Ancient Egypt and the ‘hierarchy of evidence’

30 Jun

by Jonathan Stokes, PhD student in Multimorbidity theme


One of the first things you learn when training in any sort of biomedical research is the ‘hierarchy of evidence’ (picture above), what counts as the ‘best’ type of evidence in our field. As you can see from this pyramid, randomised controlled trials (RCTs) – where a population is randomly split into two or more groups, an intervention is carried out on one, and the other acts as a ‘control’; results look at the difference between the two groups after a period of time – are at the very top (not counting ‘Systematic reviews’ which are simply compilations of many studies from the lower points of the pyramid). RCTs work so well, and are rated so highly in the hierarchy, because by definition and set-up they control for as much variation as possible. They control for variation both between the two groups involved, as well as between the settings they are subjected to. This is great from a statistical point of view, but stripping context and the variation naturally found in any population between individuals gives us extremely unnatural results i.e. the interventions assessed in this way don’t work in the same way when these are used in the much more complex, real world. And this is especially true when we look at more and more complex interventions e.g. integrated care, for more and more complex patients with multimorbidity – you can find more details on this specific issue in my personal blog page here. Not only do RCTs apply only incompletely to the actual context they’ll be used in, but they also cost an absolute fortune, and take years to do properly. Not always the most practical solution perhaps, particularly in assessing an intervention which has particularly low risk of harms for instance. This lack of real-world applicability can be a real problem, especially when working in a ‘translational’ research centre, where we try to focus on crossing the ‘translational gaps’ between evidence and actual practice. Surely, we want to create evidence here which applies and can be used in the ‘real world’. For this reason, I’d argue that the age old ‘hierarchy’ of evidence in biomedical research is as outdated as its architectural equivalent of the ancient Egyptians. By no means am I arguing not to use RCTs under any circumstances by saying this! RCTs are a great source of evidence in the appropriate circumstances, and for example, I wouldn’t want to be taking any medicine that hadn’t been thoroughly tested with one. But, we need to move away from the simplistic attitude that our evidence can sit neatly in a hierarchy. For starters, look how low down experiential, person-centred, qualitative evidence sits. This is an important area of understanding for the type of care we want to provide to patients! We have a new ‘multimorbidity’ disease paradigm. Our health system needs to adapt to this, and so does our research. Particularly in a time where budgets are being squeezed, if we can save money on a completely inappropriate RCT here and there, for instance, and can instead put some of the abundant routine data we have lying around to good use, we shouldn’t be afraid to do so because of some ancient paradigm.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: