THE QUALITY OF DISCUSSION SECTIONS IN REPORTS OF CONTROLLED TRIALS
PUBLISHED BY FIVE LEADING GENERAL MEDICAL JOURNALS
Michael Clarke1 and Iain Chalmers2
1Clinical Trial Service Unit, Harkness Building, Radcliffe Infirmary, Oxford OX 26HE, UK; 2Cochrane Centre, Oxford, UK
Objective: Several health-care journals have adopted the CONSORT recommendations to make it easier to assess the quality of controlled trials, reflecting the practical importance of these studies in guiding decisions in health care. Previous research has highlighted deficiencies in descriptions of the materials and methods used, and in the analysis and presentation of results. By contrast, the quality of discussion sections in trial reports has received little scrutiny, yet it is impossible to judge the relevance of a particular trial unless its results have been set in the context of all other relevant research. Ideally this should involve the presentation of an up-to-date systematic review, as was done in the 1986 report of the ISIS-1 trial.
Design: We assessed the Discussion sections in all 26 reports of randomized trials published during May 1997 in Annals of Internal Medicine, BMJ, JAMA, The Lancet, and the New England Journal of Medicine.
Results: In only 2 reports were the results of each new trial discussed in the context of an updated systematic review of earlier trials. In a further 4 reports, reference was made to relevant systematic reviews, but no attempt was made to integrate the results of each of the new studies in an updated review. One report appears genuinely to have been the first published trial to address the question studied. In the remaining 19 reports it does not appear that a systematic attempt had been made to set the new results in the context of previous trials.
Conclusions: These findings suggest that considerable scope remains for journals to help their readers to set the results of new trials in the context of preexisting evidence.
Return to Session Information
