Online Clinical Calculator for Predicting 28-Day Mortality in Older Adult Patients With Sepsis-Associated Encephalopathy: Retrospective Study Using MIMIC-IV
Notice bibliographique
Résumé
BACKGROUND: Sepsis-associated encephalopathy (SAE) represents a critical complication of sepsis, especially among older adults. Despite its clinical relevance, there remains a lack of accessible and practical tools specifically designed to predict 28-day mortality in this vulnerable population. OBJECTIVE: We aimed to enhance the practical applicability of the model by creating a web-based tool that allows real-time, individualized mortality risk prediction, facilitating early intervention and informed decision-making in clinical practice. METHODS: Using data extracted from the MIMIC-IV (Medical Information Mart for Intensive Care IV) database, we identified older patients (≥65 years) with SAE (n=2165) and divided them into a development cohort (n=1531) and a validation cohort (n=634). Key risk factors associated with 28-day mortality were identified, and a predictive nomogram was constructed. Model performance was evaluated using the concordance index, integrated discrimination improvement, net reclassification index, and calibration curve analysis. Clinical applicability was assessed through decision curve analysis and benchmarked against traditional intensive care unit (ICU) scoring systems. Furthermore, the nomogram was deployed as a web-based application, enabling clinicians to input data and generate individualized mortality predictions. RESULTS: A total of 2165 older patients with SAE were included, among whom 290 (13.4%) died within 28 days of ICU admission. Multivariable logistic regression identified lower body weight (odds ratio [OR] 0.985, 95% CI 0.975-0.994; P=.001), lower systolic blood pressure (OR 0.972, 95% CI 0.957-0.986; P<.001), lower hemoglobin (OR 0.984, 95% CI 0.974-0.995; P=.005), lower PaO2 (OR 0.996, 95% CI 0.994-0.997; P<.001), and lower Glasgow Coma Scale score (OR 0.825, 95% CI 0.786-0.864; P<.001) as mortality risk factors. Higher respiratory rate (OR 1.083, 95% CI 1.029-1.141; P=.002), increased anion gap (OR 1.081, 95% CI 1.031-1.135; P=.001), elevated blood urea nitrogen (OR 1.045, 95% CI 1.016-1.076; P=.002), prolonged partial thromboplastin time (OR 1.033, 95% CI 1.016-1.050; P<.001), and reduced urine output (OR>0.99, 95% CI 0.999-1.000; P=.002) were also predictive. Patients admitted to "other" ICU types had lower mortality compared with the medical ICU reference group (OR 0.327, 95% CI 0.176-0.609; P<.001). The nomogram achieved concordance index values of 0.899 (development) and 0.897 (validation), outperforming sequential organ failure assessment (0.692), Acute Physiology Score III (0.804), Logistic Organ Dysfunction System (0.771), Simplified Acute Physiology Score II (0.704), and Oxford Acute Severity of Illness Score (0.753), with significant integrated discrimination improvement and net reclassification index improvements (all P<.001). Calibration curves confirmed good agreement between predicted and observed outcomes, while decision curve analysis supported the model's superior clinical utility. CONCLUSIONS: This study presents a novel, validated nomogram for predicting 28-day mortality in older patients with SAE, integrating routinely available clinical data. The deployment of the model as a digital tool enhances its accessibility and usability, providing clinicians with a practical resource for risk stratification and individualized patient management.
Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.
Comment cette classification a été obtenuedéplier
Prédiction machine sur la base complète
Imitation des enseignantsNi prévalence calibrée, ni vérité terrain. Validation humaine à venir. Le volet Gemma est une étiquette directe du modèle pour chaque travail de la base, lue sur la notice réduite au titre. Le volet Codex est un classifieur appris des 10 348 étiquettes directes de Codex et calibré sur les taux pondérés de l'échantillon; les champs sans appui suffisant ne portent aucun appel Codex. Le mode candidate est l'union des deux volets; le consensus est leur intersection. Ces sorties portent le statut machine_predicted_unvalidated et ne sont pas des étiquettes humaines.
Scores du classifieur distillé par catégorie (deux têtes)
| Catégorie | Codex | Gemma |
|---|---|---|
| Métarecherche | 0,003 | 0,009 |
| Méta-épidémiologie (sens strict) | 0,000 | 0,000 |
| Méta-épidémiologie (sens large) | 0,001 | 0,001 |
| Bibliométrie | 0,002 | 0,002 |
| Études des sciences et des technologies | 0,000 | 0,000 |
| Communication savante | 0,001 | 0,001 |
| Science ouverte | 0,000 | 0,001 |
| Intégrité de la recherche | 0,000 | 0,000 |
| Charge utile insuffisante (le modèle a refusé de juger) | 0,001 | 0,000 |
Scores machine (provisoires)
Les deux têtes enseignantes du modèle étudiant, lues sur ce travail. Un score ordonne la base pour la relecture; il n'affirme jamais une catégorie, et le statut de validation accompagne chaque rangée tel quel.
Scores de référence d'un modèle non mature (critères de maturité non atteints, 7 itérations). Un score ordonne; il n'affirme jamais une catégorie.
score_only:v0-immature-baseline · tel quel depuis la passe de notation : score_only signifie que le nombre peut ordonner les travaux, et qu'aucune étiquette de catégorie n'en découleClassification
machine, non validéePrédiction automatique; un appel candidat d’une seule source (Gemma direct ou Codex distillé), pas un consensus.
Le détail, modèle par modèle et score par score, se trouve en fin de page sous « Comment cette classification a été obtenue ».