Table 3. Quality of Evidence and Strength of
Recommendations
Quality of
Evidence Interpretation
High-quality
evidence
Studies that provide high confidence in the effect estimate. New
data from future studies are thought unlikely to change the effect.
Moderate-quality
evidence
Studies that provide moderate confidence that the true effect is
likely to be close to the estimate but possibly could be substantially
different.
Low-quality
evidence
Studies that provide limited confidence about the effect. e true
effect may be substantially different from the estimate.
Very low-quality
evidence
Studies that provide very little confidence about the effect. e true
effect may be quite different from the estimate.
Strength of
Recommendation Interpretation
Strong
recommendation
Supported by moderate- to high-quality evidence (e.g., multiple
randomized controlled trials). Recommended course of action
would apply to all or almost all patients. Only a small proportion of
clinicians/patients would not want to follow the recommendation. In
rare instances, a strong recommendation may be based on very low- to
low-certainty evidence. For example, an intervention may be strongly
recommended if it is considered benign, low-cost, without harms,
and the consequence of not performing the intervention may be
catastrophic. An intervention may be strongly recommended against
if there is high certainty of more harm than the comparison with very
low or low certainty about its benefit.
Conditional
recommendation
Supported by lower-quality evidence or a close balance between
desirable and undesirable outcomes. Recommended course of action
would apply to the majority of the patients, but the alternative is
a reasonable consideration. Conditional recommendations always
warrant a shared decision-making approach, reflecting individual
values and preferences.
Ungraded Position
Statement
Evidence for a PICO question did not support a graded
recommendation or did not favor one intervention over the other;
guidance for this question provided by the voting panel.
Adapted from: Guyatt GH, Oxman AD, Vist GE, et al. GRADE: an emerging consensus on rating
quality of evidence and strength of recommendations. BMJ 2008; 336 :924–6, and GRADE
Working Group, list of GRADE working group publications and grants. Available at: https://www.
gradeworkinggroup.org/publications.html.
