Research published today in BMC Veterinary Research has added to a growing body of literature that suggests the evidence-base in some areas of veterinary science is still poor, or even lacking. In this guest post, researchers at the Royal Veterinary College (RVC) Canine Epilepsy Clinic describe some of the issues they encountered when conducting the first systematic review on the efficacy of antiepileptic drugs (AED) in the field of canine epilepsy, and discuss the need for trials that provide high quality evidence to achieve more reliable and objective results.
Epilepsy is not a specific disease but a chronic condition characterized by recurrent seizures, and is the most common chronic neurological condition found in dogs and humans. It affects around 50,000 canines in the UK and approximately 600,000 people. Various antiepileptic drugs (AEDs) are used for the management of idiopathic epilepsy (IE) in dogs, however, clinical information on the grounds of their efficacy still remains limited. Professor Holger Volk, Clinical Director of the RVC Small Animal Referral Hospital and Section Editor for BMC Veterinary Research added “Canine epilepsy is a complex condition and can be very distressing for the dog and their owners, therefore it is immensely important the drug treatments we give to our pets have been properly tested in reliable studies”. In light of this need, the authors performed a systematic review with the simple aim to describe the efficacy of AEDs used for the management of IE in dogs.
However, co-author Marios Charalambous discussed how much of the available evidence for the medical treatment of canine epilepsy was based on subpar testing. Only four blinded randomized clinical trials (bRCTs) were reported which were considered to offer the highest quality of evidence amongst all the studies evaluated. The majority of the studies were non-blinded non-randomized uncontrolled trials and case series with a relatively low number of animals, short investigation period and not always clear diagnostic criteria for including dogs with IE.
Co-Author Dr David Brodbelt, Senior Lecturer in Companion Animal Epidemiology, added that on the outcomes of study demonstrated that “many studies revealed a potential efficacy of a drug, but then there is a lack of follow-up studies of higher quality confirming the initial positive results”. For these reasons, the authors were unable to provide a definitive recommendation for treatment.
But what would be the reason for poor study design and reporting in these publications?
The answer to this is not easy and is multidimensional. Examples for poor study design and reporting are problems with recruitment, participation of small number of affected animals, limited or unclear diagnostic procedures, potential inadequate length of study period, costs and very often insufficient funding to conduct a high quality study, training of principal investigators and collaborators, time commitment, ethical restrictions, insufficient cooperation of the patients (animals) and/or their owners, publication bias and pressures associated with academic productivity. These are issues that are commonplace in other areas of veterinary research.
What are the future recommendations to help overcome these issues?
There are many potential recommendations one could consider to help reduce poor quality evidence-based research in veterinary medicine. The authors put forward some suggestions, focusing on the “roots” of the issue that fundamentally draws upon a closer, more transparent relationship that extends from the funders through to the publishers.
- It is important for veterinary scientists to work together with funding bodies and all major stakeholders to help improve funding structure and training with an aim to reduce subsequent publication bias and pressures as much as possible.
- International task forces, such as the International League Against Epilepsy (ILAE), need to be encouraged as they help to link experts in the field of veterinary medicine, with a goal to advance and disseminate scientific knowledge.
- It is important to teach authors of the importance of publication standards, such as the ARRIVE and CONSORT guidelines, as these standards help the author to improve the quality of their reporting from the offset, and also allows the reader to evaluate the strength of the study.
- It would also help to publish study protocols in veterinary peer-reviewed journals prior to starting the trial. This way, readers can be confident that the protocol has been critically reviewed prior to commencement and investigators are aware of potential weaknesses and limitations that can be avoided in subsequent trials.
- One of the issues when conducting this current systematic review was the lack of supporting data, which if available, helps to make scientific evaluation and reporting more transparent and trustworthy. The amount of useful data that is lost due to insufficient or no reporting is considerable. Therefore, scientists should make their supporting data open and available to the public, and publishers should work towards making it mandatory for authors to submit their raw data before publication.
- The importance of registering clinical trials in human medicine has long been established. The All Trials campaign, which is now supported by thousands of individual patients, clinicians and researchers across the world and by hundreds of organizations representing millions of people, claims that every investigator running a clinical trial should register it and report its results. This process should also be mandatory for veterinary scientists to help improve transparency in reporting.
Let us know your opinion on these matters in the comments box below.
BMC Veterinary Research Executive Editor,
Dr Hayley Henderson
Further information about the authors and the study can also be found at the RCV Press Office site