PennHIP Myths
I came across a web page that contained so much misinformation about PennHIP, I felt I needed to clear it up.
PennHIP is the better test. It's the only one that claims to be able to predict problems, not just diagnose them. The fact that you can do it at six months means, as a breeder, I can run the test on a young dog and make the decision to spay or neuter before sexual maturity. Because I do PennHIP at six months, and OFA can't tell if there's a problem unless arthritis is present 18 months later I cannot do OFA. I rather suspect the authors of this got caught up in the early PennHIP back-lash and/or got a crappy PennHIP score on on of their foundation dogs. Iris' poorest hip is .47 which is in the 100th percentile. To suggest that she needs a .30 is rediculous. She's so much tighter than the average Corgi that breeding her should lead to better hips.
| Their Statement | My Comment |
| PennHIP is an alternative method of evaluating hips used by some breeders. | True enough. It is an alternative to OFA, and OFA-like tests used in other countries. |
| It is controversial in the veterinary community and it's validity is not universally accepted. | PennHIP is not controversial in the veterinary community. PennHIP is widely acknowledge to be the superior evaluation (click here). Nothing is ever "universally accepted." It isn't universally accepted that men have been to the moon. |
| With English Setters it has little practical value for selecting breeding dogs. | What? OFA has little value for selecting breeding dogs. People breed OFA Fair every day. |
| We believe that due to the way PennHIP has been marketed as the "magicpill" substitute for OFA evaluations, it has in all likelihood set backprogress in reducing hip dysplasia. | Before PennHIP, OFA was in business for about 30 years. During it's 30 year monopoly, it made no real progress towards reducing hip dysplasia. It's hard to have a "set back." After 40 years of using OFA, you have a 1 in 6 chance of getting an English setter with DJD. |
| Because the method does not reliably diagnose the disease some breederswho rely exclusively on PennHIP will unknowingly use dysplastic dogs intheir breeding programs, | Actually the veterinary definition of Hip Dysplasia is excessive hip laxity. The PennHIP method measures hip laxity, it doesn't judge it. |
| and others are using PennHIP because they know their dogs won't pass an OFA. | This is so untrue, I don't know where to start. |
| Understanding why we say this requires an in depth look at PennHIP | Understanding why you say this would require more than that. |
| PennHIP Reports and What They Mean | |
| PennHIP is not a pass/fail screening and there is no such thing as "PennHIP Certified", or "PennHIP'ed clear" of hip dysplasia. | That's half true. PennHIP recommends that only animals in the top half of the breeds be used for breeding. OFA is also not Pass/Fail they have a seven-layer scoring system, which is remarkably similar to the sliding scale of the DI. |
| PennHIP only addresses two criteria: | That's because (until someone figures out something better) you only need two criteria. |
| 1 - Laxity Dogs are given a score called the Distraction Index, or DI,which measures the amount of laxity (also called subluxation) in thehip and compares it to other dogs from the same breed. | |
| A score of 0 would mean no laxity, a score of 1 indicates a hip that can be forced completely out of the socket. | Yes, it's a percentage. You can look at the formula by clicking here. I have heard of hips so bad they couldn't be measured. The estimated DI was 1.20. |
| Based on the dog's DI the report will show where the dog lies in relation to the rest of the breed in the PennHIP database. | |
| For instance, a dog in the 90th percentile indicates hips that are in the tightest 10% of the breed. | Yes, that's what percentile means. If your dog is in the 90th percentile, it's hips are tighter than at least 90% of the other dogs who have been tested. |
| Breeders are told that using breeding dogs that score better than themedian score for the breed will decrease the overall incidence ofdysplasia. | Yes, by limiting your breeding selection to the best half of the dogs of a breed, the hips should get tighter over generations. |
| According to PennHIP, a DI of .3 or LESS almost always indicates good hips. | Regardless of breed, there is a very near zero percent chance of developing DJD. |
| Dogs with a DI of .7 or greater virtually always have dysplasia. | Again, the definition of dysplasia is looseness of the joint. I don't think these folks know the difference between dysplasia and degenerative joint disease. Dysplasia leads to DJD. |
| Also according to PennHIP's research, scores between .4 and .6 are not reliable enough to be useful. | That's just not true. Some dogs with scores in this range will develop DJD, and some will not. At .40 about 2% of Goldens will develop DJD and at .60 about 15% will. You can click here for the graph. |
| Most English Setter scores are higher than .4. | This is true, the median as of February 5, 2009 for English setters is .58. That means half of the ES are worse than .58. |
| 2 - Presence or absence of Degenerative Joint Disease (DJD) HipDysplasia is abnormal conformation of the hip joint. DJD is arthriticchanges to the joint that are a symptom of Hip Dysplasia, not thedisease itself. | Hip Dysplasia is (again) Hip Laxity not an "abnormal conformation." Dysplasia leads to DJD. DJD is actually the disease we are worried about. If your dog's hips are really loose, but it doesn't develop DJD, it's not in pain, so nobody would care. |
| Presence of DJD confirms that the dog has Hip Dysplasia, but absence of DJD does NOT mean the dog is clear of Hip Dysplasia. | Again, the Distraction Index confirms Hip Dysplasia. OFA does not claim to accurately predict future disease. |
| The Problem With PennHIP | |
| Because arthritic changes often take years to develop | Yes, that means you have to repeat your OFA test annually, because it's really the only thing they look for. PennHIP actually predicts the likelihood of developing DJD. |
| and there are multiple aspects of hip joint conformation that can be abnormal | Yes, and they are accounted for in the measurement process. A shallow socket will cause the DI to shoot up. |
| it is entirely possible for a dysplastic dog to have a fairly good looking PennHIP report. | No it's entirely impossible. The definition of Hip Dysplasia is Hip Looseness and PennHIP measures exactly that. |
| An example would be a dog with a middle of the road DI score like .45,which would be better than average for an English Setter, and shows noDJD at the time of evaluation. | OK, we are making up a fake dog in our head. This argument would be so much better if they had an example of an actual dog who had a .45 no DJD and was OFA dysplastic. |
| Many dogs like this would fail an OFA evaluation due to subluxation orone or more of the other abnormalities in the hip joint that PennHIPdoes not consider. | Earlier they said "Dogs are given a score called the Distraction Index, or DI,which measures the amount of laxity (also called subluxation)...." So PennHIP would have measured this and would not have a .45. In most of the comparisons I've seen, dogs are more likely to be judged OFA Good or Fair and then have a DI in the bottom half of their breed. Hip laxity is the best predictor of developing DJD. |
| Unfortunately such dogs would also meet PennHIP's recommended breeding criteria. This is a major short coming of the PennHIP method. | Yes, but such dogs do not exist except in the mind of the authors. |
| Unless the hips are very tight, very loose, or arthritic, there is NOway to tell from the report whether a dog has good or bad hips. | That's just not true. In fact, it's basically the truth about OFA. You need to wait two years before you can get an evaluation because basically the only thing they look for is DJD. OFA scores Excellent, Good, Fair, Borderline, and Dysplastic. PennHIP actually gives you a predictive number for the ones in the middle. |
| We don't use PennHIP because of this and for the following reasons: | |
| There is no breeding study or data which shows whether selecting for lower DI's decreases the incidence of HD in offspring. | After 40 years of OFA, HD has not been decreased. |
| The last time we talked to PennHIP (personal comments from Dr. GailSmith, the founder of PennHIP) they indicated no interest in pursuing abreeding study. | OFA had not done one either and is not pursuing such a study. The cost of such a study and the time involved would be long. OFA points to improved numbers of "excellent" ratings for dogs as improvement in the breeds. Labrador Retrievers had a median DI in 2001 of .50 and by 2008 the median was .46 according to my research. |
| All of the published studies that fully support the PennHIP method were authored or coauthored by Gail Smith. | The OFA does a lot of research too. Dr. Smith also participates in studies on other methods for predicting DJD. |
| Independent studies contradict PennHIP's claims. | Which? I can get into JAVMA archives. I haven't seen them. Independent studies have confirmed the claims. |
| PennHIP does not allow independent access to, or review of, it's database. | I'd like to see this too, but it doesn't mean the method is not good. |
| In English Setters, a breed which has a very high median DI score, | English Setters Median DI is .58. Average of all dogs is .48. I have compiled 19 breeds with better DI's and only 4 with worse DI's. |
| simply following PennHIP's recommendation of picking breeding dogsbetter than the median for the breed would guarantee using a highpercentage of dysplastic dogs in a breeding program. | Actually, no. It would probably eliminate many dogs that are graded OFA Fair, and some that are OFA Good or Excellent. As a breeder, you are free to use higher or lower criteria than the recommended minimum. |
| The median score is so high because the database contains only a fewhundred English Setters, a large percentage of which are from onekennel that has an exceptionally high incidence of HD (personalcomments Dr. Pamela McKelvie/PennHIP), making the median score uselessfor this breed. | It contains 644 dogs, which is enough to be statistically relevant. Again, if you don't like the median, breed tighter dogs. |
| Because there is no pass/fail, | There is no Pass/Fail in OFA either. |
| and because buyers do not understand how PennHIP works, | Buyers can look at the test themselves and figure out how it works. It's no more complicated than OFA. |
| using PennHIP as the ONLY hip screening method gives dishonest breedersa way to hide the fact that they are using dogs in their breedingprogram that would fail OFA evaluation. | Only using OFA gives dishonest breeders a way to hide the fact that they are using dogs who would be deemed undesiredable if a PennHIP evaluation was done. PennHIP is Measured. OFA is Judged. It's like the difference between getting a speeding ticket with a radar gun vs. an estimated speed. Estimated is unreliable. |
| This does not mean we believe all breeders who use PennHIP are intentionally dishonest. | Just most of us right? |
| In fact, a breeder who also has OFA certifications may be especially conscientious. | Unfortunately, I'm not subjecting my dogs to anesthesia risk twice, because OFA can't be done until 24 months. Heck, we bred Iris prior to her second birthday thanks to PennHIP. |
| PennHIP DOES NOT, as has sometimes been claimed, do an OFA - like evaluation for anyone, so any "certification" of this type must come from the OFA to be meaningful. | PennHIP does receive the "OFA view" x-ray. They (again) Measure it. You don't need an "evaluation" when you can pull out calipers and measure something. |
| PennHIP's database is fully closed and there is no way to verify DI's for an individual dog. | This is purportedly coming. I've signed off for participation in the open database for all of my dogs. With OFA veterinarians are not required to submit all films, so if a veterinarian or owner can see a problem, they don't submit it, rendering OFA's database useless as well. |
| OK, time to change the formatting, as the argument goes to a different level. Their advice is to the left, mine to the right, with rebuttal in italics. | |
| If you are considering buying a pup from a litter out of parents with PennHIP evaluations and no OFA's, here is what you need to look for: | If you are considering buying a pup from a litter out of parents with OFA evaluations and no PennHIP's, here is what you need to look for: |
| You need to see the actual PennHIP reports for both parents. | You need to see the actual OFA reports for both parents, or verify them on the internet. |
| The DI's of both hips of both parents need to be .3 or less. | The OFA's of both parents need to be Excellent. |
| PennHIP's should be in the top half for their breed, and tighter is even better. | |
| Both reports must show no DJD present. | Both reports must show no DJD present. |
| Both parents should be at least 2 years old at the time of the evaluation. | Both parents' OFA evaluations should be repeated annually and be done less than 12 months ago. |
| Research has proven that DI's don't change over time, they only need to be done once. Responsible breeders run PennHIP at six months of age. | |
| If all of these criteria are met it is very likely the parents of the litter have normal hips. | Actually, if all these criteria are met, the parents of the litter have "perfect" hips. As near to perfect as exist. |
PennHIP is the better test. It's the only one that claims to be able to predict problems, not just diagnose them. The fact that you can do it at six months means, as a breeder, I can run the test on a young dog and make the decision to spay or neuter before sexual maturity. Because I do PennHIP at six months, and OFA can't tell if there's a problem unless arthritis is present 18 months later I cannot do OFA. I rather suspect the authors of this got caught up in the early PennHIP back-lash and/or got a crappy PennHIP score on on of their foundation dogs. Iris' poorest hip is .47 which is in the 100th percentile. To suggest that she needs a .30 is rediculous. She's so much tighter than the average Corgi that breeding her should lead to better hips.
Really, I've yet to see a good reason to use OFA in lieu of PennHip. As a veterinarian I've done both, and PennHip just offers so much more. It's a 3 view process, and the first view is the classic OFA shot. As a breeder I can evaluate a bitch at a young age and if her hips aren't up to my standards I can spay promptly and essentially eliminate the risk of mammary cancer. I can also evaluate what for many breeds is the primary limiting factor for many breeds prior to investing heavily in training and showing. My honest response to a kennel, especially of hunting dogs, which so blasts PennHip--which is the standard for NAVHDA--is that this is a kennel which is trying to hide something.
Reply to this
The September 2010 JAVMA has a study that settles it. OFA is wrong over half the time. The give Excellent ratings to dogs who are not excellent and less than excellent ratings to dogs who are excellent OVER HALF THE TIME. Flip a coin to determine if your OFA is worth the paper it is printed on. There is a 52% chance your OFA excellent dog isn't excellent, and an even greater chance that your good or fair dog is actually excellent.
Reply to this