Table 35Strength of evidence: Communicating strength of recommendation

Number of Studies; Subjects; DesignRisk of BiasConsistencyDirectnessPrecisionResults and Strength of Evidence
1; 341126
RCT
ModerateConsistency unknown (single study)DirectImpreciseRelatively few medical residents who received strong non-numeric recommendations for or against care (e.g., “we recommend”) reported that they would adhere to guideline-concordant care; there was little difference among guideline concordant care using language from various existing grading schemes (e.g., “we recommend, “clinicians should”).

Weak recommendation wording resulted in somewhat higher guideline-concordant care; effects differed by specific wording. For weak recommendations for care, “we suggest” performed best. For weak recommendations against care, “we conditionally recommend against” performed best.

Insufficient

Note: RCT = randomized controlled trial.

From: Results—Key Question 3: Communicating Uncertainty

Cover of Communication and Dissemination Strategies to Facilitate the Use of Health-Related Evidence
Communication and Dissemination Strategies to Facilitate the Use of Health-Related Evidence.
Evidence Reports/Technology Assessments, No. 213.
McCormack L, Sheridan S, Lewis M, et al.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.