AGMR

Background: The electronic Frailty Index (eFI) was developed and validated using routinely available primary care electronic health record data. This study aims to develop mapping function for mapping between the eFI and a research standard frailty index validated as part of the English Longitudinal Study of Ageing (ELSA). Mapping eFI and ELSA frailty index (ELSAFI) measures could help facilitate the translation of research findings into clinical practice.Method: Data from the Community Ageing Research 75+ (CARE75+) UK prospective cohort was used in this research. Regression analysis was used to develop mapping functions that can be used for translating the eFI into ELSAFI and vice versa. The predictive performance of the mapping function was evaluated using the Root Mean Square Error (RMSE), Mean Average Error (MAE), adjusted R square, and parsimony. Results: A total of 896 individual cases was included in the analysis. The mean age of participants was 82.9 years (SD = 5.1) and 54.9% were females. There was positive correlation between the eFI and ELSAFI, r = 0.60. The mapping function of ELSAFI to eFI explained 38% of the variance in the derivation sample whilst the mapping algorithm of eFI to ELSAFI explained 40%. Both mapping functions had RMSE of 0.09. Conclusion: The eFI can accurately be mapped on to a research standard ELSAFI and vice versa at the group level but not at individual level. Further testing is required to validate the mapping functions developed in this study in independent samples.

mapping; electronic frailty index; frailty; older people