Rapid Responses to:

PAPERS: Peter C Gøtzsche and Helle Krogh Johansen Meta-analysis of short term low dose prednisolone versus placebo and non-steroidal anti-inflammatory drugs in rheumatoid arthritis BMJ 1998; 316: 811-818 [Abstract] [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

Has there been a labelling problem ?

John Higgie   (29 January 2001)

Has there been a labelling problem ? 29 January 2001
John Higgie, GP Harley House Surgery Send response to journal: Re: Has there been a labelling problem ?	Revisiting this article I noticed that the labelling at the bases of the scattergram is correct for the first two sets of data, but not for the grip strength data, where the rating scale works in reverse. For pain and joint swelling less is better, for grip strength more is better, so the caption at the bottom "steroids better...NSAID better" should be the other way round. Apologies if you have published a correction or explanation, but I don't keep my backcopies because your website is so good. Regards. John Higgie