Chapter 10: Overall Certainty of Evidence

Summary

When a systematic review is made to support recommendations, systematic review authors will rate the certainty of the body of evidence that informs each critical and important outcome12. When moving from the evidence to decision-making, one overall certainty in the evidence value (high, moderate, low, very low) is determined from all of the individual outcomes. This is informed by only the outcomes deemed critical, not the important outcomes. The overall certainty of evidence is typically made based on the critical outcome with the lowest certainty of evidence rating. For example, if the evidence profile presents the following outcomes, the overall certainty would be moderate: High certainty in a mortality reduction (Critical outcome), Moderate certainty in reduced incidence of hospitalization (Critical outcome), Low certainty in improvement in quality of life (Important outcome), Moderate certainty in increased serious adverse events (Critical outcome). This is because the lowest of the critical outcomes (mortality, hospitalization, and serious adverse events) is Moderate.

In certain situations, the overall confidence might not be based on outcomes which were pre-determined as critical for decision-making. The guideline panel may change what is considered to be critical based on the results of the systematic review. Certain positive outcomes or negative effects might have been found to occur infrequently; it is acceptable to decrease the importance of these outcomes. Similarly, the panel may identify specific critical outcomes that inform the recommendation and that may influence the overall certainty of the outcome; however, the caveat is that the certainty of the evidence cannot be higher than the critical harm.