Article type
Abstract
Background:The Cochrane Risk-of-Bias Tool is used for evaluating the risk of bias of randomised-controlled trials included in systematic review, which contains six domains (including 7 items): selection bias (random sequence generation and allocation concealment), performance bias, detection bias, attrition bias, reporting bias and other sources of bias. The criteria of the six former items are defined explicitly in Cochrane Handbook, but without those for 'Other bias'.
Objectives:Aim to investigate the criteria of 'Other bias' of the Cochrane Risk-of-Bias Tool for further assessment of risk of bias of randomised-controlled trials in systematic reviews.
Methods:We hand-searched the systematic reviews published from 1 January 2012 to 31 December 2014 in Cochrane Database of Systematic Review, Annals of internal medicine, Annals of Surgery, The Journal of the American Medical Association, The Lancet, British Medical Journal, of which the included randomised-controlled trials were assessed by the Cochrane Risk-of-Bias Tool. Two reviewers completed the hand-searching, screening and data extraction independently.
Results:340 systematic reviews were included, of which 250 (74%) were Cochrane reviews and 90 (26%) non-Cochrane reviews. 3342 primary studies included in 233 (69%) systematic reviews were estimated with 'Other bias', of those 723 (22%) were assessed as high risk, of which 538 (16%) were reported with the causes that were integrated into 67 causes. 55 causes (from 167 original studies included in 108 systematic reviews) were overlapped with the former six items such as random sequence generation, etc. In addition, the top cause of 'Other bias' was the potential conflicts of interest.
Conclusions:The causes of 'Other bias' are various in the current systematic reviews, but most of them are overlapped with the other six items including random-sequence generation, etc.
Objectives:Aim to investigate the criteria of 'Other bias' of the Cochrane Risk-of-Bias Tool for further assessment of risk of bias of randomised-controlled trials in systematic reviews.
Methods:We hand-searched the systematic reviews published from 1 January 2012 to 31 December 2014 in Cochrane Database of Systematic Review, Annals of internal medicine, Annals of Surgery, The Journal of the American Medical Association, The Lancet, British Medical Journal, of which the included randomised-controlled trials were assessed by the Cochrane Risk-of-Bias Tool. Two reviewers completed the hand-searching, screening and data extraction independently.
Results:340 systematic reviews were included, of which 250 (74%) were Cochrane reviews and 90 (26%) non-Cochrane reviews. 3342 primary studies included in 233 (69%) systematic reviews were estimated with 'Other bias', of those 723 (22%) were assessed as high risk, of which 538 (16%) were reported with the causes that were integrated into 67 causes. 55 causes (from 167 original studies included in 108 systematic reviews) were overlapped with the former six items such as random sequence generation, etc. In addition, the top cause of 'Other bias' was the potential conflicts of interest.
Conclusions:The causes of 'Other bias' are various in the current systematic reviews, but most of them are overlapped with the other six items including random-sequence generation, etc.