German Shepherd Hip Dysplasia by the Numbers: Six Countries, Five Different Rates, and Why They Disagree

There is no single number for hip dysplasia in German Shepherds, and anyone who gives you one without saying where it came from is hiding the most useful part. The published rate runs from 18.9% in the American database, where owners decide which films to send, to 25.6% in Sweden, where every official film is sent, to 32.4% in Switzerland over two decades, to 55% in a closed guide-dog breeding colony scored for its own records rather than for a registry. The breed did not change between those studies. The census did. Every table on this page is reconciled against that one fact, and the practical reading is short: treat every published number as a floor, assume your dog's odds are one in four to one in three unless her parents' scores say otherwise, and know that only two rows in these tables are still yours to edit: the age at neutering, decided once, and the ration, decided every day.

How common is hip dysplasia in German Shepherds? Every prevalence figure, in one table#

These are the population-level figures we could verify at source. Each row states who was counted, at what age, and on which scale, because those three things move the number more than the breed does.

Read the last row carefully, because it is the one that explains the others. In one ordinary year at ordinary clinics, 2.65% of German Shepherds had hip dysplasia written in their record. On the films, one in four to one in three have it. Those are not the same measurement, and the gap is not all one thing: some of it is dogs whose grade C hips never trouble them, some of it is diagnoses made in an earlier year and not written again in 2013, and some of it is dogs whose hips hurt and whose record says nothing. Nobody has measured how the gap divides. The third part is the one this site exists for, because it is measured in years per dog.

The OFA row deserves a note. The Orthopedic Foundation for Animals publishes live breed statistics on its own site, and those figures move as new evaluations arrive. OFA states that its website content, including database results, is copyrighted and may not be reproduced without permission, so the figure above comes instead from the peer-reviewed analysis of the same database through April 2015 by Oberbauer, Keller and Famula, which is open access and which we could read in full.^[1] If you want today's OFA number, go and read it there; expect it to be near, not identical to, 18.9%.

Why do the numbers disagree so much?#

Four levers move a prevalence figure, and none of them is the dog.

dogorthopedic.com

Who submits. This is the largest lever and it has been measured directly. In a prospective study at the University of Pennsylvania, clinically normal Golden Retrievers and Rottweilers referred for hip evaluation had a dysplasia prevalence of 53% to 73% and 41% to 69% respectively when every film was read. Among a second group of dogs radiographed for OFA certification, only 53% of the films were actually sent to OFA. Of the films submitted, 92% were normal; of the films withheld, 50% were normal. A radiograph with normal-appearing hips was 8.2 times as likely to be submitted.^[7] The authors of the 60-breed OFA analysis say the same thing about their own data: dogs never evaluated, and dogs evaluated but never submitted, are not in the database, and without mandatory participation the data are incomplete.^[1] An earlier analysis of 760,455 OFA hip scores put it more bluntly, describing the voluntary reports as severely biased against exposing dogs with problems.^[8]

Sweden removes the post-film half of this lever. From the clinics performing the radiographs, all obtained films are submitted to the kennel club for central evaluation, and all results are mandatorily recorded.^[2] That is why the Swedish 25.6% is the single most trustworthy population figure in the table, and why the same authors still warn that severe cases may be under-represented, because a dog with clinical signs before screening age may never be screened at all. Even the honest census misses the dogs that hurt earliest.

Age at the film. OFA's final evaluation is of radiographs taken at 24 months of age or older.^[20] Sweden screens at 12 to 24 months, median 1.3 years.^[2] The Seeing Eye scored at 12 to 16 months. The subjective hip-extended score is least stable in young dogs: in a three-year prospective study of 142 large-breed dogs filmed at 4, 6, 12, 24 and 36 months, the OFA-style score's consistency over time was 0.06 to 0.39 against 0.55 to 0.91 for the distraction index, and the authors questioned its predictive use before two years.^[9] A one-year-old and a two-year-old census are not counting the same thing.

The scale. OFA uses seven categories and calls three of them normal. The FCI scale runs A to E, and a grade C, "mild", is counted dysplastic in the European tables above. The British scheme does not classify at all; it sums nine traits per hip into a number from 0 to 106 and publishes a median, which over the last five years is 11 for the German Shepherd against 8 for the Labrador and 11 for the Golden Retriever.^[4] PennHIP measures laxity as a distraction index, a continuous number rather than a category,^[9] and so does not produce a prevalence at all. Which test you should choose for your own dog is a separate decision with its own page.

The year. The breed has been improving for decades in every scheme that has tracked it, so a figure pooled over 1970 to 2015 is an average across decades in which the breed was measurably improving, and understates where the breed stands now. The next section is about that.

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

Are German Shepherd hips getting better over time?#

Yes, in every scheme that has measured the same population twice. The table gives the trend where the source reported actual figures.

Two things to hold alongside the good news.

The first is that improvement in a voluntary registry is partly improvement in what gets sent. The Swiss authors note that films with clinical or obvious radiographic signs are submitted less often, and the OFA analysts say the voluntary bias was worst at the beginning of their period.^[3][8] The Seeing Eye result is the cleanest: a closed colony scored for its own breeding decisions, with no registry to impress, and the rate more than halved through selection on estimated breeding values. That is the proof that selection works, and it is also the reason the 55% matters. It is the starting figure on this page with the least reason for anyone to have edited it.

The second is the direction of the clinical numbers. In 1,243,681 dogs reported to the Veterinary Medical Database between 1964 and 2003, the diagnosed prevalence of hip dysplasia rose significantly across the four decades.^[11] That is not a contradiction of the screening trend. It measures how often the condition is found, in a population that is brought to a hospital, in decades when more owners went looking. Screening data say the breed's hips are tighter than they were. Hospital data say more of the loose ones are being named. Both can be true, and the second is the one you can act on.

How does the German Shepherd compare with other breeds?#

The breed is high-risk but not the worst, on any scale, and a reader who has been told "German Shepherds are the hip dysplasia breed" should see the company it keeps.

Blank cells are figures the source did not report or that we did not verify, not zeros. In the OFA analysis the German Shepherd ranks eighth of 60 breeds for hip dysplasia and sixth for elbow dysplasia at 17.8%.^[1] In Sweden it sits almost exactly beside the Golden Retriever, and both are far below the Dogue de Bordeaux, the Bullmastiff and the American Staffordshire Terrier.^[2] Elbows are the breed's other joint and, in the Swedish data, 16.9% of 12,309 German Shepherds had elbow dysplasia, and a worse hip grade co-occurred with a worse elbow grade in this breed specifically.^[2] The rest of the skeleton, including the lumbosacral spine that gets mistaken for hips in middle age, is covered in German Shepherd joint problems.

There is one number in which the German Shepherd is genuinely unusual, and it points the other way. In the PennHIP database analysis of 15,742 dogs, German Shepherds had the tightest mean hips of the four breeds studied, a mean distraction index of 0.41 against 0.48 to 0.53 for Labradors, Rottweilers and Goldens, yet 4.95 times the probability of radiographic joint disease of the other three breeds combined, with no joint disease observed below a breed-specific index of 0.28.^[12] For the same looseness, a German Shepherd is more likely to develop arthritis than the other three breeds. That is the fact behind the breed's reputation, and it means a distraction index that would be unremarkable in a Labrador is over the line in this breed.

How heritable is it, and what does that mean for a buyer?#

Heritability is the share of the variation in hip scores that is explained by genes. It is not the chance a puppy inherits the disease from a dysplastic parent, and it depends on which score was measured.

dogorthopedic.com

The spread from 0.2 to 0.58 is mostly a difference in scoring systems and statistical models, not a disagreement about the breed. Every estimate is high enough that selecting parents on hip scores works, which is what the trend table shows. And the German analysis found something a buyer can use directly: the "borderline" grade was genetically distinct from "free", so a borderline parent is not a clean parent, and the authors argued against reading borderline as healthy.^[14] The German analysis adds the other half: selection on relative breeding values, which combine a dog's own score with its relatives', was more efficient than selection on the parents' own records.^[10] The Seeing Eye colony is what that looks like when it is done: parents chosen on estimated breeding values, and the rate more than halved.^[5] A breeder who can show you the scores of both parents and of the grandparents and siblings is giving you a better number than one who shows you two OFA certificates.

What the heritability also says, in every version, is that somewhere between 42% and 80% of the variation in the score is not explained by the genes it measures. Some of that is measurement noise. Much of it is environment, and the largest single piece of environment has a trial behind it. That is where the rest of this page goes.

Does sex or neutering change the odds, and does it change life expectancy?#

One of the searches that reaches this site asks whether a German Shepherd's sex changes life expectancy with hip dysplasia. Here is what has been measured and what has not.

So the honest answer to the question as typed is in three parts. Sex changes the odds of hip dysplasia a little in the largest dataset, in favour of males, and not detectably in the clinical sample or in the joint-disease models; it is not a number to choose a puppy on. Sex does change how long a German Shepherd lives, by 1.4 years in favour of females in the one large study that measured it, but that study found no difference between the sexes in what they died of, and nobody has measured longevity in dysplastic German Shepherds separately by sex, so the specific number the question asks for does not exist. And the decision that does move joint risk in this breed is not the puppy's sex but the timing of neutering, which is a conversation to have with your vet before the appointment is booked rather than after.

The causes-of-death row is the one to keep. Musculoskeletal disorder and inability to stand were the first and second reasons a German Shepherd's life ended in that study, a combined 31.2%.^[6] Hips are not the only reason a Shepherd cannot stand; the spine, degenerative myelopathy and the elbows all contribute. But a dog whose hips were loose at one year and whose weight drifted up for nine is the ordinary route to that row, and the route has a lever on it.

The row you edit every day#

dogorthopedic.com

Everything above is a description of a population you cannot change. Genes were chosen by the breeder. The film has been, or will be, read. The scheme is whichever country you live in. Neutering age is yours, but it is decided once. The one input with controlled-trial evidence that the owner controls every day is how much the growing dog eats.

The trial was done in Labradors, not German Shepherds, and it should be described exactly. Forty-eight puppies were paired at six weeks; from eight weeks one dog in each pair was fed 25% less than whatever its pair-mate ate freely. At two years, by the OFA method, 7 of the 24 restricted dogs were dysplastic against 16 of the 24 free-fed dogs.^[16] Followed for life, the median age at which the restricted dogs first showed radiographic hip arthritis was 12 years against 6 for the controls.^[17] Put beside the tables above: a single feeding decision moved the two-year rate in that 48-dog colony from 67% to 29%, a bigger swing than two decades of selection produced across the Swiss registry. A trial and a registry are not a fair comparison, but the scale is the right one to hold in mind.

The German Shepherd-specific reason to care is in the primary-care data. Overweight or obesity was the fourth most common disorder recorded in UK German Shepherds, at 5.18%,^[6] and in a study of 50,787 neutered dogs at around 900 North American hospitals, overweight German Shepherds carried a 1.35-fold hazard of death compared with normal-condition ones, the smallest penalty of the twelve breeds studied but still a penalty.^[18]

Restriction in that trial was a weighed ration, and the household cup has been tested against a scale: 100 owners asked to measure a quarter, a half and one cup of kibble with three common devices ranged from 47.83% under to 152.17% over.^[19] A 25% reduction cannot survive an instrument with that error. The free version comes first: any gram scale already in the kitchen, and a weigh-in on the clinic scale at every visit, which tracks the thing that matters. If there is no gram scale in the house, this is the one we card.

What the scale cannot do is pick the number. The 25% in the trial was 25% less than a free-fed littermate, not a figure you can read off a bag, and the target for a growing German Shepherd is a lean body condition score agreed with your vet and re-checked at each visit. We have not used this unit; it is carded for its 1 g resolution, its tare button and its price, $13.99 on the day we checked, and its page says so. It also does nothing for a dog already grown: for an adult whose hips are already arthritic, weight loss still reduces lameness, and that evidence lives on the breed guide.

What to do this week#

If you are choosing a puppy. Ask for the hip scores of both parents, then ask for the grandparents' and the siblings'. A breeder who has them will not mind. Read a borderline as borderline. If the scheme is OFA, remember the number you are looking at was volunteered, and ask whether every litter-mate was submitted.

If your German Shepherd is under two. Decide this week which screening you will do and when; the PennHIP versus OFA page is the decision in one place, and if you are insuring, read what the policy says about hips before the film is taken. Weigh the food from tonight. Have the neutering-age conversation with your vet before the appointment exists.

If your dog is older and you came here for odds. The odds were one in four to one in three at the film and 2.65% in the record, and an unmeasured share of the space between those numbers is dogs being read as stiff, slow or just old. The signs by age are specific enough to check against tonight. The master guide covers what comes after the name.

Every number on this page is a floor, and the dogs missing from it are the ones who hurt. Nobody counted them because nobody sent the film, and in a house the same thing happens for the same reason. The delay is the injury. Go and count.

Sources#

1. 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 database 1970 through April 2015, restricted to dogs 2 years or older; 1,056,852 hip and 275,129 elbow records across 60 breeds; German Shepherd Dog 107,048 hip evaluations, 18.9% dysplastic (eighth of the fifteen highest-prevalence breeds; Newfoundland 24.8%, Bloodhound 24.7%, American Staffordshire Terrier 24.4%, Bullmastiff 24.0%, Rottweiler 20.1%, Chesapeake Bay Retriever 19.4%, Chow Chow 19.2%, Golden Retriever 18.8%) and 17.8% elbow dysplasia; heritability of hip score in the German Shepherd 0.58 ± 0.03 against a 60-breed average of 0.57; sex effect on hip ratings breed-dependent (males more favourable in 14 breeds, females in 15); S2 Table gives the German Shepherd sex impact on hip score as −0.141 ± 0.034, with the footnote that negative values reflect a decreased risk in males, and the age impact as 0.003 ± 0.001; the hip genetic trend for the three largest-population breeds (German Shepherd, Golden Retriever, Labrador Retriever) "did not deviate from the trends seen in the breeds reported above indicating that slow, substantive progress has been achieved in hip improvement"; "dogs that were never evaluated or those that were evaluated but for whom owners elected not to submit were not be captured... without mandatory participation, the data available in the OFA database are incomplete."
2. 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 — Swedish Kennel Club screening 2007–2016, dogs 12–24 months (median 1.3 years); 114,568 dogs screened for hips. German Shepherd Dog hips: 12,335 screened, A 4,876 (39.5%), B 4,305 (34.9%), C 2,175 (17.6%), D–E 979 (7.9%), C–E 3,154 (25.6%); elbows 12,309 screened, ED 1–3 2,077 (16.9%); worse hip grade co-occurred with worse elbow grade in the German Shepherd Dog. Golden Retriever hips 25.2% C–E; Dogue de Bordeaux 75.4%, Bullmastiff 65.1%, American Staffordshire Terrier 64.2%. "From the veterinary clinics performing the radiographic examinations, all the obtained radiographs are submitted to the Swedish Kennel Club for central evaluation... All results are mandatorily recorded." Limitation: severe cases may be under-represented because dogs with clinical signs before screening age might not be screened.
3. Ohlerth S, Geiser B, Flückiger M, Geissbühler U. "Prevalence of Canine Hip Dysplasia in Switzerland Between 1995 and 2016—A Retrospective Study in 5 Common Large Breeds." Frontiers in Veterinary Science. 2019;6:378. doi:10.3389/fvets.2019.00378 — official scores from the Swiss kennel clubs; 12,034 dogs of five breeds; German Shepherd Dog 5,326: overall prevalence 32.4% (C 22.5%, D or E 9.9%; A 27.9%, B 39.7%); 46.2% dysplastic in 1995–1999 falling to 18.0% in 2010–2016; A-rated rising from 13.0% to 46.5% and C-rated falling from 31.1% to 14.8%; minimum age for official scoring 12 months (18 in giant breeds); "Radiographs of dogs with clinical or obvious radiographic signs of CHD are less commonly submitted for official evaluation."
4. British Veterinary Association. "Hip Dysplasia Scheme Breed Specific Statistics 2024." bva.co.uk (PDF), and "BVA Hip Dysplasia Scheme for dogs," bva.co.uk — both read 16 September 2026. German Shepherd Dog: 10,388 tested over 15 years, 2,525 over five years, 309 in 2024; 15-year mean 15, median 11; 5-year mean 14, minimum 0, maximum 104, median 11. Comparison 15-year medians: Labrador Retriever 9 (49,137 tested; five-year median 8), Golden Retriever 10 (16,325; five-year 11), Rottweiler 8 (2,650; five-year 8), Newfoundland 10 (1,578; five-year 10). Scheme page: each hip scored on nine parameters, minimum 0 and maximum 53 per hip, total 0 to 106; dog must be at least 1 year old; "CHS recommends only breeding dogs with hip scores below or close to the published breed medians."
5. Leighton EA. "Genetics of canine hip dysplasia." Journal of the American Veterinary Medical Association. 1997;210(10):1474-1479. PubMed 9154200 — closed breeding colony producing guide dogs; hip quality assessed for 2,037 German Shepherd Dogs and 1,821 Labrador Retrievers 1980–1996; parents selected on estimated breeding values; "In fewer than 5 generations of selection, the percentage of German Shepherd Dogs with canine hip dysplasia at 12 to 16 months of age decreased from 55 to 24%"; Labradors from 30 to 10%.
6. O'Neill DG, Coulson NR, Church DB, Brodbelt DC. "Demography and disorders of German Shepherd Dogs under primary veterinary care in the UK." Canine Genetics and Epidemiology. 2017;4:7. doi:10.1186/s40575-017-0046-4 — VetCompass, 12,146 German Shepherds among 455,557 dogs under care in 2013; disorder prevalence from a random sample of 1,660 records (median age 4.7 years): hip dysplasia 44 dogs, 2.65% (95% CI 1.93–3.54), females 2.42% versus males 2.89%, P = 0.546; overweight/obesity 86, 5.18%; osteoarthritis 92, 5.54%. Mortality from 272 deaths: median longevity 10.3 years (IQR 8.0–12.1), females 11.1 (IQR 9.1–12.5) versus males 9.7 (IQR 7.5–11.7), P = 0.001; "No differences in probability were identified between males and females for the 14 most common causes of death"; most common causes musculoskeletal disorder 16.3%, inability to stand 14.9%, neoplasia 14.5%, spinal cord disorder 13.6%.
7. Paster ER, LaFond E, Biery DN, Iriye A, Gregor TP, Shofer FS, Smith GK. "Estimates of prevalence of hip dysplasia in Golden Retrievers and Rottweilers and the influence of bias on published prevalence figures." Journal of the American Veterinary Medical Association. 2005;226(3):387-392. doi:10.2460/javma.2005.226.387 — prospective; 200 clinically normal Golden Retrievers and 140 Rottweilers aged 24–60 months screened (prevalence 53–73% and 41–69%), plus 93 dogs radiographed for OFA certification: 49 (53%) submitted; 45 of 49 submitted (92%) normal against 22 of 44 not submitted (50%); "Radiographs with normal-appearing hips were 8.2 times as likely to be submitted to the OFA"; "Results suggest substantial bias in the OFA database, which causes lower estimates of prevalence of CHD."
8. 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 — OFA registry 1974–2009; 760,455 hip and 135,409 elbow scores across 74 breeds; genetic improvement 16.4% of a phenotypic standard deviation for hips; heritability 0.22 for hips; the voluntary reports "have been severely biased against exposing dogs with problems, especially at beginning period."
9. Smith GK, Gregor TP, Rhodes WH, Biery DN. "Coxofemoral joint laxity from distraction radiography and its contemporaneous and prospective correlation with laxity, subjective score, and evidence of degenerative joint disease from conventional hip-extended radiography in dogs." American Journal of Veterinary Research. 1993;54(7):1021-1042. doi:10.2460/ajvr.1993.54.07.1021 — 142 large-breed dogs of 14 breeds radiographed at 4, 6, 12, 24 and 36 months; intraclass correlation over time 0.55–0.91 for the distraction index against 0.06–0.39 for the subjective OFA-style score; "The associated large error questions the predictive use of the 7-point, subjective hip-scoring scheme, particularly prior to the age of 2 years."
10. Janutta V, Hamann H, Distl O. "Genetic and phenotypic trends in canine hip dysplasia in the German population of German shepherd dogs." Berliner und Münchener Tierärztliche Wochenschrift. 2008;121(3-4):102-109. PubMed 18412027 — 157,455 German Shepherd Dogs of the German population born 1985–2003 and radiographed for hip dysplasia; "an almost linear and significant decline in both phenotypic and genetic trends for CHD scores (scale 1 to 5 for CHD free to severe CHD) over a period of 19 years, with b = −0.021 ± 0.002, and b = −1.69 ± 0.05, respectively"; selection on relative breeding values shown to be more efficient than selection on parents' phenotypes.
11. Witsberger TH, Villamil JA, Schultz LG, Hahn AW, Cook JL. "Prevalence of and risk factors for hip dysplasia and cranial cruciate ligament deficiency in dogs." Journal of the American Veterinary Medical Association. 2008;232(12):1818-1824. doi:10.2460/javma.232.12.1818 — cross-sectional; 1,243,681 dogs reported to the Veterinary Medical Database 1964–2003; castrated males more likely to have hip dysplasia (OR 1.21); "Prevalences of HD and CCLD increased significantly over the 4 decades for which data were examined."
12. Smith GK, Mayhew PD, Kapatkin AS, McKelvie PJ, Shofer FS, Gregor TP. "Evaluation of risk factors for degenerative joint disease associated with hip dysplasia in German Shepherd Dogs, Golden Retrievers, Labrador Retrievers, and Rottweilers." Journal of the American Veterinary Medical Association. 2001;219(12):1719-1724. doi:10.2460/javma.2001.219.1719 — 15,742 dogs in the PennHIP database; mean distraction index 0.41 in German Shepherds against 0.48–0.53 in the other three breeds; 4.95-fold probability of joint disease; no joint disease below a breed-specific index of 0.28; sex not significant and dropped from the final models.
13. Wilson BJ, Nicholas FW, James JW, Wade CM, Tammen I, Raadsma HW, Castle K, Thomson PC. "Heritability and phenotypic variation of canine hip dysplasia radiographic traits in a cohort of Australian German shepherd dogs." PLoS ONE. 2012;7(6):e39620. doi:10.1371/journal.pone.0039620 — 13,124 Australian German Shepherd Dogs born 1976–2005 scored on the nine BVA hip traits; heritability of individual traits 0.14–0.24 (ordinal), 0.14–0.25 (linear), 0.12–0.40 (binary); summed-score heritability 0.30 ± 0.02.
14. 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 — 48,367 German Shepherd Dogs born 2001–2007 in the German scheme, hip scores for 47,730; moderate heritabilities of 0.2–0.3 for the quasi-continuous traits; genetic differences between borderline and affected status (correlation of breeding values 0.5 for hips); results "argued for both diseases against interpretation of borderline findings as healthy."
15. Hart BL, Hart LA, Thigpen AP, Willits NH. "Neutering of German Shepherd Dogs: associated joint disorders, cancers and urinary incontinence." Veterinary Medicine and Science. 2016;2(3):191-199. doi:10.1002/vms3.34 — hospital records over 14.5 years on 1,170 intact and neutered German Shepherd Dogs, joint disorders followed through 8 years; intact males 7% with one or more joint disorders against 21% in males neutered before one year; intact females 5% against 16%; "The increased joint disorder incidence mostly associated with early neutering was CCL."
16. 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." Journal of the American Veterinary Medical Association. 1992;201(6):857-863. doi:10.2460/javma.1992.201.06.857 — 48 Labrador Retrievers paired from six weeks, one of each pair fed 25% less than its ad libitum pair-mate from eight weeks; at two years, 7 of 24 restricted and 16 of 24 free-fed dogs dysplastic by the OFA method.
17. Smith GK, Paster ER, Powers MY, Lawler DF, Biery DN, Shofer FS, McKelvie PJ, Kealy RD. "Lifelong diet restriction and radiographic evidence of osteoarthritis of the hip joint in dogs." Journal of the American Veterinary Medical Association. 2006;229(5):690-693. doi:10.2460/javma.229.5.690 — the same cohort followed for life; median age at first radiographic hip osteoarthritis 6 years control-fed against 12 years restricted-fed.
18. Salt C, Morris PJ, Wilson D, Lund EM, German AJ. "Association between life span and body condition in neutered client-owned dogs." Journal of Veterinary Internal Medicine. 2019;33(1):89-99. doi:10.1111/jvim.15367 — 50,787 middle-aged neutered dogs at around 900 North American hospitals, 1,811 German Shepherds; hazard ratio for death in overweight versus normal body condition, German Shepherd 1.35 (99.79% CI 1.05–1.73), the lowest of the twelve breeds studied.
19. Coe JB, Rankovic A, Edwards TR, Parr JM. "Dog owner's accuracy measuring different volumes of dry dog food using three different measuring devices." Veterinary Record. 2019;185(19):599. doi:10.1136/vr.105319 — 100 owners; individual accuracy from −47.83% to +152.17%.
20. Orthopedic Foundation for Animals. "Hip Screening Procedures." ofa.org — read 16 September 2026: "Radiographs of animals 24 months of age or older are independently evaluated by three randomly selected, board-certified veterinary radiologists from a pool of 20 to 25 consulting radiologists throughout the USA in private practice and academia."

This page is a data reference and does not replace veterinary advice. Screening choices, neutering age, feeding targets and surgical timing are decisions to make with your own vet, and where surgery is on the table, with a board-certified surgeon and your dog's radiographs on the screen.