Risk prediction of 30-day in-hospital all-cause mortality among critically Ill patients transferred from the emergency resuscitation room: development and validation of a nomogram model
Year: 2026
Type: article
Abstract: Background The emergency resuscitation room serves as the site for initial assessment and resuscitation of critically ill patients. Most patients are transferred to specialized wards for further management after emergency intervention. Currently, prediction tools for mortality risk in this critically-ill population remain scarce. This study aimed to develop a nomogram model integrating early biomarkers to predict 30-day in-hospital mortality among patients hospitalized following emergency resuscitation room care. Methods A single-center retrospective study was performed. A total of 986 adult critically ill patients transferred from the emergency resuscitation room to specialized wards at Zunyi First People’s Hospital between January 2023 and December 2025 were enrolled. Missing data were handled by multiple imputation (m = 5). Candidate predictors were screened by least absolute shrinkage and selection operator (LASSO) regression with dual-stability criteria (≥ 70% selection across imputed datasets and bootstrap resamples). Multivariable logistic regression, with estimates pooled via Rubin’s rules, was performed to identify independent predictors of 30-day in-hospital all-cause mortality. Bootstrap internal validation with optimism correction was performed; discrimination, calibration, continuous net reclassification improvement (NRI), integrated discrimination improvement (IDI) and DeLong’s test were used to evaluate model performance. Stratified analyses across five disease categories were implemented to assess effect heterogeneity. Results The 30-day in-hospital mortality was 21.0%. Age, lactate-to-albumin ratio (LAR), D-dimer-to-fibrinogen ratio (DFR), and glucose were identified as independent risk predictors. A nomogram was constructed to visualise the final prediction model. The nomogram achieved an optimism-corrected AUC of 0.82 with satisfactory calibration (calibration-in-the-lare 0.00, 95% CI, − 0.17–0.17; calibration sloe 0.96, 95% CI, 0.85–1.17). It exhibited superior predictive performance compared with a simple readily-available bedside age-plus-lactate reference model (ΔAUC 0.06, P < 0.001). Age, LAR and glucose showed consistent prognostic effects across disease subgroups, while DFR presented significant heterogeneity ( P -interaction = 0.008), with diminished predictive effect in trauma patients. Conclusions This internally validated nomogram based on readily available routine laboratory indices reliably predicts short term mortality risk for resuscitation room originated critically-ill patients. It may serve as a non-specific early warning triage tool to support emergency risk stratified clinical decision making. External validation in independent cohorts is required before broad clinical application. Clinical trial number Not applicable.
Source: BMC Emergency Medicine
Language: English
Field: Medicine
Domain: Health Sciences
Open Access status: gold
Estimated APC paid: $2,790
OpenAlex ID: W7225806317
FWCI:
Cites:
Cited by:
Related to: