OFA Statistics: What do you think?

[Gerry Clinchy]

The OFA has posted its most recent statistics for hip & elbow dysplasia. What do you think?

(alpha order)
Chessies Elbows
1507 evaluated 94.4% Normal 5.6% Dysplastic
Grade I (mild) 4.1% Grade II (moderate) 1.2% Grade III (severe) .3%

Curly Coated
204 evaluated 98.5% Normal 1.5% Dysplastic (AKC website says .5%)
Grade I .5% Grade II 0% Grade III 0%

Flat Coats Elbows
1508 evaluated 99.2% Normal .8% Dysplastic
Grade I .7% Grade II .1% Grade III 0%

Goldens Elbows
21,739 evaluated 88.4% Normal 11.4% Dysplastic
Grade I 9.0% Grade II 1.9% Grade III .5%

Labs Elbows
47,067 evaluated 88.7% Normal 11.3% Dysplastic
Grade I 8.3% Grade II 2.0% Grade III .8%

Tollers Elbows
264 evaluated 96.6 Normal 3.4% Dysplastic
Grade I 3.9% Grade II .4% Grade III 0%

Worst of the most numerous breeds:
Rottweiler 11,820 evaluated 58.8% Dysplastic 40.7% (AKC math error?)
Grade I 30.5% Grade II 8.6% Grade III 1.5%

Chessies Hips
11,484 evaluated Dysplastic 20.9% Excellent 11.8%

Curly Coated Hips
1,030 evaluated Dysplastic 15.3% Excellent 7.9%

Flat Coat Hips
4.651 evaluated Dysplastic 4.3% Excellent 18.8%

Goldens Hips
121,067 evaluated Dysplastic 20.0% Excellent 3.8%

Labs Hips
202,065 evaluated Dysplastic 12.1% Excellent 17.4%

Worst of a more numerous breed:
Newfoundland
13,572 evaluated Dysplastic 25.3% Excellent 7.7%

Tollers Hips
1,316 evaluated Dysplastic 6.8% Excellent 16.6%

Personally, I think the OFA stats are way off. I find it hard to believe that Newfoundlands have only 5.3% more incidence of CHD than Goldens. I think that the incidence of CHD in both breeds is way higher than this, even among carefully bred dogs. I believe that the obviously dysplastic xrays simply aren't sent to OFA. I believe this probably accounts for the smaller numbers of Grade III elbows as well. It's hard to believe that if 58.8% of Rottweilers have elbow dysplasia, only 1.5% of those are Grade III (severe). I'm stunned that Goldens have continued to fall behind on their percentage of Excellent ratings (1/4 as many as Labs or Tollers).

[Franco]

I think that many with bad hips and/or elbows don't send thier xrays to OFA.

[Losthwy]

I would be interesting to see how these stats compare to the stats of 10, 20 years ago.

[firehouselabs]

It would be nice if submission by the vet was mandatory to OFA for evaluation and for study without having to post it on the website if the owner did not want to. That way the numbers would be more accurate? and OFA studies would be helped. The owners could opt to post the results just as they do now by initialing the box on the form.

[Swampcollie]

Gerry I think you're correct, the vast majority of obvious failures are not submitted. From an informational standpoint, it would be nice to have all dogs submitted, but many people just don't want to spend the money to send them in for evaluation when it's obvious the dog won't pass.

I don't doubt that numbers OFA published are accurate. But they're only representative of the dogs that were actually submitted, not the overall breed as a whole.

[Montview]

I think that many with bad hips and/or elbows don't send thier xrays to OFA.

I agree. I don't know of many personally who will actually submit films that obviously say a dog is dysplastic (hips, elbows, or otherwise), although that would probably really help with "real" numbers in our breeds. I also don't know of many veterinarians who will encourage pet owners to submit films on dysplastic pets, which would further help determine true percentages.

[gsc]

Many of the health guarantees require OFA report before they will replace a puppy. It is something that we can do to help improve OFA numbers, but unless they choose the have it posted, it doesn't help with the litter research.

[Gerry Clinchy]

It would be nice if submission by the vet was mandatory to OFA for evaluation and for study without having to post it on the website if the owner did not want to. That way the numbers would be more accurate? and OFA studies would be helped. The owners could opt to post the results just as they do now by initialing the box on the form.

This is exactly the procedure used by PennHIP. You can agree to have them published on their website (or not). In fact, OFA also will record your PH #s on the OFA website!

The vet must send the xrays to them. So, no matter what they look like, once the films are shot, they must go to PH, or the vet could lose his certification with PH. With PH the fee is about the same as OFA. The cost of PH xrays (there are 3 films shot) is significantly more than OFA. PH is very picky about the quality of their xrays. If they return them for improper positioning the vet basically has to re-do them at no additional charge to the patient (that happened to me once). When you pay $300 for a procedure & they send them back for improper position, the patient is not likely to be a happy camper, so the vet would probably lose a patient if they didn't re-do at no charge. That is surely motivation to get it right the first time. Has anyone heard of OFA refusing to read an xray for poor positioning?

Almost all the experts agree that one should use OFA and one other mode of xray for the most accurate picture. PH's 3 views include the OFA view as well. The reason I was given (at a lecture by Dr. Smith) is that the OFA view is the best for detecting DJD, though not the best way to assess hip joint laxity.

Also OFA results are impugned by lack of consistency in the type of sedation used. A study has been done (funded in part by Golden Retriever Foundation) that shows that if a parent is xrayed without anesthesia or with ACE as the sedative, and the offspring are done with different sedative, the expected hip results are not consistent (offspring are not as good as the parents). Reasoning: muscle tension with a conscious or mildly sedated dog will exert influence over the appearance of the tightness of the hip joint. Even though the dog may "appear" relaxed, there is muscle tension present which will tend to hold the hips tighter in the joint. With more complete muscle relaxation, you will see the hip joint as it truly is.

The same kind of incomplete results shows up with the prcd-PRA testing for Goldens. By best count, at least 550 Goldens have been tested. As of 4/4/09 only 102 appear on the OFA website. Of those only 7 are "carriers", all the rest are clear. There are 318 Goldens listed on www.GoldenDNA.com, of which 63 are carriers and 3 affected. There is no charge for listing a tested dog on this website. So, even when it's "free", the figures are lacking not reflecting the total numbers tested.

[Gerry Clinchy]

I would be interesting to see how these stats compare to the stats of 10, 20 years ago.

From GRCA News, March-April 1977

Quoting information in RFT News of January 1977:

AKC registrations from 1/74 to 12/76 for Labs were 109,924; for Goldens 68,866. Percentage of xrays submitted v. registration was 2.5% for Labs and 3.8% for Goldens.

At the time OFA believed it received only 1/3 of the xrays actually taken by vets. Vets were essentially acting as pre-screeners of the obviously dysplastic hips.

12.7% of Labs rated "Excellent" v. 2.8% of Goldens.
compared to Labs Excellent today 17.4% Goldens 3.8%

58.9% of Labs rated "Good" v. 46.2% of Goldens
12% of Labs rated "Fair" v. 27.2% of Goldens
Total not dysplastic: Labs 83.6% Goldens 76.2%

7.8% Labs rated "mild" v. 14.7% of Goldens
6.1% Labs rated "moderate" v. 7.4% of Goldens
1.1% Labs rated "severe" v. 1.0% of Goldens

Total dysplastic: Labs 15% Goldens 23.1%
compared to Labs dysplastic today 12.1% Goldens 20%

Even back 30 years, the Labs were doing better in hips than Goldens. They have continued to gain more in the "Excellent" ratings than Goldens. But overall dysplastic has decreased about the same amount for both breeds. (which might mean that local vets are about the same in ability to pre-screen the obviously dysplastic hips).

If anyone is a packrat and has the Jan. 1977 issue of RFTN there may have also been mention of the other retriever breeds, but only Labs and Goldens were mentioned in the GRCA article because they were of somewhat comparable size in population, while the other breeds were of much smaller numbers back then.

[HiRollerlabs]

What if OFA were to offer "free" evaluation/recording if the hips/elbows were dysplastic? More people would be inclined to send in the xrays for OFA evaluation and we'd have better statistics.

Is there any information on how many times xrays are resubmitted and the second submission is not dyplastic? From other posts on this topic "you can make good hips look bad with poor positioning but you cannot make bad hips look good"--so not trying to stir that pot. Just saying, we have a dog that graded "excellent" hips on 2nd set of xrays and "dysplastic" on 1st set that were poorly positioned. On the second set we used a Stillwater, MN vet who knows how to position the dog for hip xrays.