Tuesday, September 29, 2015

Critical appraisal: not just seeing what you agree with

It's the Labour Party Conference this week, and so on the Andrew Marr Show on the BBC on Sunday last, Jeremy Corbyn, party leader, was interviewed.   I was pleased to see some of the interview, and remarked to the family how good it was and how much sense I thought was in what Mr Corbyn said (see declaration of interest below!).   Ah, well, said someone close to me, you would think that, because you agree with him.

This set me thinking.  Did I have a biased view of what he said because I only saw those things that I agreed with?   Did I have a biased view because I overlooked things that I did not agree with?   And that set me thinking about critical appraisal.

Reading the whole paper

One reason for critically appraising a research article is to see if it applies to your research or clinical practice.   Is another to make sure that you don't see only the things in it that you agree with?   I do think (and have said so to students) that critically appraising a paper is a way to make sure that you read the whole thing, and not just the bits that you find easiest.  For me, the statistical parts of a paper are more difficult, and are likely to be the bits I skip over.   If I am reading with a checklist like CASP, then I am made to read those bits, as I can't complete the checklist if I don't.   Does a checklist help you see all things in a paper and not just those things you agree with?

Which papers to read?

Is there a danger that when selecting papers from a list of search results, that I will still choose those items that I agree with, whose titles reinforce things I already "know" or have decided?

Are there criteria to use to select references from a list to read in full?   If you are doing a systematic review, then you look for things that match your inclusion criteria.  If you are looking for high quality evidence to inform patient care or clinical practice, then you critically appraise the evidence.  But if your search finds 50, how do you choose which to select for that detailed critical appraisal?   I have talked in teaching sessions about looking at titles, and then abstracts, to decide which to look at in full, but what are you looking for when you review titles and abstracts?   You are comparing them with your answerable question, the search question you formulated using PICO or a similar scheme.  Do you think they may contain an answer to that question?

Declaration of interest: as you might be suspecting by now, I am a party member (I joined a year ago or so) and I did vote for Jeremy Corbyn. 

No comments: