Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
Frailty Assessment in Older Adults: Reconstruction of a Risk Analysis Tool using Nursing Data
ABSTRACT
Background:
Frailty screening for older adults is of particular importance for those with declining health and social risk factors. However, the screening tools currently available to assess frailty do not offer automated real-time appraisal in the clinical setting. 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 investigated the accessibility of frailty components in the EHR of older medical/surgical patients.
Methods:
This was a developmental study that included encounters of older adult patients admitted to a medical/surgical unit at two academic hospitals in North Florida between January 2012 and May 2021. We used secondary data from a large retrospective observational study of EHR data to develop a frailty index that not only exhibited statistical and computational rigor but also possessed clear clinical utility. Among the frailty elements included were shortness of breath, functional status, cognitive decline, and unexplained weight loss. Total frailty risk index scores were calculated and their correlation to age and hospital death explored.
Results:
A total of 10,863 patients aged 65 years and above were included. The average age was 75.4 (SD = 7.7) years. Of the included participants, 4,938 (45.5%) were male. With regard to race-ethnicity, 7,724 participants (71.1%) were non-Hispanic White, 2,406 participants (22.1%) were Non-Hispanic Black, 314 participants (2.9%) were Hispanic, and the remaining 419 participants (3.9%) were other or unknown race-ethnicity. Our frailty index scores ranged from 24 to 84 and indicated a right skew. The mean frailty score for our sample was 45.3 (SD = 8.7). The Pearson correlation coefficient between our frailty index scores and age was found to be 0.297 (95% CI: 0.280, 0.314) and between frailty index scores and hospital death was found to be 0.097 (95% CI: 0.078, 0.115) in our sample. A univariate logistic regression analysis revealed a statistically significant association between the frailty score and hospital mortality (p<0.001).
Conclusions:
Our study applied a practice-based evidence framework to operationalize a frailty risk index utilizing the elements of frailty captured in the EHR data of older adult patients. The generated frailty index met our standard for a valid frailty index, and represented a proof of concept for leveraging existing assessment data to capture frailty in older patients. This provides the foundation for an operational support tool that makes the use of data collected through routine care available to the care team without increasing provider workload or interfering with workflow. Future study of the overall reliability and validity of these derived frailty scores is warranted.
Citation
