External validation of the Electronic Screening Index of Frailty (e-SIF) in a population of 1.4 million inhabitants aged 65 years and older.

The Electronic Screening Index of Frailty (e-SIF) was first validated in 2022 in a sample of 9315 people, but further studies in larger populations were recommended to corroborate those findings. The objective was to evaluate e-SIF construct validity, in a large and independent sample from that used for its initial validation, by analysing score associations with age, sex, hospitalizations, institutionalizations, mortality, and health resource use. An observational 2-year longitudinal study (2018-2019) was conducted of the Catalan population aged ≥65 years (1.4 million people) using retrospectively collected data. Frailty was established according to e-SIF score. Study variables included sociodemographic characteristics, mortality, hospitalizations, institutionalizations, primary care visits, emergency visits, and day hospital sessions during the study period. The study included 1 465 312 people (mean age 75.8 years, 57.2% women). Frailty prevalence was 14.0% in women and 9.1% in men (P < .001) and progressively increased with age from 2.2% in the 65-69 age group to 38.9% in ≥95 age group. As frailty status increased, mortality, hospitalizations, institutionalizations, and health resource use also increased. Hospitalization-free survival and institutionalization-free survival worsened as frailty increased. Multivariate logistic regression models based on all 42 e-SIF items showed areas under the curve of 0.85, 0.75, and 0.82 in predicting 1-year mortality, hospitalizations, and institutionalizations, respectively. This large study corroborates the findings of the initial e-SIF validation study and reaffirms its good construct validity.