A goal-oriented framework for specifying clinical guidelines and handling medical errors.

Area of Science:

• Medical Informatics
• Clinical Decision Support Systems
• Artificial Intelligence in Healthcare

Background:

• Computer-interpretable guidelines (CIGs) are crucial for evidence-based care and error reduction.
• Existing CIGs may not adequately address exceptional patient behaviors or unforeseen circumstances.
• This limitation can lead to deviations from optimal care pathways and potential medical errors.

Purpose of the Study:

• To develop a novel framework for monitoring, detecting, and handling exceptions during CIG execution.
• To ensure that deviations from expected CIG behavior are identified and managed proactively.
• To prevent exceptions from escalating into medical errors and maintain high standards of patient care.

Main Methods:

• A framework was designed to specify guideline goals and link them to executable tasks.
• Exceptions were defined and associated with specific goals for management.
• A state-based approach was employed to link tasks, plans, goals, monitored effects, and exceptions.
• The framework was demonstrated using a generic chronic disease management plan and a specific hypertension management case.

Main Results:

• The developed framework successfully monitors CIG execution for deviations.
• It can detect and link exceptions to relevant guideline goals.
• The state-based approach facilitates the management of exceptions through defined tasks or plans.
• The system demonstrated its utility in managing exceptions within a chronic disease context.

Conclusions:

• The proposed framework enhances the robustness of computer-interpretable guidelines.
• It provides a mechanism to proactively manage exceptions, thereby reducing the risk of medical errors.
• This approach contributes to safer and more reliable clinical decision support systems.