Use of declarative statements in creating and maintaining computer-interpretable knowledge bases for guideline-based...

Area of Science:

• Medical Informatics
• Artificial Intelligence in Medicine

Background:

• Developing computer-interpretable clinical practice guidelines (CPGs) for decision support is labor-intensive.
• Existing methods require significant modeling expertise and effort.

Purpose of the Study:

• To reduce the effort required to create and maintain computer-interpretable knowledge bases for decision support.
• To formulate substantial portions of CPGs as computable statements.

Main Methods:

• Formulated CPGs as computable statements expressing declarative relationships between patient conditions and interventions.
• Developed query and expression languages for decision-support systems (DSS) to evaluate guideline statements.
• Utilized DSS for decision-making and targeted clinical information system commentaries.

Main Results:

• Declarative statements significantly reduce modeling expertise and effort for creating and maintaining decision-support knowledge bases.
• Enabled a DSS to evaluate guideline statements in specific patient situations.

Conclusions:

• The use of declarative statements offers a more efficient approach to developing computer-interpretable CPGs.
• Discusses implications for sharing computable knowledge bases for decision support.