Trade-offs in producing patient-specific recommendations from a computer-based clinical guideline: a case study

Area of Science:

• Medical Informatics
• Clinical Decision Support Systems

Background:

• Clinical practice guidelines (CPGs) are essential for evidence-based medicine.
• Computer-based CPGs aim to provide patient-specific recommendations.
• The availability and richness of online clinical data influence CPG output.

Purpose of the Study:

• To determine the amount of online clinical data needed for patient-specific CPG recommendations.
• To assess if increased online data enhances recommendation specificity.
• To evaluate the desirability of heightened specificity in CPG recommendations.

Main Methods:

• Analysis of the Agency for Health Care Policy and Research's quick reference guide for acute postoperative pain management.
• Categorization of patient-specific data items based on online availability and determinability.
• Evaluation of how data availability affects the specificity and text volume of computer-generated recommendations.

Main Results:

• The amount of patient-specific recommendation text varied with the online availability of different data categories.
• High specificity in recommendations, driven by extensive data, may not always be clinically advantageous.
• Design trade-offs exist when converting static CPGs into interactive, data-driven systems.

Conclusions:

• The richness of online clinical data significantly impacts patient-specific recommendations from computer-based CPGs.
• Balancing data requirements with the clinical utility of recommendation specificity is crucial.
• This case study highlights key considerations for developing effective digital clinical decision support tools.