Improving GRADE evidence tables part 2: a systematic survey of explanatory notes shows more guidance is needed
Author
dc.contributor.author
Langendam, Miranda
Author
dc.contributor.author
Carrasco Labra, Alonso
Author
dc.contributor.author
Santesso, Nancy
Author
dc.contributor.author
Mustafa, Reem
Author
dc.contributor.author
Brignardello Petersen, Romina
Author
dc.contributor.author
Ventresca, Matthew
Author
dc.contributor.author
Heus, Pauline
Author
dc.contributor.author
Lasserson, Toby
Author
dc.contributor.author
Moustgaard, Rasmus
Author
dc.contributor.author
Brozek, Jan
Author
dc.contributor.author
Schunemann, Holger
Admission date
dc.date.accessioned
2016-12-06T18:57:08Z
Available date
dc.date.available
2016-12-06T18:57:08Z
Publication date
dc.date.issued
2016
Cita de ítem
dc.identifier.citation
Journal of Clinical Epidemiology 74 (2016) 19-27
es_ES
Identifier
dc.identifier.other
10.1016/j.jclinepi.2015.12.008
Identifier
dc.identifier.uri
https://repositorio.uchile.cl/handle/2250/141693
Abstract
dc.description.abstract
Objectives: The Grading of Recommendations Assessment, Development, and Evaluation (GRADE) working group has developed GRADE evidence profiles (EP) and summary of findings (SoF) tables to present evidence summaries in systematic reviews, clinical guidelines, and health technology assessments. Explanatory notes are used to explain choices and judgments in these summaries, for example, on rating of the quality of evidence.
Study Design and Setting: A systematic survey of the explanations in SoF tables in 132 randomly selected Cochrane Intervention reviews and in EPs of 10 guidelines. We analyzed the content of 1,291 explanations using a predefined list of criteria.
Results: Most explanations were used to describe or communicate results and to explain downgrading of the quality of evidence, in particular for risk of bias and imprecision. Addressing the source of baseline risk (observational data or control group risk) was often missing. For judgments about downgrading the quality of evidence, the percentage of informative explanations ranged between 41% (imprecision) and 79% (indirectness).
Conclusion: We found that by and large explanations were informative but detected several areas for improvement (e.g., source of baseline risk and judgments on imprecision). Guidance about explanatory footnotes and comments will be provided in the last article in this series. (C) 2016 Elsevier Inc. All rights reserved
es_ES
Patrocinador
dc.description.sponsorship
Cochrane Collaboration's Methods Innovation Fund
McMaster GRADE Center