Room for improvement? A survey of the methods used in systematic reviews of adverse effects

Ref ID	432
First Author	S. Golder
Journal	BMC MEDICAL RESEARCH METHODOLOGY
Year Of Publishing	2006
URL	https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1402311/
Keywords	• Harms • Sponsorship bias • Risk of bias • Low reporting quality
Problem(s)	• Reliance on randomised controlled trials for harms / safety data • Search strategy not provided • Conflict of interest statement or disclosures for review authors missing • Financial conflicts of interest of review authors • No quality assessment undertaken or reported
Number of systematic reviews included	256
Summary of Findings	48% of included reviews provided information on the source of funding. Very few (5%) provided sufficient information to reproduce the literature search. Of the reviews that did report financial support 85 were independent sources, 32 could represent a conflict of interest (largely a drug manufacturer), and 6 said they received no financial support. 63% of reviews sought to include studies that compared the intervention with a control, and 28% of reviews limited themselves to data from randomised controlled trials. Less than half (41%) of the reviews specified assessment of the quality of the included studies in the methods section, or alluded to it by describing in the methods data extraction that included indicators of study quality.
Did the article find that the problem(s) led to qualitative changes in interpretation of the results?	N/A
Are the methods of the article described in enough detail to replicate the study?	Yes