LAUSR

Objective: To develop and validate the Social Frailty Index, a summary measure of social risk in older adults, and determine its ability to risk stratify beyond traditional medical risk models Design: Prognostic model development and validation using demographics and a comprehensive inventory of social characteristics Setting: The Health and Retirement Study, a longitudinal, nationally representative survey of U.S. adults >50 years. We developed the model using the 2010 wave and validated it using the 2012 wave; there was no overlap between 2010 and 2012 respondents. Participants 8250 adults aged [≥]65 years who completed the Psychosocial and Lifestyle Questionnaire (4302 in the 2010 development cohort, 3948 in the 2012 validation cohort). Main exposure: Demographic and social characteristics Main outcome: 4-year mortality Results: Within 4 years of the baseline interview, 22% of study participants in both the development and validation cohort had died. Drawn from 183 possible predictors, the final model included age, gender, and 8 social predictors: lacking neighborhood cleanliness, low perceived control over financial situation, having children and meeting with them less than yearly, not working for pay, less active with children (grandchildren, neighborhood, nieces/nephews), no volunteering or charity work, feeling isolated from others, being treated with less courtesy or respect. In the validation cohort, the model discriminated well (c-statistic 0.73) and was strongly associated with 4-year mortality (7.3% in the lowest decile, 59.9% in the highest decile). Also, the Social Frailty Index meaningfully risk-stratified participants beyond the Charlson score, a commonly used medical comorbidity index, and the Lee Index score, a comorbidity and function model. Conclusion: This prognostic index, which includes age, gender, and 8 social characteristics, accurately risk stratifies older adults and refines the prediction of commonly used comorbidity- and function-based risk models.