Identifying Key Psychosocial Predictors of Hospital Readmission Among Older Adults: Application of Textual Analytics in Electronic Health Records

Objective

​This project is funded by the ARISE-GERI Joint Research Fund. The specific aims of our study are: (1) to assemble a lexicon of electronic health records-available search terms that define key psychosocial factors, namely depressive symptoms, anxiety, poor social support, financial strain, living alone, and caregiver stress; (2) to develop lexicon-based case detection algorithms for identification of each of these psychosocial factors; 3) to ascertain whether psychosocial factors identified by textual analysis add to existing prediction models for 30-day hospital readmission which are based on clinical and administrative data of older adults.

Collaborators

1. NTFGH

2. NTU

3. SUSS