Reconstruction of a Risk Analysis Tool That Uses Nursing Data for Frailty Assessment Among Older Adults: Derivation and Internal Evaluation Study
Bibliographic record
Abstract
Background: Frailty screening for older adults is of particular importance for those with declining health and social risk factors. However, many existing frailty tools do not offer automated appraisal in clinical settings due to challenges in data collection and the complexity of current approaches. Although routine frailty screening is inconsistently implemented, elements of frailty are captured in the electronic health record (EHR) from hospital admissions data. Thus, further adjustments and adaptations are required to correctly identify frailty. Objective: This study investigated existing frailty-related EHR data at two hospitals within a single health system to (1) identify key data elements and establish their availability in existing clinical workflows, and (2) use these data elements to reconstruct a validated and widely used frailty index. Methods: This developmental study included encounters of older adult patients (aged ≥65 years) admitted to medical-surgical units at 2 academic hospitals in North Florida between January 2012 and May 2021. EHR data were used to reconstruct a frailty index modeled after the Risk Analysis Index. Optimal cut points for frailty classification were determined through receiver operating characteristic 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 to address rare outcome events. Results: A total of 10,863 hospital patients (45.5% male; mean age 75.4, SD 7.7 years) were included. Using optimal cut points, patients were classified as not frail (40.6%), prefrail (43.5%), or frail (14.9%), with corresponding mortality rates of 0.45%, 0.82%, and 3.24%. After adjustment for confounders, frail patients had significantly higher odds of mortality compared to not frail patients (odds ratio 7.14, 95% CI 4.30-11.8; P<.001). Component importance analysis identified shortness of breath (area under the receiver operating characteristic curve reduction: 0.051) and functional status (area under the receiver operating characteristic curve reduction: 0.047) as the strongest predictors. Age-stratified analyses confirmed the classification system's validity, with frail patients aged 65 to 75 years (mortality 3.97%) having higher mortality risk than not frail patients aged ≥86 years (mortality 0.48%). Conclusions: This study developed and validated a 3-category frailty classification system with strong predictive validity for hospital mortality among older adults. The generated frailty index leverages existing EHR data to capture frailty without increasing provider workload or interfering with workflow. Component importance analysis identified respiratory dysfunction and functional limitations as key predictors. This automated approach to frailty assessment could improve risk stratification, providing critical information to clinical teams with minimal burden and supporting clinical decision-making for hospitalized older adults. Future study of the overall reliability and validity of these derived frailty scores is warranted.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".