Using EHR Data to Identify Patient Frailty and Risk for ICU Transfer

Area of Science:

• Gerontology
• Healthcare Analytics
• Clinical Risk Prediction

Background:

• Frailty is a significant concern in older medical-surgical patients.
• Predicting intensive care unit (ICU) transfer is crucial for patient care planning.

Purpose of the Study:

• To evaluate the predictive capabilities of four frailty risk score (FRS) definitions.
• To assess FRS for predicting time to first ICU transfer in adults aged 50 and over.

Main Methods:

• Utilized nursing flowsheet data, laboratory tests, and ICD-10 codes to construct FRS.
• Employed Cox regression and Schemper-Henderson variance to model and assess predictive accuracy.
• Analyzed data by age group and controlled for clinical characteristics and sex.

Main Results:

• All FRS definitions significantly predicted time to first ICU transfer.
• Multivariable modeling controlling for clinical factors enhanced predictive accuracy.
• Highest risk for ICU transfer was observed in the 50 to <60 years and ≥80 years age groups.

Conclusions:

• Frailty, as measured by FRS, is a significant risk factor for ICU transfer in medical-surgical patients.
• FRS should be integrated into patient assessment and care planning for high-risk individuals.
• Frailty prevalence varied, with the highest risk for ICU transfer observed in specific younger and older age brackets.