Article type
Year
Abstract
Background: in recent years, quality of evidence assessment of Cochrane Reviews has been advocated in order to increase the degree of certainty and confidence about the clinical outcome results presented from the analyzed studies. The GRADE approach considers the limitations in study designs, inconsistency, indirectness of evidence, imprecision of results and risk of publication bias. The evaluation of each outcome comparison is presented in a 'Summary of findings' table, providing readers (clinicians or consumers) with a more transparent and objective presentation on the main results.
Objectives: to analyze the quality of evidence presented in the 'Summary of findings' tables of Cochrane Review main comparisons.
Methods: we randomly screened published review data between 2017 and 2018. We excluded Cochrane Reviews of methodology, diagnostic interventions, overviews, prognosis, network meta-analysis and withdrawn reviews. Two authors extracted GRADE data related to the quality of evidence in each review from the 'Summary of findings' table for the main comparisons, as well as the reasons why the evidence quality was downgraded.
Results: we included 280 reviews from 50 different Cochrane Groups, and we excluded 22 reviews due to exclusion criteria. Twelve reviews did not present data in the 'Summary of findings' table, and so we included 246 reviews and 1254 outcomes in the analysis. With regard to the level of evidence, we identified 88 outcomes that had been assessed as high quality (7%), 328 as moderate quality (26%), 466 as low quality (37%), and 372 as very low quality (30%). The most frequent reasons for downgrading the level of quality of evidence were limitations in study designs and imprecision of results.
Conclusions: the quality of evidence for the main comparisons in Cochrane Reviews is predominantly low or very low.
Patient or healthcare consumer involvement: future primary studies need improved methodological design and greater precision to achieve results that will improve quality. Interventions that are investigated in systematic reviews of high quality can provide safer and more consistent information to consumers and decision-makers.
Objectives: to analyze the quality of evidence presented in the 'Summary of findings' tables of Cochrane Review main comparisons.
Methods: we randomly screened published review data between 2017 and 2018. We excluded Cochrane Reviews of methodology, diagnostic interventions, overviews, prognosis, network meta-analysis and withdrawn reviews. Two authors extracted GRADE data related to the quality of evidence in each review from the 'Summary of findings' table for the main comparisons, as well as the reasons why the evidence quality was downgraded.
Results: we included 280 reviews from 50 different Cochrane Groups, and we excluded 22 reviews due to exclusion criteria. Twelve reviews did not present data in the 'Summary of findings' table, and so we included 246 reviews and 1254 outcomes in the analysis. With regard to the level of evidence, we identified 88 outcomes that had been assessed as high quality (7%), 328 as moderate quality (26%), 466 as low quality (37%), and 372 as very low quality (30%). The most frequent reasons for downgrading the level of quality of evidence were limitations in study designs and imprecision of results.
Conclusions: the quality of evidence for the main comparisons in Cochrane Reviews is predominantly low or very low.
Patient or healthcare consumer involvement: future primary studies need improved methodological design and greater precision to achieve results that will improve quality. Interventions that are investigated in systematic reviews of high quality can provide safer and more consistent information to consumers and decision-makers.