Quantifying care coordination using natural language processing and domain-specific ontology.

Area of Science:

• Nursing
• Healthcare Management
• Gerontology

Background:

• Older adults living in the community require coordinated care.
• Effective care coordination is crucial for managing health conditions in community-dwelling older adults.
• Existing methods for quantifying care coordination are limited.

Purpose of the Study:

• To identify specific care coordination activities used by Aging in Place (AIP) and home healthcare (HHC) nurses.
• To propose a method for quantifying the extent of care coordination provided to older adults.
• To compare the intensity of care coordination between AIP and HHC models.

Main Methods:

• Developed a care coordination ontology using activities from 11,038 case management notes.
• Mapped patient narrative notes to the ontology to compute problem profiles.
• Quantified care coordination for patients in both AIP and HHC groups.

Main Results:

• Compared 217 patients in AIP with 217 patients in HHC using 128,135 narrative notes.
• Patients in AIP received a higher "dose" of care coordination compared to HHC patients.
• AIP demonstrated larger care coordination doses across all four Omaha categories.

Conclusions:

• "Communicate" and "manage" are the most frequent care coordination activities identified.
• Nurse care coordinators primarily focus on patient communication and problem management.
• The overall aggregated dose of care coordination is significantly larger in the AIP model than in HHC.