Validation of an Electronic Health Record Algorithm for Identifying Housing-Related Needs in a Safety-Net Health...

Area of Science:

• Health Informatics
• Public Health
• Health Services Research

Background:

• Identifying patients with housing-related needs is crucial for health systems addressing health-related social needs (HRSN).
• Electronic health records (EHRs) offer a potential avenue for efficient identification of these needs.
• Existing methods may lack the necessary efficiency or accuracy in diverse healthcare settings.

Purpose of the Study:

• To develop and validate a novel, pragmatic EHR-based algorithm for identifying patients with housing-related needs.
• To assess the algorithm's sensitivity and specificity within a safety-net hospital setting.
• To compare the algorithm's performance against individual data elements and traditional methods.

Main Methods:

• Retrospective evaluation of an EHR algorithm at Boston Medical Center (BMC), a safety-net hospital.
• The algorithm incorporated six EHR structured data elements, including HRSN screening results.
• Validated against two reference groups: housing program enrollees and a primary care patient sample with manual chart review.

Main Results:

• The algorithm achieved a sensitivity of 62% among housing program enrollees and 81% among primary care patients.
• Specificity was high at 97% in the primary care group.
• While HRSN screening had the highest single-element sensitivity, the algorithm including multiple elements outperformed individual components and diagnostic codes.

Conclusions:

• Pragmatic EHR algorithms using locally tailored structured data are an accessible and efficient method for identifying patients with housing-related needs.
• This study is the first to validate such an algorithm in a safety-net setting, demonstrating moderate sensitivity and high specificity.
• Further improvements necessitate enhancements in HRSN screening and EHR documentation practices.