Identification of Risk Factors and Development of Prediction Models for Glasgow Coma Scale Score Deterioration in Stroke Patients: A Study Based on the MIMIC-IV Database

Objective Stroke remains a major cause of mortality and disability worldwide. This study aims to identify the risk factors associated with Glasgow Coma Scale (GCS) deterioration in stroke patients using data from the MIMIC-IV database. Methods We conducted a retrospective cohort study based on the MIMIC-IV database, involving 1984 adult stroke patients. The main exposure variables included age, Charlson Comorbidity Index (CCI), and Sequential Organ Failure Assessment (SOFA) score. The outcome variable was GCS deterioration during hospitalization. Covariates included demographic information, comorbidities, and laboratory indicators. Multivariate logistic regression models were used to analyze risk factors. Results Each 1-point increase in SOFA score was associated with a 50.69% increased risk of GCS score deterioration (OR = 1.5069, 95% CI: 1.2641-1.7964, P < .0001). Each year increase in age was associated with a 24.19% increased risk (OR = 1.2419, 95% CI: 1.0496-1.4695, P = .0116). Each 1 mEq/L increase in minimum anion gap was associated with a 24.56% increased risk (OR = 1.2456, 95% CI: 1.0076-1.5398, P = .0424). Risk factors varied significantly among disease subtypes. Conclusion SOFA score, age, and anion gap are important predictors of GCS score deterioration in patients with neurological diseases. These findings may help identify high-risk patients early and optimize treatment strategies.