 |
 |
Rod: you are right. However categorial hierarchies are no longer used. GRADE allows for up- and down-grading, depending on the quality of evidence, and the OCEBM Levels of Evidence<https://www.cebm.net/2016/05/ocebm-levels-of-evidence/> went as far as to avoid the term ‘hierarchy’ altogether (which was on purpose). Somehow, the idea that hierarchies persist, persists. And, to be fair, EBM is often practiced as if there were strict hierarchies. Still, it is the perception that categorical hierarchies exist, rather than the need to ‘do away with [straw man] hierarchies’. One way to move beyond hierarchies is to reconsider Bradford Hill<http://journals.sagepub.com/doi/10.1258/jrsm.2009.090020>. Warm regards, Jeremy From: "[log in to unmask]<mailto:[log in to unmask]>" <[log in to unmask]<mailto:[log in to unmask]>> on behalf of Rod Jackson <[log in to unmask]<mailto:[log in to unmask]>> Reply-To: Rod Jackson <[log in to unmask]<mailto:[log in to unmask]>> Date: Saturday, 22 September 2018 at 03:41 To: "[log in to unmask]<mailto:[log in to unmask]>" <[log in to unmask]<mailto:[log in to unmask]>> Subject: Re: Removing. SR from the evidence hierarchy Hi Craig. In my opinion, it’s time to get rid of all evidence hierarchies. While they had a role early on in the development of EBM, they are simplistic and often wrong. For example it is impossible to undertake a good long term RCT of, say, high versus low dietary saturated fat with CHD as the outcome. There is so much noncompliance/crossover/contamination over a period of, say 5 years, that its completely unsurprising that the trials either show no effect or very small effects. It’s hard enough to get even modest compliance in a long term double blind placebo controlled trial of a low dose drug without meaningful side effects. Why researchers have believed they could do a long term dietary trial is difficult to understand, although many have tried. Long term cohort studies are even worse (eg PURE, just published in the Lancet) because as well as having the same problem I mention above for RCTs, they suffer badly from misclassification of exposure (most people are unable to tell you what they usually eat with any degree of accuracy) and they are seriously confounded (middle class health conscious people like me eat a Mediterranean diet). So the two theoretically highest quality types of studies for determining the effects of interventions should be near the bottom of any evidence hierarchy for the question ‘does a diet high in saturated fat cause CHD?’ along with any SRs of such studies. In many situations RCTs (and SRs of high quality RCTs) are the most appropriate type of study to answer a question, in others cohort studies are the most appropriate, and in others economic or qualitative studies may be the most appropriate. It depends completely on the question. I believe we now do our students, our colleagues and the public far more harm than good by promoting evidence hierarchies. My friend and colleague Professor Nick Wald uses the analogy of a carpenter’s tool belt. Carpenters don’t have a hierarchy of tools. If they want to nail two pieces of wood together, they use a hammer, but if they want to cut a piece of wood in two, they use a saw. Regards Rod Jackson Professor of Epidemiology University of Auckland New Zealand On 22/09/2018, at 9:37 AM, Craig Lockwood <[log in to unmask]<mailto:[log in to unmask]>> wrote: Dear All, A concept paper came out recently on hierarchies of evidence which repositioned systematic reviews..... https://ebm.bmj.com/content/21/4/125 They highlight how quality issues can be represented then propose that Sr be removed from the hierarchy and instead be used as the lens through which other study designs in the pyramid are considered. I think this repositions evidence within a very narrow band, it seems to argue that the primary utility of Sr is as a tool to critique other study designs and I’m not sure if others on this list have issue with this but I also note the proposed pyramid is limited to a narrow range of quantitative studies. Where is the qualitative evidence? Where is economics etc? I recall seeing variations of the evidence pyramid that had clinical decision support systems at the top and other variations which were multi sided and had other types of evidence on the other sides. Conceptually This one seems too narrow to be useful. https://ebm.bmj.com/content/21/4/125 Kind regards, Craig Director implementation Science, Joanna Briggs Institute ________________________________ To unsubscribe from the EVIDENCE-BASED-HEALTH list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=EVIDENCE-BASED-HEALTH&A=1 ________________________________ To unsubscribe from the EVIDENCE-BASED-HEALTH list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=EVIDENCE-BASED-HEALTH&A=1 ######################################################################## To unsubscribe from the EVIDENCE-BASED-HEALTH list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=EVIDENCE-BASED-HEALTH&A=1