Improving completeness of electronic problem lists through clinical decision support: a randomized, controlled trial.

Area of Science:

• Clinical Informatics
• Health Services Research
• Primary Care Medicine

Background:

• Accurate clinical problem lists are essential for patient care, decision support, and research.
• Problem lists are frequently incomplete or outdated, hindering effective healthcare delivery.
• Improving problem list documentation is crucial for quality patient management.

Purpose of the Study:

• To evaluate if a clinical alerting system can enhance problem list documentation.
• To determine the effectiveness of an inference rule-based system in prompting providers.
• To assess the impact of alerts on the accuracy and completeness of electronic problem lists.

Main Methods:

• A cluster randomized trial involving 28 primary care clinical areas across 11 clinics.
• Intervention group received electronic health record-based alerts suggesting problems based on inference rules.
• Primary outcome was alert acceptance rate; secondary outcome was the number of documented problems.

Main Results:

• Alerts were accepted in 41.1% of 17,043 instances.
• The intervention group documented significantly more problems (adjusted OR=3.4, p<0.001).
• Problem inference alerts led to 6277 additional documented problems, with 70.4% added via alerts.

Conclusions:

• Problem inference alerts substantially increase the notation of important patient problems in primary care.
• This improvement in documentation has the potential to significantly facilitate quality improvement efforts.
• The system demonstrated effectiveness in addressing incomplete and outdated problem lists.