Evidence-based medicine (EBM) aims to assess the strength of proof behind medical interventions in terms of risks and benefits, and therefore can be used to inform clinical decision making on both an individual and a population basis. As such, EBM is crucial in maintaining quality medical care and ensuring good clinical outcomes. Many parties are involved in EBM. Firstly, researchers and publishers are involved in the conduct and dissemination of medical evidence. Then, policy makers and clinicians are responsible for the eventual implementation of changes in healthcare decisions that may occur.
This week, BMC Medicine attended the Evidence Live 2013 conference in Oxford, held on the 25th and 26th of March. Attracting around 1000 delegates, the conference provided stimulating debate over the current status and future directions of EBM and evidence-based practice. The topic that provoked substantial debate was regarding breast cancer screening. Peter Gøtzsche from the Nordic Cochrane Centre provided convincing evidence against the use of mammography. However, it seems that the evidence is still far from clear – this debate has been the subject of several articles in BMC Medicine this year, which were highlighted in a recent blog.
EBM in low income settings was also discussed, which can sometimes be problematic due to lack of resources. However, Kath Maitland gave an excellent example of EBM in Kenya. She discussed the results from the Fluid Expansion as Supportive Therapy (FEAST) study, where African children with severe febrile illnesses given the intervention of rapid fluid bolus exhibited excess mortality compared with controls. This was exactly the opposite result to what was expected. She presented the initial trial, and also the secondary analysis of this data, which was recently published in BMC Medicine. Here, authors explored the mechanisms behind the excess mortality, finding that it was due to cardiovascular collapse rather than fluid overload.
Another hot topic in EBM touched upon publication, and publication bias. Ben Goldacre highlighted the importance of publishing negative results. Unfortunately, many authors, and many journals, are not keen to publish negative trial results as these are deemed to be not as exciting or citeable. As a consequence, there is a bias caused by over-representation of positive effects of a particular intervention, which could be potentially harmful when applied to clinical decision making based on this ”evidence”. In an attempt to prevent this, he advocates that all trials should be both published and registered, which is the basis of the AllTrials campaign.
Of course, individuals are only able to make informed decisions based on appropriate evidence if they are able to access the data in the first place. This may be a problem if that data is behind a subscription wall. In an exciting session on the latest developments in publishing, the benefits of open access (OA), as well as open peer review, were discussed by Ginny Barbour and Trish Groves. As part of BioMed Central, which is the founder of OA, BMC Medicine is a strong supporter of this mode of publishing, promoting “knowledge for all”.
Evidence Live prompted much needed debate in the field of EBM within the key stakeholder groups. It is hoped that continued discussion among the different disciplines involved will further improve EBM and, consequently, healthcare.