Email recruitment to use web decision support tools for pneumonia

Area of Science:

• Medical Informatics
• Clinical Decision Support
• Pulmonology

Background:

• Effective management of Community Acquired Pneumonia (CAP) relies on timely clinical decisions informed by accurate patient data.
• Existing clinical information systems (CIS) and decision support tools can be integrated to aid guideline application.

Purpose of the Study:

• To develop and evaluate a system integrating CIS, email, and a web-based pneumonia severity evaluation tool (SET) for CAP classification and guideline support.
• To assess physician response rates and the accuracy of CAP classification using the developed system compared to chart review.

Main Methods:

• Integration of existing CIS, email, and a web-based SET using secure tokens for data linkage.
• Physician engagement through email solicitations and assessment of SET utilization.
• Comparison of pneumonia classification accuracy against expert chart review as the gold standard.

Main Results:

• Physician response rate to email solicitations averaged 50% over 14 months.
• Professional coding of pneumonia as a diagnosis was a sensitive indicator for CAP.
• Pneumonia classification using the SET demonstrated high reliability compared to expert chart review.
• A trend towards admitting more severe CAP cases was observed, with lengths of stay exceeding guideline expectations.

Conclusions:

• The integrated system reliably classifies CAP severity, but physician-reported guideline misfit and extended lengths of stay indicate potential limitations in current guidelines.
• Further refinement of classification systems and guidelines is needed to better align with current clinical practice for CAP management.