Revisiting the social frailty index in aging adults: relevance and implications for an emerging construct.

Frailty is an established benchmark of aging-related decline, yet most measures of frailty focus on physical decline. Increasing evidence that social environments influence trajectories of aging has led to growing interest in social frailty. Recently, a 10-item Social Frailty Index (SFI-10) was developed using data from the Health and Retirement Study (HRS) that used 8 social items plus chronological age and sex to predict mortality. This study evaluates the validity of the SFI-10, providing considerations for social frailty measurement. Utilizing HRS cohorts (n = 8264, ages 66 to 101) and cohorts from the Midlife in the United States (MIDUS) study (n = 5619, ages 40 to 75), the SFI-10 was evaluated alongside an index including only its social-environmental factors (SFI-8). Logistic regressions were used to predict mortality from the SFI-10, SFI-8, and only age and sex. Within MIDUS, the SFI-10, SFI-8, and age were further evaluated against external criteria relevant for social frailty. Findings indicate that chronological age is the most important predictor of mortality risk across both samples. Social-environmental factors predict mortality only marginally in older adults but more strongly in midlife. When associated with external criteria relevant for social frailty, the SFI-8 exhibited stronger criterion validity than the SFI-10. These findings suggest that social frailty may be best reflected by measures focusing exclusively on social behaviors and resources, excluding demographic characteristics. Our work underscores the need for well-designed measures of social frailty.