The BMJ has been supportive of evidence-based medicine, one aspect of
which is the assiduous searching for and citing of clinical data. When
ideas are discussed there should be similar acknowledgement of what has
gone before. In the editorial on structured discussions [1], the
conventional IMRaD structure is credited to a book published in 1996,
coincidentally published by the BMJ publishing group. I do not know who
invented the acronym, but I have three medical style manuals in front of
me [1-3] published between 1985 and 1991: all of them describe the
conventional structure and two of them have IMRAD in the index. Is a BMJ
editorial a fit place for what is in fact an advertisement for one of its
own books?
The discussion structure put forward as "Our proposal..." also
appears in all the style manuals. As the editorialists judged references
were needed, it would have been polite to have cited an antecedent: "We
propose a structure similar to Huth's be used…". I suggest Huth because
when he warns of stealing others' credit by burying their previous similar
findings deep in the discussion, he cites a BMJ editorial in support.[5]
On the more general point: I disagree that the case for structuring
discussions is the same as structuring abstracts. The discussion is the
only part of the paper that lets readers anywhere near the minds of the
investigators. Attempts to constrain investigators denies the bias and
uncertainty of human discourse, and seems to me based on the pretence that
facts are value-free. Better to give advice to writers on how to write
well; warn readers, as many clearly already know, that "All that
glisters..."; and let the investigators take their chances with electronic
rapid responses.
1 Docherty M, :Smith R. The case for structuring the discussion of
scientific papers. BMJ 1999;318:1224-5.
2 Huth EJ. How to write and publish papers in the medical sciences, 2nd
ed. Baltimore: Williams & Wilkins, 1990. pp 67-8.
Rapid Response:
An old idea, but is it a good one?
Editor,
The BMJ has been supportive of evidence-based medicine, one aspect of
which is the assiduous searching for and citing of clinical data. When
ideas are discussed there should be similar acknowledgement of what has
gone before. In the editorial on structured discussions [1], the
conventional IMRaD structure is credited to a book published in 1996,
coincidentally published by the BMJ publishing group. I do not know who
invented the acronym, but I have three medical style manuals in front of
me [1-3] published between 1985 and 1991: all of them describe the
conventional structure and two of them have IMRAD in the index. Is a BMJ
editorial a fit place for what is in fact an advertisement for one of its
own books?
The discussion structure put forward as "Our proposal..." also
appears in all the style manuals. As the editorialists judged references
were needed, it would have been polite to have cited an antecedent: "We
propose a structure similar to Huth's be used…". I suggest Huth because
when he warns of stealing others' credit by burying their previous similar
findings deep in the discussion, he cites a BMJ editorial in support.[5]
On the more general point: I disagree that the case for structuring
discussions is the same as structuring abstracts. The discussion is the
only part of the paper that lets readers anywhere near the minds of the
investigators. Attempts to constrain investigators denies the bias and
uncertainty of human discourse, and seems to me based on the pretence that
facts are value-free. Better to give advice to writers on how to write
well; warn readers, as many clearly already know, that "All that
glisters..."; and let the investigators take their chances with electronic
rapid responses.
1 Docherty M, :Smith R. The case for structuring the discussion of
scientific papers. BMJ 1999;318:1224-5.
2 Huth EJ. How to write and publish papers in the medical sciences, 2nd
ed. Baltimore: Williams & Wilkins, 1990. pp 67-8.
3 Farr AD. Science writing for beginners. Oxford: Blackwell, 1985. pp 33-
5.
4 O'Connor M. Writing successfully in science. London: Chapman and
Hall, 1991. pp 66-7.
5 Anon. Reference 13. BMJ 1985;291:1746.
Competing interests: No competing interests