and set up at every stop. Our car-loading guide has the measurements.

We card the Pet Gear tri-fold because at 71 inches it reaches a 30-inch cargo floor at about 25 degrees, and at 19.5 inches wide with raised side edges it gives a hesitant dog room to place his feet. Its faults are on the card, starting with the 30-day warranty. One finding cuts against us and we print it here: walking up a ramp asks for more elbow extension than climbing stairs, in a study of eight sound dogs on a custom-built ramp and stairs.^[21] That is range of motion, not impact, it was measured going up, and that study's ramp was not a car ramp. The load we are trying to remove is the landing, so for an elbow dog the ramp's job is the way down, and here there is a second finding against us. In three dogs trotting down a 15-degree grade on a force plate, the front legs took a larger share of the vertical load than on the level, 0.06 more of the total, and did most of the braking.^[24] A slope down puts more of the dog on his front end, not less. That was trotting, not walking, three dogs, and not a car ramp, and we found no study that measured a dog walking down one. Our reasoning is that a controlled walk at a shallow angle, even front-heavy, replaces a single landing at roughly three times body weight; that is an inference, not a measurement, so keep the ramp shallow and the pace slow. At a 30-inch SUV floor even this 71-inch ramp sits at about 25 degrees, steeper than the 15 degrees in that study, which is one more reason the curb and the lower door come first.

dogorthopedic.com

The other household lever is weight, and here we owe you the honest version. The lifelong Labrador colony above does not settle it for elbows. Fewer lean dogs had arthritis on film, though not significantly fewer, and their radiographic arthritis was significantly less severe at six years; under the microscope at the end of life, more lean dogs had arthritis, which the authors say is likely attributable to their greater age at death, and significantly more of two specific changes in the joint lining and cartilage. The authors read the study as support for "calorie restriction as a clinical tool to slow progression of elbow OA."^[16] We would put it more cautiously: for elbows, weight is a reasonable lever with mixed direct evidence, and the stronger reasons to keep a Shepherd lean are the ones on our hip pages. If you want to measure the ration rather than scoop it, we describe a gram scale elsewhere; any scale already in the house does the same job.

What to do this week#

If you are holding a 0, or an OFA Normal. File it. If it was a preliminary read or a European grade at 12 months, note the date on it; it describes the elbow at that age. Nothing else needs doing.

If you are holding a 1. It is dysplasia on paper. In the Swedish follow-up, across five breeds and counting only the dogs still alive at interview, grade 1 dogs were limping in the past month about as often as grade 0 dogs, but were more often treated for their elbows over the years, and euthanasia for orthopedic disease was associated with grade 1. That is a population, not a forecast for him. Do not breed from him on the strength of "basically normal." Do learn what a front-end problem looks like, which is not the limp you are expecting. Our page on limp versus stiffness versus weakness has the distinctions. Take a short phone video of him trotting towards you on a flat surface now, while he is sound, so there is something to compare against in two years.

If you are holding a 2 or 3. Ask your vet what the film showed, arthritis alone or a primary lesion, and whether CT would change anything. In the Swedish follow-up, most surviving grade 2 dogs, across the five breeds, were never taken to a vet for orthopedic reasons. Book the conversation now, while nothing is wrong enough to force it. If you have not yet insured him, the policy page explains how pre-existing-condition rules treat a dysplasia finding already in the record; it is written about hips, so read it with his elbow grade in mind.

If you are reading a parent's grade. Both elbows, the scheme, final or preliminary, and the date the dog was born if the stamp is German.

The delay this site exists to shorten is usually the months between the first quiet sign and the day someone names it. With elbows the name often comes first, on a certificate at one year old, and then the delay starts. The owner files "grade 1, basically normal" and reads the next five years of a slightly shorter stride as the dog getting older. The certificate did its job on the day it was written. Watching his front end is how you keep it up to date.

Sources#

1. Orthopedic Foundation for Animals. "Elbow Dysplasia." ofa.org — read 28 September 2026: "the OFA will not certify a normal elbow until the dog is 2 years of age"; "The OFA also accepts preliminary elbow radiographs. To date, there are no long-term studies for preliminary elbow examinations like there are for hips"; "there are no grades for a radiographically normal elbow. The only grades involved are for abnormal elbows with radiographic changes associated with secondary degenerative joint disease"; Grade I "Minimal bone change along anconeal process of ulna (less than 2mm)", Grade II 2–5 mm with trochlear notch sclerosis, Grade III over 5 mm; "If fragmentation of the medial coronoid only involves the cartilage, it may not be seen radiographically"; "The OFA started its elbow database in 1990 using a modified protocol of the IEWG"; the extreme flexed medial-to-lateral view required.

2. Orthopedic Foundation for Animals. "Elbow FAQs." ofa.org — undated, read 28 September 2026: three board-certified radiologists per evaluation, "single reading on preliminary reports"; "The OFA elbow protocol is based on the IEWG recommendations"; "It is not a gradual continuum from normal to abnormal"; "it is incorrect to assume that a dog with Grade I ED will eventually develop Grade II or III ED"; 13,151 breeding pairs: normal × normal 12.2%, normal × dysplastic 26.1–31.3%, dysplastic × dysplastic 41.5% of offspring affected; more than 19,000 German Shepherd elbow evaluations, 1,326 resubmitted, 87% unchanged, 10% normal to dysplastic (majority after preliminary reports), 3% dysplastic to normal; Grondalen, 207 Rottweilers, "68% of the non-lame dogs had DJD of the elbow"; "The OFA does not participate in the decision process whether or not to breed an animal."

3. Tellhelm B, Amort K, Ondreka N. "Grading primary elbow dysplasia lesions and elbow osteoarthrosis according to the IEWG protocol." WSAVA World Congress Proceedings, 2014, including the IEWG table "Scoring (Updated 2010)". VIN — two-stage reading ("a) to assess the degree of secondary joint disease (arthrosis) and b) to check for signs of a primary lesion"); grade definitions quoted in the text; borderline grade used "in some countries (e.g., Germany, France, Italy)"; The IEWG's own site (vet-iewg.org) could not be reached on 28 September 2026.

4. Flückiger M. Screening for elbow dysplasia. WSAVA World Congress Proceedings, 2004. VIN — "Minimal age for routine screening is 12 months."

5. British Veterinary Association. "Elbow Dysplasia Scheme." bva.co.uk — read 28 September 2026: "The dog must be at least 1 year old"; grades 0 radiographically normal, 1 mild osteoarthritis, 2 "Moderate osteoarthritis or a primary lesion with no osteoarthritis", 3 "Severe osteoarthritis or primary lesion with osteoarthritis"; "the overall elbow grade is determined by the higher of the two individual grades"; "CHS recommends only breeding from dogs that have an elbow grade of 0."

6. British Veterinary Association. "Elbow Dysplasia Scheme Procedure Rules," effective 1 January 2026. bva.co.uk (PDF) — two mediolateral views per elbow (extended about 110°, flexed about 45°); "Two scrutineers agree the grade for each radiograph."

7. British Veterinary Association. "Elbow Dysplasia Scheme Breed Specific Statistics 2024." bva.co.uk (PDF) — read 28 September 2026, dogs tested since 1999: German Shepherd Dog 8,804 graded; grade 0 7,126 (81%), grade 1 902 (10%), grade 2 501 (6%), grade 3 275 (3%), grades 1–3 1,678 (19%).

8. 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 — Table 2, German Shepherd Dog elbows 2007–2016, dogs 12–24 months: n 12,309; ED 0 10,232 (83.1%), ED 1 1,384 (11.2%), ED 2 383 (3.1%), ED 3 310 (2.5%), ED 1–3 2,077 (16.9%); median age at screening 1.3 years; "All radiographs obtained by the veterinary clinics must be sent to SKC for evaluation", ensured through a referral the owner buys before booking, and "All results are mandatorily recorded"; elbow screening "is based on guidelines from the International Elbow Working Group", and "dogs with obvious presence of a primary lesion (UAP, MCPD or OCD) will be graded as ED 3"; "dogs with severe ED often seek veterinary care before 12 months of age, and hence do not participate in official screening." Table 4, German Shepherd ED by hip grade: HD A 87.0% ED 0 and 4.1% ED 2–3 (2.3 + 1.8); HD D/E 74.5% ED 0 and 9.6% ED 2–3 (5.4 + 4.2); association significant after Bonferroni correction in the breed.

9. 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 — OFA data 1970 through April 2015, public and withheld records, dogs 2 years or older; S1 Table, German Shepherd elbows n 37,233, 82.17% normal, 17.83% dysplastic (DJD Grade I–III); S2 Table, German Shepherd sex impact on elbow score 0.364 ± 0.051, "Positive values reflect a decreased risk of CHD or ED for females."

10. James HK, McDonnell F, Lewis TW. "Effectiveness of canine hip dysplasia and elbow dysplasia improvement programs in six UK pedigree breeds." Frontiers in Veterinary Science. 2019;6:490. doi:10.3389/fvets.2019.00490 — three-year rolling mean of 6.35% of Kennel Club-registered German Shepherds born 2012–2014 elbow-graded; "Participation in the BVA/KC hip and elbow screening schemes is voluntary."

11. 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 read via Europe PMC (PMID 21554416): 48,367 German Shepherds born 2001–2007, elbow scores for 28,011; "positive additive genetic correlation of 0.3 between CHD and both CED and CED-ARTH"; results "argued for both diseases against interpretation of borderline findings as healthy."

12. Verein für Deutsche Schäferhunde. SV-Zeitung 10/2014, "Gesundheit," feature including an interview with B. Tellhelm. schaeferhunde.de (PDF) — "(ED 1/noch zugelassen, ED 2/mittlere ED, ED 3/schwere ED)". The order of stamps (normal, fast normal, noch zugelassen, mittlere, schwere) is also listed in the Austrian SVÖ Zuchtordnung 2025.

13. Verein für Deutsche Schäferhunde. "Wichtiger Hinweis," 18 June 2025. schaeferhunde.de — "Für die Zuchtzulassung wird für Hunde ab Wurftag 1.1. 2025 ein HD- und ED-Befund von 'normal' oder 'fast normal' gefordert. Die Befundung mit 'noch zugelassen' genügt nicht mehr." Effective on entry of the new breeding regulations in the club's register.

14. Villamonte-Chevalier A, van Bree H, Broeckx B, Dingemanse W, Soler M, Van Ryssen B, Gielen I. "Assessment of medial coronoid disease in 180 canine lame elbow joints: a sensitivity and specificity comparison of radiographic, computed tomographic and arthroscopic findings." BMC Veterinary Research. 2015;11:243. doi:10.1186/s12917-015-0556-9 — 180 elbow joints of 90 lame dogs, two experienced observers in consensus; "For the radiographic evaluation, a sensitivity of 98 % and specificity of 64 - 69 % using CT as the technique of reference."

15. Vali Y, Villamonte Chevalier A, van Bree H, Broeckx BJG, Gielen I. "Cross-modality diagnostic agreement in canine medial coronoid disease in a screening population: radiographs vs. CT in 424 elbow joints." Veterinary Sciences. 2025;12(9):883. doi:10.3390/vetsci12090883 — 212 dogs, median age 21 months, 13 German Shepherds; radiographic sensitivity 65% (17/26), specificity 93% (369/398), positive predictive value 37%, negative predictive value 98% (369/378) at a prevalence of 6%; "subtle or early-stage MCD changes are often missed"; Figure 1, a 24-month-old German Shepherd normal on radiographs with a fissure on CT.

16. Huck JL, Biery DN, Lawler DF, Gregor TP, Runge JJ, Evans RH, Kealy RD, Smith GK. "A longitudinal study of the influence of lifetime food restriction on development of osteoarthritis in the canine elbow." Veterinary Surgery. 2009;38(2):192-198. doi:10.1111/j.1532-950x.2008.00487.x — 48 Labrador Retrievers, restricted dogs fed 25% less; end of life "when a dog either died spontaneously or was euthanatized as a result of severe disease"; radiographic elbow OA at end of life 12 of 23 control-fed (52.2%) and 5 of 22 restricted (22.7%), "no statistical difference in radiographic OA frequency between groups at any of the time points"; histopathologic OA 12 of 23 and 18 of 22, "no significant difference"; restriction "resulted in a 1.8-year extension in median lifespan"; radiographic OA severity greater in control-fed dogs "at 6 years only" (P < .05); restricted dogs had higher frequency of synovitis (P = .021) and chondro-osseous metaplasia (P = .016), while "The increased frequency of histopathologic OA in the DR group is likely attributable to the greater age of the DR dogs at EOL"; Clinical Relevance: "These findings support calorie restriction as a clinical tool to slow progression of elbow OA"; "Fragmented medial coronoid process, un-united anconeal process, and osteochondrosis were not present in any elbow"; funded by and conducted at Nestlé Purina.

17. Kubiak-Nowak D, Hebel M, Prządka P, Antończyk A, Glińska-Suchocka K, Jankowski M, Spużak J. "Frequency and coexistence of ununited anconeal process with other primary elbow dysplasia in dogs." Life (Basel). 2026;16(7):1063. doi:10.3390/life16071063 — 108 dogs aged 3–15 months referred for lameness, joint pain or routine breed screening, 69 German Shepherds; "UAP was identified in 9 cases and only in German Shepherds"; coexisting lesions in 8 of 9; radiography and CT κ = 1.0 for UAP; fusion "may not occur until between 20 and 24 weeks of age"; diagnostic "in dogs exceeding 20 weeks of age"; the authors' caution on unequal breed distribution.

18. Bergström A, Johard S, Lee MH, Comin A. "Long-term prognosis of quality of life in dogs diagnosed with mild to moderate elbow dysplasia in Sweden." Frontiers in Veterinary Science. 2020;7:572691. doi:10.3389/fvets.2020.572691 — five breeds including 139 German Shepherds (53 ED 0, 40 ED 1, 46 ED 2), owners interviewed in 2017; Table 4: lameness in the past month 42/269 ED 0, 40/238 ED 1, 59/197 ED 2; sought veterinary care 68/269, 91/238, 84/197; elbow surgery 4, 17, 22; the "Impact" summary quoted in the text; odds ratio for NSAID treatment 7.0 (ED 1) and 10.4 (ED 2) against ED 0; 61 dogs euthanized for orthopedic reasons, "a significant statistical association between the occurrence of euthanasia due to orthopedic disease and ED scores of 1 and 2 in the sample (chi-square p < 0.001)", excluded from "both the orthopedic index part of the study and the statistical analyses"; "It can be speculated that owners who paid closer attention to orthopedic symptoms ... were also more likely to seek veterinary care and get treatment, irrespective of the ED grading"; "Dogs with ED2 had moderate OA at the time of radiography, yet only 43% received veterinary care for orthopedic reasons at any point during the following 2- to 6-year period."

19. Pardey D, Tabor G, Oxley JA, Wills AP. "Peak forelimb ground reaction forces experienced by dogs jumping from a simulated car boot." Veterinary Record. 2018;182(25):716. doi:10.1136/vr.104788 — 15 healthy adult dogs; the study "did not attempt to investigate the consequences of vGRF on joints and soft tissues"; platforms of 0.55, 0.65 and 0.75 m; peak forelimb vertical force 27.35 ± 4.14, 30.84 ± 3.66 and 34.12 ± 3.63 N/kg. The conversion to multiples of body weight (÷ 9.81) is ours.

20. Fish FE, Sheehan MJ, Adams DS, Tennett KA, Gough WT. "A 60:40 split: differential mass support in dogs." The Anatomical Record. 2021;304(1):78-89. doi:10.1002/ar.24407 — 552 dogs of 123 breeds; grand mean 60.4% of body mass supported by the forelimbs while standing, individual range 47.6–74.4%; no breed-specific German Shepherd figure.

21. Carr JG, Millis DL, Weng HY. "Exercises in canine physical rehabilitation: range of motion of the forelimb during stair and ramp ascent." Journal of Small Animal Practice. 2013;54(8):409-413. doi:10.1111/jsap.12111 — eight dogs; "ramp ascent had significantly greater range of motion compared to stair ascent", including greater elbow extension.

22. Hou Y, Wang Y, Lu X, Zhang X, Zhao Q, Todhunter RJ, Zhang Z. "Monitoring hip and elbow dysplasia achieved modest genetic improvement of 74 dog breeds over 40 years in USA." PLoS ONE. 2013;8(10):e76390. doi:10.1371/journal.pone.0076390 — Figure 2 legend: "Few elbow scores were released except for category 'Normal' before 2000."

23. Verein für Deutsche Schäferhunde. "ED-Verfahren" information brochure (ZBA 917), October 2001. schaeferhunde.de (PDF) — minimum age "mindestens 12 Monate"; categories normal, Grenzfall (borderline), ED 1, ED 2, ED 3.

24. Lee DV. "Effects of grade and mass distribution on the mechanics of trotting in dogs." Journal of Experimental Biology. 2011;214(3):402-411. doi:10.1242/jeb.044487 — three adult dogs, 27–34 kg, trotting on a 15° grade; downhill, the forelimb share of normal impulse rose 0.06 relative to level trotting, and the limbs exerted only braking force apart from slight late-stance hindlimb propulsion.

This page is a reading guide and does not replace veterinary advice. What a result means for breeding, whether further imaging is worthwhile, and whether a front-leg change needs examination are questions for your own vet, with your dog's films on the screen.