Conclusiveness resolves the conflict between quality of evidence and imprecision in GRADE

The objective of our article is to show how "quality of evidence" and "imprecision," as they are defined in Grading of Recommendations Assessment, Development, and Evaluation (GRADE) articles, may lead to confusion. We focus only on the context of systematic reviews.

We analyze, with the aid of standard probabilistic and statistical concepts, the concepts of quality of evidence and imprecision as used in the GRADE framework. This enables us to point out some weaknesses in the relation between "quality of evidence" and "imprecision."

The GRADE framework contains terms familiar from classical statistics, but these terms are used in nonstandard ways. Notably, "imprecision" does not have the meaning in the GRADE framework that it has in statistics, and the well-known table of "evidence levels" wrongly suggests that "quality of evidence" and "accuracy" express the same concept-they do not.

We believe that "conclusiveness" rather than "imprecision" would be a suitable term to use when the question whether the CI excludes or includes certain critical margins is being addressed. Conclusiveness could also replace quality of evidence as the final step for a systematic reviewer.