Admission Red Cell Distribution Width-Albumin Ratio and 1-Year Mortality in Chronic-Wound Inpatients: Prediction-Model Development, Internal Validation, and Exploratory Temporal Evaluation

Admission-day RAR discriminated 1-year all-cause mortality better than its components, albumin and red cell distribution width. Admission-day mortality tools have been lacking for chronic-wound inpatients; this study evaluated RAR, a ratio of two routine admission markers, as a candidate predictor paired with ACCI. Single-center retrospective cohort of 584 first-admission patients, with admission markers compared by AUC; 53 patients (9.1%) died within 1 year.

A two-variable logistic model of RAR and ACCI showed strong discrimination and good calibration, stratifying patients into low-, intermediate-, and high-risk tiers. The model uses only two variables available on the day of admission, providing a patient-level 1-year mortality stratification tool for mixed-etiology chronic-wound inpatients. Internally validated by bootstrapping, with an optimism-corrected C-statistic of 0.855 and good calibration; observed mortality across tiers was 1.3%, 9.0%, and 36.8%.

A high admission RAR (≥5.2) or low albumin (<3.0 g/dL) likewise flagged high-risk patients. Provides simplified threshold cues usable on the day of admission without the full model. Based on admission-marker analysis in the same cohort, consistent with the model's risk stratification.

In an exploratory temporal evaluation of a later same-center cohort of 124 patients, discrimination was similar to the development cohort, but calibration was imprecise. First exploratory temporal check of the model within the same center at a later period. Exploratory temporal evaluation with 124 patients; discrimination similar, calibration imprecise.

The model is intended for adults first admitted with a chronic wound of any etiology, using two variables available on the day of admission, RAR and ACCI, for 1-year all-cause mortality risk stratification in similar single-center inpatient settings; development and internal validation were based on 584 patients from 2021 to 2024, and the exploratory temporal evaluation on a later same-center cohort of 124 patients.

Calibration was imprecise in the temporal evaluation, suggesting model performance may vary across time periods; the single-center design means transportability to other institutions, case mixes, and care pathways still requires validation; and the clinical utility of the RAR threshold (≥5.2) and albumin threshold (<3.0 g/dL), as well as the model's effect on clinical decisions and patient outcomes, awaits prospective confirmation.