What's the evidence for the EBM evidence hierarchy?
The EBM evidence hierarchy has become very familiar over the last few years. It looks something like this:
1a Evidence from a meta-analysis of RCTs
1b Evidence from at least one RCT
2a Evidence from at least one controlled study without randomization
2b Evidence from at least one other quasi-experimental study
3 Evidence from non-experimental descriptive studies, such as comparative studies, correlation studies and case-control studies
4 Evidence from expert committee reports, or opinions and/or clinical experience of respected authorities.
This is a radical list. For 2000 years, respect for authority – whether for Aristotle or Galen – lay at the heart of medical care or healthcare. But in this list, whilst it is still on the list, it comes bottom. The aim of this session is to think about why the list works as it does and what evidence or argument could be used to support it.
Suppose one were to test the ordering by examining the success of different approaches to care based on different levels: comparing a health service which insisted on evidence from a meta-analysis of RCTs with one which relied on evidence from non-experimental descriptive studies, such as comparative studies, correlation studies and case-control studies. How would such a comparison itself be ranked? Suppose that other studies based on different approaches within the list produced different results, how, without begging the question, could one decide which was the more reliable study?
On the other hand, it may be possible to work out from first principles how it is that rising up through the list removes particular kinds of bias or error. If so, what are the limits of that approach?
Reading the extract from
Thornton, T. (2006) Essential Philosophy of Psychiatry, Oxford: Oxford University Press