Article Text


Using the evidence
  1. David Mills


THIS series gives readers the opportunity to consider and contribute to discussion of some of the ethical dilemmas that can arise in veterinary practice. Each month, a case scenario is presented, followed by discussion of some of the issues involved.

In addition, a possible way forward is suggested; however, there is rarely a cut-and-dried answer in such cases, and readers may wish to suggest an alternative approach. This month's dilemma, ‘Using the evidence’, was submitted and is discussed by David Mills. Readers with comments to contribute are invited to send them as soon as possible, so that they can be considered for publication in the next issue. Discussion of the dilemma ‘Refusing to take part in euthanasia’, which was published in the May issue of In Practice, appears on page 311.

The series is being coordinated by Steven McCulloch, a practising vet with a PhD in the ethics of veterinary policy. It is hoped it will provide a framework that will help practitioners find solutions when facing similar dilemmas.

Statistics from

David Mills qualified from the University of Cambridge in 2008. After a period in mixed practice, he has worked at RSPCA hospitals for the last three years. He is nearing completion of the CertAVP in welfare, ethics and law. He is currently studying part time for a PhD in the ethics of veterinary interventions and evidence-based medicine.

Using the evidence

A lively, 60 kg dog presents to your clinic with an obvious anterior cruciate rupture that requires surgical repair. You are aware that published evidence indicates osteotomy procedures have a higher success rate in large dogs, but in your own self-audited, unpublished experience, a lateral fabellar suture performed by yourself carries as high a success rate. Your clinic does not offer osteotomy procedures but the owner can afford referral. What operation should you recommend to the owners of this dog?

Issues to consider

The key ethical problem here lies in the inherent tensions encountered in the application of evidence-based veterinary medicine (EBVM). EBVM is ‘the use of best relevant evidence in conjunction with clinical expertise to make the best possible decision . . . the circumstances of each patient, and the circumstances and values of the owner/carer, must also be considered’ (University of Nottingham 2016). According to an ‘ideal’ EBVM process, a clinician presented with a problem searches for the relevant literature, appraises its quality and applicability according to the hierarchy or pyramid of evidence (systematic reviews, meta-analyses, randomised controlled trials at the top, anecdote and experience at the bottom), uses this to make the ‘best’ decision, follows this treatment plan, and audits the outcome.

It all sounds eminently sensible and ethical: after all, who would not want to base their decisions on best evidence? Indeed, there appears to be an RCVS imperative for this: ‘in order to be considered fit-to-practise, veterinary practitioners hold the responsibility to ground their decisions on sound, objective and up-to-date evidence’ (RCVS 2013).

The problem in this scenario is that the clinician's ‘evidence’ – unpublished but self-audited – clashes with published evidence. The former lies at the bottom level of the hierarchy/pyramid, the published evidence is higher up (Christopher and others 2013, Gordon-Evans and others 2013). One cannot escape the conflict on applicability (the dogs in the study are representative of your patient) or patient/owner circumstances as there are no financial or logistical restrictions on care.

The interests of the patient, the veterinary surgeon and the owners also require consideration. The dog's interests lie in returning to pain-free function as quickly, and with as little suffering, as possible. Based on these criteria, there may be little difference between procedures for the dog. The veterinary surgeon may have financial interests in performing the surgery. Equally, following an EBVM ‘cookbook’ may disincentivise the veterinary surgeon to audit other outcomes, and referring future cases may lead to deskilling to the detriment of other patients. The owners may have interests in minimising cost, and therefore may be attracted by the in-house procedure; they may not have the interest or the capacity to understand or appraise the evidence.

Possible way forward

On a strict reading of EBVM, the choice seems obvious: one should refer for an osteotomy procedure. But is this ethically the right thing to do? Whether it is or not depends on our conception of ‘evidence’, and the relative strengths of different forms of evidence. The published evidence draws its strengths from a sense of objectivity. Its methods and conclusions can be scrutinised, and we can appraise its quality and applicability. It offers (ideally) larger sample sizes than can our own experience, from which strong conclusions can be drawn by applying population data to the individual.

However, it can also be argued that the ‘objectivity’ afforded by publication elevates this form of evidence too highly. What is to say, epistemologically, that published evidence trumps unpublished evidence? Statistics are usually invoked, but the medical sciences use a frequentist approach that simply tells us whether a null hypothesis can be rejected, not which alternative hypothesis is to be believed – therefore choosing the ‘best’ approach would entail a very large number of studies in a process of exclusion to identify the last standing, true hypothesis (Lindley 2000). Indeed, we can justifiably ask what objectivity means, and actually how useful it is to a discipline such as medicine, where subjective aspects such as pain, emotion, suffering are beyond objectivity's reach (Chalmers 2013).

There appears to be no a priori reason to reject the individual's carefully and faithfully recorded case series in favour of published studies, despite EBVM theory endowing the latter with ‘higher level’ evidence. To do so is to ride roughshod over clinical expertise, which arguably in veterinary medicine should take precedent over external evidence where the latter continues to be of questionable applicability (referral centre studies applied to general practice), low power, and low volume.

Changing our conception of EBVM, or modifying it to a repository of external evidence to be referenced if required rather than obligatorily may help to clear the ethical waters in decision making until such time the evidence base can match its lofty ambitions. In my view, the surgeon should operate in-house.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.