Frail older people have a high risk of poor health outcomes. GPs are encouraged to identify those who are moderate-severely frail and provide personalised care.
The electronic frailty index (eFI) supports this, but has limitations, including insufficient attention to mental health and time-varying conditions.
This project set out to improve the eFI’s clinical utility: to determine the best combination of mental health variables to predict frailty outcomes, and also to optimise time-dependent variables.