Neil Chanchlani is a CMAJ Associate Editor and clinical research fellow at the University of Exeter, UK. He recently attended the 8th EBMLive conference in Oxford, England.
All healthcare practitioners are encouraged to make decisions that are based on strong evidence. But often we don’t – sometimes because the evidence is poor and conflicting, other times because we are ignorant and unaware and, rarely, because we aren’t comfortable with updating our practice. So we need to keep Evidence-Based Medicine (EBM) on the frontline and not on the (academic) shelves - we need to remind clinicians, researchers, and patients that decisions should be based on the best possible data. What those data look like, which research practices are used to interrogate them, and how results gets disseminated are the questions that EBMLive tries to answer.
This year, the focus of EBMLive was on updating and improving the EBM Manifesto, which aims to take forward and implement better solutions for better evidence and healthcare. This is a well-intentioned effort, but difficult to enact. The relatively small group of attendees - 150 to 200 people - are highly motivated and interested. EBMLive is a conference about generating ideas and conversations about improving the state of medical research – for example, how to reduce research waste, or understanding how conflicts of interest can impact research outcomes. Keynote speakers this year were EBM celebrities Ben Goldacre, Trish Greenhalgh, and John Ioannidis.
Arguably, the UK is lagging behind Canada when it comes to progressing the EBM movement. The Knowledge Translation (KT) Program at St. Michael’s Institute is a trailblazer, and KT Program scientist Dr. Andrea Tricco is leading the national Canadian-wide Strategy for Patient-Orientated Research Evidence Alliance. She described how the program encourages patients to be involved in all aspects of research to ensure questions and results are relevant. In addition, the program is committed to prioritizing the needs of Indigenous peoples in Canada, by recruiting researchers to actively consult with communities to find out the research questions that Indigenous peoples would like to have answered.
Find out more at https://sporevidencealliance.ca/
EBM is not always about conducting new research to answer novel questions; it can be about identifying poorly conducted research. We must acknowledge that some research practice is poor, be transparent about biases, and demonstrate honesty about the limitations of study conduct in interpreting results, says Sir Terence Stephenson, professor of child health at UCL Great Ormond Street Institute of Child Health. He suggested that rewarding peer review, improving resources for tracking patients through observational studies and trials, and blockchain technology are all possible solutions to reducing questionable research practices.
A summary of Sir Terence's talk can be found here.
Professor Will Dixon, Chair in Digital Epidemiology at the University of Manchester, spoke about how data from a national patient drug safety registry has been used to improve clinical outcomes and reduce adverse events for patients. However, there are challenges when collecting registry data from multiple hospitals, nationally and internationally, such as being careful of how much to ask users to fill out, quality control of the data coming in, and potential for selection and attrition biases. Prof. Dixon commented that social media platforms, such as Twitter and Facebook, offer patients the opportunities to share their experiences with each other and healthcare staff, and can help aid clinicians’ and researchers’ understanding of patients' journeys.
Nick DeVito discussed the promising work coming out of the dedicated team at the Oxford EBM Data Lab. The group aims to improve trial transparency by looking at how clinical trials are conducted and reported. Common problems that occur include not reporting primary outcomes and switching pre-specified outcomes to ones that have more favourable results after data analysis has occurred.
His top tips for readers, peer reviewers and editors to help misreporting of trials and observational studies? 1) Check the study was registered prospectively 2) Check outcomes match what was pre-specified in the protocol, and if they do not, question why not. 3) Use reporting guidelines, such as the CONSORT and STROBE checklists, when critically appraising an article.
“Language is everything when a doctor talks to a patient”, says Dr. Richard Stevens, associate professor of medical statistics at the University of Oxford. Patients do not have the same health literacy as doctors and we need to better understand how to bridge the gap with current medical nomenclature and the vocabulary we use in day-to-day clinical practice. His team became particularly interested in the term ‘chronic kidney disease’ and how that can cause confusion amongst patients and affect the relationship they have with their doctor.
Last year, he published an analysis article in the CMAJ advocating for the use of the term ‘Kidney age, not kidney disease’ when communicating with patients. He has built on this work by piloting a kidney communication tool in the form of a leaflet and website. His team has gained interest from UK general practitioners to pilot their tool. Next, they will be carrying out a study to ask patients whether they prefer to be told their ‘kidney age’ or stick to traditional terminology of ‘stage of kidney disease’.
The principles of EBM are not just applicable to medicine, but other disciplines too. This year, the conference had speakers and attendees from veterinary medicine, including a session on how to adapt the EBM manifesto for animal welfare. Suzanne Jarvis, managing editor of BMJ veterinary medicine journals, and Dr. Rachel Dean, Director of the Centre for Evidence Based Veterinary Medicine at the University of Nottingham, UK, highlighted the lessons that can be learned from EBM and applied to their practice, both academically and clinically. Important differences include the language used across disciplines and the differences in industry set-up in the UK with the public health care service versus the largely private service of veterinary medicine. Is there a need for a veterinary-medicine based manifesto? Yes, they argue.
EBM has come a long way, but there is still progress to be made. There are many reasons why we need better evidence, but the solutions required are more challenging to devise and enforce, particularly on the front line where clinical decisions are made.
To join the discussion, consider attending EBMLive next year in Toronto, Ontario from July 6 – 8th, 2020.