External validation of the Hospital Frailty Risk Score among older adults receiving mechanical ventilation

Abstract Purpose:To externally validate the Hospital Frailty Risk Score (HFRS) to predict clinical outcomes in critically ill patients.Methods:We selected index hospitalizations of older adults (≥75 years old) receiving mechanical ventilation, using the United States Nationwide Readmissions Database (NRD) from January 1, 2016, to November 30, 2018. Frailty risk was determined by the HFRS using International Classification of Diseases, Tenth Revision Clinical Modification (ICD-10-CM) codes, and further subcategorized into low-risk (score <5), intermediate-risk (score 5-15), and high-risk (score >15) for frailty. We evaluated the HFRS to predict prolonged hospitalization, in-hospital mortality, and 30-day emergency hospital readmissions, using multivariable logistic regression, after adjustment for patient and hospital characteristics. Model performance was assessed using the c-statistic, Brier score, and calibration plots.Results:Among the 649,330 weighted index hospitalizations in the cohort, 50.0% were female, the median (interquartile range [IQR]) age was 81 (78-86) years old, and the median (IQR) HFRS was 10.8 (7.7-14.5). Among the cohort, 9.5%, 68.3%, and 22.2% were subcategorized as low-, intermediate-, and high-risk for frailty, respectively. After adjustment, high-risk patient hospitalizations were associated with increased risks of prolonged hospitalization (adjusted odds ratio [aOR] 5.59 [95% confidence interval [CI] 5.24-5.97], c-statistic 0.694, Brier score 0.216) and 30-day emergency hospital readmissions (aOR 1.20 [95% CI 1.13-1.27], c-statistic 0.595, Brier score 0.162), compared to low-risk hospitalizations. Conversely, high-risk hospitalizations were inversely associated with in-hospital mortality (aOR 0.46 [95% CI 0.45-0.48], c-statistic 0.712, Brier score 0.214). Calibration plots demonstrated good calibration for the adjusted analyses.Conclusions:In this external validation study, the HFRS was not successfully validated in the NRD to predict in-hospital mortality in critically ill older adults receiving mechanical ventilation. While the HFRS may predict prolonged hospitalization and 30-day readmission, its use should be avoided in the critically ill.