A descriptive feast but an evaluative famine - A systematic review of published papers on primary care computing 1980-1997

Report from the BMJ’s Information in Practice editorial committee

Members of the committee were:
Douglas Carnall
Jeremy Wyatt
Nick Booth

Decision: Revise to author

1 The editorial committee felt this was a valuable update to work previously published in the BMJ providing the following criticisms can be addressed.

2 Search criteria: The number of studies found and found eligible should be more clearly expressed in the abstract. It would be more appropriate to use the figure of 1892 abstracts in the abstract. Why was CINAHL omitted from the initial search?

3 We discussed whether it was justifiable to call this a systematic review: it does not synthesize data, and the inclusion criteria for studies were not always explicit. Please consider the QUOROM guidelines on systematic reviews (attached) and ensure that your paper conforms to them, or justify why this is not appropriate. It may be more appropriate to title your paper as a review (with methods section, which we encourage in every case anyway).

4 The paper claims that 5 papers show that computerisation increases consultation length. Was this only measured in 5 studies? Or did 10 studies find computers shortened consultation length. Was time measured objectively? There is evidence that use of a computer subjectively lengthens the perception of time. In other words, were clear prior questions framed before resorting to the literature, and if so could this be reflected in the paper?

5 The diversity of applications for which the computer is used in primary care presents a very real difficulty for the authors. Re-organising the paper by application--e.g. mail-merging, clinical coding--would make the paper more readable.

                 6 We will publish the extended table of references on the web. 7 The last points relate to our processes: a We are now processing all manuscripts electronically, so please could you provide us with a copy of your article on disk as well as in hard copy. Please see the enclosed guidance about our preferred formats. If you cannot provide one of these please send a disk anyway with a note of the software used.

b It would help us greatly if you would send with your revised paper a covering letter explaining how you have responded to all the points raised in this report.

c The BMJ aims to include the design of each study in the title; please amend your title accordingly.

d Before this report leaves our offices your paper will be checked through by one of our technical editors. He or she may enclose some checklists and guidance notes to help you revise the paper into BMJ style; please respond to these if you decide to revise your paper.