A LEADING member of the Council of Docked Breeds has highlighted serious flaws in the ‘scientific evidence’ written up over twenty years ago in the so-called ‘Edinburgh Report’ (by vets Darke, Thrusfield & Aitken ‘Association Between Tail Injuries And Docking In Dogs’), being cited by MPs and anti-docking campaigners as a definitive seven-year study against the practice in their plans to ban the practice of tail docking in the UK.
Ron Henney told OUR DOGS: "I’m so tired of hearing this ‘evidence’ trotted out time and time again by anti-dockers including the British Veterinary Association, as well as the RSPCA and I wondered if any of them had actually analysed what was said in the original report. This ‘evidence’ was quoted by the Australian veterinary authorities in the evidence given to the Australian Government in its own plans to ban docking – and no-one has seriously questioned it."
Mr Henney contacted JRS Hales, a former biomedical Research Professor, Faculty of Medicine, University of New South Wales and asked him for permission to quote from his own in-depth analysis of the original study. Professor Hales gave his permission and now Mr Henney has written to the BVA’s official publication The Veterinary Record to set the record straight.
As yet, Mr Henney has received no indication as to whether his letter and Dr Hales’ report will be published in an upcoming issue, but OUR DOGS is pleased to reproduce Mr Henney’s letter to The Veterinary Record, together with Dr Hales’ analysis of the Edinburgh Report.
Letter to The Veterinary Record
Docking & The Edinburgh ‘Study’
The one and only piece of dog related ‘scientific evidence’, as opposed to mere opinion, that has been presented by the BVA in 36 years as part of the docking debate is that termed by the organisation As: "A seven year study at the University of Edinburgh Royal (Dick) School of Veterinary Studies" circa 1985. Which, in spite of the recommendation that "Vets quote it to MP’s", though few will have read and even less studied.
Firstly this was not a dedicated ‘seven year study’. Had it been so the researchers would have known which dogs visiting the clinic possessed tails. Instead they had to guess! It was an ‘analysis of data’ which concluded: "There is insufficient evidence at the 5% level to suggest a positive association between tail injuries and an undocked tail." The fact that at the time dogs correctly docked, to a greater or lesser extent, would not have had much of a tail to damage, does not seem to have been entered into the equation (Reminds me of the search for WMD’s).
Since publication of the paper I have tried to draw attention to what I believe to be its defects. Some of those defects, carefully worded, are admitted by the authors. I find it strange, given the profession considers this paper so important, that in the twenty years following its publication, it has failed to revisit and rectify those defects by carrying out a dedicated study.
Who am I a mere dog breeder concerned about the welfare of the breeds in which I have an interest to challenge the BVA? However the ‘ANALYSIS’ is now being used as gospel evidence in Australia. As a result a far better qualified critique has emerged. Maybe this academic criticism will prove more acceptable, convincing and worthy of consideration.
Please read and absorb Professor Hale’s critique, which is attached herewith. I hope the profession will find it interesting and informative.
Critique of the publication: Darke, Thrusfield & Aitken (1985), Association Between Tail Injuries And Docking In Dogs, Veterinary Record, 116: 409.
By Dr JRS. Hales B.Sc., M.Sc., PhD.
Protagonists of docking maintain that it reduces the risk of tail injuries, while antagonists argue that it is such a barbaric procedure that it cannot be justified on prophylactic grounds. In view of the fact that no quantitative data were available to support either claim, Darke et al said they would "investigate whether docking is associated with a reduced occurrence of damage to tails".
They did this by examining the database of clinical case record of the small animal practice teaching unit at the Univ. of Edinburgh. The study may be criticised from three aspects:
1. Strictly speaking, traditionally docked breeds are not directly comparable to non-docked breeds in terms of tail injury, because the principal original purpose of docking was to remove the tail from those breeds, which were originally found to be susceptible to tail injury. Therefore, the truly valid test of the prophylactic value of docking is to compare docked with undocked individuals within traditionally docked breeds. For example, it is pointless to compare tail injuries in German Shorthaired Pointers with Pointers because the much greater prevalence of injuries in the GSP (when originally being developed) is precisely the reason it began to be docked. There were & still are few injuries in the EP & therefore it does not require & has not ever required docking to protect it. This difference is presumably due to the lower tail-set in the EP.
Consequently, the Edinburgh data actually indicate a 15-times higher rate of tail injury in traditionally docked than in non-docked breeds. That is, the most valid comparison based on the data presented by Darke et al is to take the commonly assumed estimate that 5% of the pups of traditionally docked breeds are not docked, i.e., 5% of 2616 = 131. Eight injuries of 131 = 6% of these with tail injuries. The traditionally non-docked group had 39 of 9513 with injuries, = 0.4%, a 15-fold difference. In fact, the "0.4%" is the value employed by the Australian Veterinary Association supposedly to illustrate how few tail injuries there are & therefore how unnecessary docking is. What the AVA failed to point out is the fact that their data came from breeds which were not docked because they never needed to be docked.
2. The Edinburgh data did not specify whether or not each individual dog had been docked & therefore the Darke et al assumed there had been normal docking practice according to breed, i.e., all Boxers were assumed to have been docked whereas all Whippets were assumed to not be docked.
What bias could any misclassification cause & will it influence conclusions based on the data? The sole effect which any misclassification can have had on the "docked" data, is a negative one, i.e., a dog with tail injury which they assumed to have been docked (because of its breed) might not have been docked, leading to an over-estimate of the number of injuries in docked dogs and the conclusion that docking is not beneficial in terms of avoiding injury. Likewise, or conversely, the sole effect which any misclassification can have had on the "non-docked" data, is a positive one i.e., a dog which had been docked (despite being of a non-docked breed) did not have a tail which could be injured, leading to an under-estimate of the number of injuries in non-docked dogs and the conclusion that a full length tail is not more likely to be injured.
Thus, misclassifications will have narrowed the difference between injuries to docked & non-docked dogs & caused a bias towards the conclusion that docking is not an effective prophylactic measure. This begs the question: "Because, for example, a dog cannot develop testicular cancer if it has been castrated earlier in life, how can a dog’s tail be injured if it has been docked?" From this perspective, there is absolutely no need for such a study as that by Darke et al.
3.Notwithstanding the above insurmountable problems, it is finally noteworthy that although Darke et al found the "odds ratio" which they employed as a test of the strength of association between tail injuries & the presence of a full tail, to be 1.28, they concluded there was not a positive association between tail injuries & an undocked tail.
PLEASE NOTE: A hitherto unmentioned highly significant outcome from the data is the much greater number (a factor of x3.6) of non-docked dogs which presented for treatment of any kind.
If, as maintained by antagonists of docking, there are untoward side effects of docking (e.g., urinary incontinence, attacks by non-docked dogs because of communication difficulties, etc), a relatively greater number of visits to the vet by docked breeds must be expected. That is, from Jan. 1977 to Oct. 1984, the Small Animal Practice Teaching Unit at the Univ. of Edinburgh recorded 12,129 consultations, of which 9513 were dogs of traditionally non-docked & 2616 traditionally docked breeds.
Because around one-third of all breeds are docked, if there was no bias towards either non-docked or docked breeds, theoretical subdivision of the total would yield values of 8005 & 4003 (respectively). These data clearly indicate a distinct bias towards bad health in non-docked breeds & good health in docked breeds, however, we would not suggest that this is a reality, but that the bias may well be due to some other factor such as cross-breeds in the non-docked group. What we are undeniably left with, is the fact that the numbers of visits to the veterinarian shows that there is absolutely no bias towards there being more health problems in traditionally docked than in non-docked breeds.
J R S Hales, B.Sc., M.Sc., PhD,
(Retired) biomedical Research Professor, Faculty of Medicine, Univ. New South Wales; Visiting Professor, Faculty of Veterinary science, Sydney University; Chief Research Scientist, Commonwealth Scientific & Industrial Research Organization of Australia.
*Dr JRS. Hales may be contacted by e-mail at: Giantpaws@Bigfoot.com