Improving EBM’s Hierarchy of Evidence

Our third EBM+ video features Professor Jon Williamson – PI for the EBM+ project – discussing the downsides of the EBM pyramid (pun intended).


Is the pyramid hierarchy of evidence the best model to use when making medical decisions? At EBM+, we think there is room for improvement.



This hierarchical approach of EBM is generally used to provide the ‘best’ available evidence. However, we know that such evidence could just be the ‘best of a bad bunch’, or that ‘good quality’ evidence can be mishandled. In the video, Jon uses a well-known Streptomycin trail to demonstrate how results can be misinterpreted when judging a drug’s long-term effectiveness. Therefore, in order to sufficiently weigh up short-term improvement and long-term survival we need to consider all the best available ‘types’ of evidence.


What’s the point in looking for the ‘best’ evidence of only one type – e.g. statistics gained from trials – to support our medical decisions, when such evidence only provides part of the story? When put like that, it sounds like current EBM mantras might be doing lots of researchers and patients a real disservice.


This is where EBM+ comes in. By combing evidence of mechanisms with correlations from clinical trials, we can reinforce the ‘best’ current evidence by understanding, for instance, how a drug works as well as knowing its correlative effect – i.e. what happens when a patient takes it. Jon also wrote an earlier blog using this analogy of reinforced concrete – Penguins and Causality (I’d read it just for the name).


Moreover, the reason we call this a pluralist approach is because we welcome different types of evidence when it comes to medical decision-making. And yes, we know in some cases both types of evidence will not be readily available, but this shouldn’t discourage us to actively seek out different evidence types – correlations AND mechanisms – in order to further legitimise patient treatments and seek more reliable outcomes.


Overall, different evidence types can help us different ways and “it’s only by using trials and mechanisms together that we can improve our clinical decision-making”.


Like what you see? What to know more? Get in touch and let us know what you think.



Leave a Reply

Your email address will not be published. Required fields are marked *