TY - GEN
T1 - Using the Hospital Frailty Risk Score to predict in-hospital mortality across all adult ages
AU - Kutrani, Huda K. Saleh
AU - Briggs, Jim
AU - Prytherch, David
AU - Andrikopoulou, Elisavet
AU - Spice, Claire
N1 - Will be Gold OA
PY - 2025/1/13
Y1 - 2025/1/13
N2 - The objective is to explore the relationship between the Hospital Frailty Risk Score (HFRS) and in-hospital mortality for all adults ages. Methods: This retrospective cohort study included patients aged 16 years and older who were admitted to Queen Alexandra Hospital in Portsmouth, UK from 01/01/2010 to 31/12/2019. We calculated HFRS for patients who had been previously admitted to the hospital within the previous 2 years. The study developed Logistic Regression models (crude and adjusted) for nine prediction periods of in-hospital mortality to assess model performance between (HFRS and in-hospital mortality) and (Charlson Comorbidity Index-CCI and in-hospital mortality), using AUROC. Results: The proportion of intermediate and high risk of frailty increased with prediction periods of in-hospital mortality. Crude HFRS models (AUROCs ranging from 0·782 to 0·829) were superior to crude CCI models (AUROCs ranging from 0·690 to 0·708) for all in-hospital mortality periods. However, HFRS combined with CCI was a better predictor of in-hospital mortality in all prediction periods. Conclusions: This study has demonstrated the utility of HFRS to predict in-hospital mortality in patients across all adult ages.
AB - The objective is to explore the relationship between the Hospital Frailty Risk Score (HFRS) and in-hospital mortality for all adults ages. Methods: This retrospective cohort study included patients aged 16 years and older who were admitted to Queen Alexandra Hospital in Portsmouth, UK from 01/01/2010 to 31/12/2019. We calculated HFRS for patients who had been previously admitted to the hospital within the previous 2 years. The study developed Logistic Regression models (crude and adjusted) for nine prediction periods of in-hospital mortality to assess model performance between (HFRS and in-hospital mortality) and (Charlson Comorbidity Index-CCI and in-hospital mortality), using AUROC. Results: The proportion of intermediate and high risk of frailty increased with prediction periods of in-hospital mortality. Crude HFRS models (AUROCs ranging from 0·782 to 0·829) were superior to crude CCI models (AUROCs ranging from 0·690 to 0·708) for all in-hospital mortality periods. However, HFRS combined with CCI was a better predictor of in-hospital mortality in all prediction periods. Conclusions: This study has demonstrated the utility of HFRS to predict in-hospital mortality in patients across all adult ages.
KW - Frailty
KW - Hospital Frailty Risk Score
KW - Age
KW - In-hospital mortality
UR - https://ebooks.iospress.nl/bookseries/studies-in-health-technology-and-informatics
UR - https://mie2025.efmi.org/
M3 - Conference contribution
T3 - Studies in Health Technology and Informatics
BT - Proceedings of MIE 2025
PB - IOS Press
T2 - Medical Informatics Europe 2025
Y2 - 19 May 2025 through 21 May 2025
ER -