Reconstruction of a Risk Analysis Tool That Uses Nursing Data for Frailty Assessment Among Older Adults: Derivation and Internal Evaluation Study
Background: Frailty screening for older adults is of particular importance for those with declining health and social risk factors. However, numerous screening tools developed to assess frailty currently available do not offer automated appraisal in the clinical setting largely due to the challenges of data collection and the complexity of existing approaches. Thus, further adjustment and adaptation are required to correctly identify frailty. Although routine frailty screening is sporadic and inconsistently implemented, elements of frailty are captured in the electronic health record (EHR) from hospital admissions data. Objective: This study aimed to develop and validate a frailty classification system using readily available EHR data. Methods: This developmental study included encounters of older adult patients (≥65 years) admitted to medical/surgical units at two academic hospitals in North Florida between January 2012 and May 2021. Based on a practice-based evidence framework, EHR data was used to reconstruct a frailty index modeled after the Risk Analysis Index. Optimal cutpoints for frailty classification were determined using Youden's index through ROC analysis. Multiple logistic regression models were compared to evaluate predictive performance for hospital mortality, and component importance was assessed by sequentially removing each frailty parameter from the comprehensive model. Age-stratified analyses were performed to evaluate the robustness of classifications across age groups (and race/ethnicity). All regression models were estimated using Firth's penalized likelihood approach to address rare outcome events. Results: A total of 10,863 hospital patients (mean age 75.4±7.7 years, 45.5% male) were included. Using optimal cutpoints, patients were classified as Not Frail (40.6%), Pre-Frail (43.5%), or Frail (14.9%), with corresponding mortality rates of 0.45%, 0.82%, and 3.24%, respectively. After adjustment for confounders, Frail patients had significantly higher odds of mortality compared to Not Frail patients (OR=7.14, 95% CI: 4.30-11.84, p
