Diagnostic testing and treatment under ambiguity: using decision analysis to inform clinical practice

Area of Science:

• Medical Decision Making
• Health Economics
• Clinical Informatics

Background:

• Partial knowledge of patient health status and treatment response complicates optimal medical decision-making.
• Clinical practice guidelines (CPGs) aim to optimize patient care but are often limited by incomplete information.
• Current medical decision-making frameworks may not fully account for uncertainty in patient outcomes.

Purpose of the Study:

• To analyze a common clinical scenario involving diagnostic testing and treatment under conditions of partial knowledge.
• To characterize optimal patient care by integrating clinician objectives, knowledge, and decision criteria.
• To re-evaluate the role and limitations of clinical practice guidelines in the context of decision analysis.

Main Methods:

• Developed a decision analysis framework for a patient presenting with initial evidence of health status.
• Modeled the choice between immediate treatment and diagnostic testing followed by treatment.
• Incorporated partial knowledge of treatment and testing response, and considered various decision criteria.

Main Results:

• Optimal care is determined by the joint influence of a clinician's objective, knowledge, and decision criterion.
• Partial knowledge of treatment and testing response significantly impacts the decision-making process.
• The analysis highlights scenarios where diagnostic testing can improve patient outcomes despite uncertainty.

Conclusions:

• Clinical practice guidelines may need refinement to better incorporate decision analysis under partial knowledge.
• Integrating decision theory can lead to more personalized and effective patient care strategies.
• Understanding the interplay of knowledge and decision criteria is crucial for optimizing treatment pathways.