Certainty ranges facilitated explicit and transparent judgments regarding evidence credibility

Area of Science:

• Medical Evidence Synthesis
• Clinical Decision Making
• Health Research Methodology

Background:

• The Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach has limitations in quantifying certainty of evidence due to non-quantitative domains.
• Existing methods struggle to incorporate factors like risk of bias, inconsistency, indirectness, and publication bias into evidence certainty ratings.

Purpose of the Study:

• To apply the concept of "certainty ranges" to quantify evidence certainty for venous thromboembolism (VTE) and bleeding risks in urological surgery.
• To determine if expanding plausible effect ranges could justify upgrading the quality of evidence in specific patient populations.

Main Methods:

• Developed and applied "certainty ranges" by expanding confidence intervals by 20% for four non-quantitative GRADE domains (risk of bias, inconsistency, indirectness, publication bias).
• Assessed the net benefit of VTE prophylaxis by comparing the plausible range of VTE reduction against the plausible range of increased bleeding.
• Utilized examples of open radical cystectomy and laparoscopic partial nephrectomy to test the robustness of the certainty range method.

Main Results:

• In high-VTE risk laparoscopic partial nephrectomy and high/medium-VTE risk open radical cystectomy patients, results were robust to expanded certainty intervals, supporting an upgrade in evidence quality.
• In low-VTE risk patients, the results were not robust, and upgrading evidence quality was deemed inappropriate.
• The "certainty range" approach demonstrated its utility in empirically assessing evidence certainty in a clinical decision-making context.

Conclusions:

• The "certainty range" concept provides a practical method for enhancing the quantification of evidence certainty in clinical practice.
• This empirical application supports the use of certainty ranges to objectively upgrade evidence quality when net benefits are unequivocally positive.
• Further theoretical and practical exploration of certainty ranges is warranted to refine evidence synthesis and guideline development.