Multidimensional frailty score predicts the risk of in-hospital complications and adverse clinical outcomes in elderly hospitalized patients: a retrospective cohort study

To investigate the predictive efficacy of multidimensional frailty assessment scoring for in-hospital complications (IHC) risk and clinical outcomes in elderly inpatients.

A retrospective analysis was conducted on 136 elderly patients aged ≥65 years who were hospitalized in Department of Geriatrics of First Affiliated Hospital of Army Medical University from May 2022 to September 2022. Demographic characteristics and clinical outcome indicators (critical illness, IHC, length of hospital stay, and discharge outcomes) were extracted from the medical record system. Nine frailty factors, in aspects of social, psychological, and physiological dimensions were identified from medical records within 72 h after admission, and each positive factor scored as 1 point. The multidimensional frailty score was calculated as a total score for all 9 factors. IBM SPSS statistics 23.0 was used for data analysis to evaluate the predictive value of the multidimensional frailty score for critical illness during hospitalization. This study was approved by the Ethics Committee of the First Affiliated Hospital of Army Medical University [approval number (B) KY2025245].

① IHC occurred in 36 patients (26.5%) during hospitalization; Those developing IHC had a higher proportion of critical illness on admission, longer hospital stays, and higher multidimensional frailty scores. ② The AUC values of the multidimensional frailty score in predicting IHC and critical illness during hospitalization were 0.712 (95%CI: 0.614 to 0.810) and 0.738 (95%CI: 0.644 to 0.832), respectively, with optimal cutoff values of 2 and 3 points. ③ The patients with higher multidimensional frailty scores had an increased risk for in-hospital IHC (β =0.487, OR=1.628, P < 0.001) and critical illness on admission (β=0.557, OR=1.746, P < 0.001); ④ Conversely, those with elevated frailty scores showed reduced probability of favorable discharge (β=-0.602, OR=0.547, P=0.040).

Multidimensional frailty assessment is simple to perform and demonstrates good predictive value for the risk of IHC and critical illness on admission in elderly patients, and can be used for rapid assessment of newly admitted patients.