Uncertainty-Adjusted Translation for Preference-Sensitive Decision Support.

Area of Science:

• Decision Science
• Health Informatics
• Patient-Centred Care

Background:

• Current Multi-Criteria Decision Analysis (MCDA) tools for person-centred care use central estimates for option performance, ignoring reported uncertainty intervals.
• Existing methods fail to integrate parameter uncertainty into decision analyses, leaving decision-makers unsupported.
• There's a need to address the gap in decision support tools regarding the handling of uncertainty in performance ratings.

Purpose of the Study:

• To propose and demonstrate a method for integrating parameter uncertainty into MCDA-based decision support.
• To enhance decision-making in person-centred care by incorporating uncertainty into the analysis.
• To provide decision-makers with a more comprehensive understanding of option performance.

Main Methods:

• Treating both the means and uncertainties of outcomes as separate criteria within the MCDA framework.
• Incorporating these uncertainty criteria into the trade-off process, making it preference-sensitive.
• An empirical proof of method using an online example of bone health medications with six options and four criteria.

Main Results:

• Demonstrated a novel approach to integrate uncertainty into MCDA for decision support.
• The proposed method allows for preference-sensitive trade-offs involving uncertainty.
• Empirical validation through a bone health medication example showcases the method's applicability.

Conclusions:

• Integrating uncertainty as a criterion within MCDA provides a more robust decision support system for person-centred care.
• This approach empowers decision-makers by explicitly addressing and trading off performance uncertainties.
• The method offers a practical solution for enhancing personalized care decisions by accounting for variability in evidence.