Decision making with multiple goals: extending the threshold approach

Area of Science:

• Medical decision-making
• Clinical informatics
• Health services research

Background:

• Optimizing patient treatment requires considering complex interactions between various diagnostic and therapeutic procedures.
• Existing methods for treatment selection may not fully account for these interdependencies, potentially leading to suboptimal outcomes.

Purpose of the Study:

• To present a novel multiple-threshold approach for determining the optimal combination of treatment actions.
• To provide a precise computational method for selecting interventions in complex patient cases.

Main Methods:

• Developed a multiple-threshold approach, extending the single-condition threshold method.
• Incorporated the analysis of interactions among therapeutic and diagnostic procedures.
• Validated the method through an example case study.

Main Results:

• The multiple-threshold approach demonstrated a precise way to compute optimal treatment actions.
• An 11-29% increase in the expected utility of treatments was observed in the example case.
• The method effectively accounts for interactions between different medical procedures.

Conclusions:

• The multiple-threshold approach offers a significant advancement in optimizing treatment selection.
• This method enhances expected treatment utility by systematically considering procedural interactions.
• The approach provides a valuable tool for clinical decision support systems.