Beyond chronological age: the self-administered multidimensional prognostic index (SELFY-MPI) as an independent predictor of prolonged hospitalization in older surgical patients

Frailty is a multidimensional geriatric syndrome that substantially increases perioperative morbidity and resource utilization in older surgical patients. The Self-Administered Multidimensional Prognostic Index (SELFY-MPI), a validated patient-reported tool derived from the Comprehensive Geriatric Assessment, has not been systematically evaluated in a surgical context. We aimed to determine whether the SELFY-MPI score independently predicts length of hospital stay (LOS) in older patients undergoing general surgery. This prospective observational cohort study enrolled 101 consecutive patients aged ≥ 65 years admitted to a general surgery ward. SELFY-MPI was administered preoperatively. Primary endpoint was length of hospital stay. Pearson correlation analysis and multivariable linear regression were performed to assess the independent association between SELFY-MPI and LOS. Median age was 82 (IQR 78–86) years; 54 patients (53.5%) were classified as higher frailty risk (SELFY-MPI ≥ 0.34). Frail patients exhibited significantly longer LOS compared to non-frail patients 12 (7–18) vs. 9 (7–14) days ( p = 0.048). Pearson correlation confirmed a significant positive association between SELFY-MPI score and LOS ( r = 0.313; p = 0.002). In multivariable regression analysis adjusted for age (Model 1: β = 13.18, SE = 4.84, p = 0.008, R²=0.10) additionally for sex (Model 2: β = 9.86, SE = 4.77, p = 0.041, R²=0.18), and surgical procedures (Model 3: β = 9.62, SE = 4.76, p = 0.046), the SELFY-MPI score emerged as the only independent predictor of LOS, while age did not reach statistical significance. SELFY-MPI is a rapid, self-administered, multidimensional frailty assessment tool that results significantly associated with prolonged hospitalization in older surgical patients. Its systematic preoperative integration may enable risk stratification, personalize perioperative pathways, and optimize resource allocation in geriatric surgical populations.