Simple Multimorbidity Count vs Charlson Comorbidity Index for 28-Day Mortality  Prediction in Critically Ill Patients: A Head-to-Head Comparison

Abstract

Objective: Comorbidity burden is an important determinant of prognosis in critically ill patients. Although the Charlson Comorbidity Index (CCI) is widely used for mortality prediction, simpler multimorbidity-based approaches may provide similar prognostic information with greater practicality. This study compared the prognostic performance of the age-adjusted CCI and a simple multimorbidity count for predicting 28-day mortality in critically ill adults. Methods: This retrospective observational cohort study included 198 adult patients admitted to a general medical-surgical intensive care unit between June and December 2025. Comorbidity burden was assessed using the age-adjusted CCI and a simple unweighted multimorbidity count based on the same 19 chronic conditions. Receiver operating characteristic (ROC) curve analysis was used to evaluate discriminative performance. Independent associations with 28-day mortality were assessed using multivariable logistic regression. Results: The overall 28-day mortality rate was 23.2%. Non-survivors had significantly higher APACHE II scores and greater comorbidity burden according to both measures. The simple multimorbidity count showed numerically better discrimination than the CCI for predicting 28-day mortality (AUC, 0.67 vs 0.62), although the difference was not statistically significant (P = .11). In multivariable analysis, both the multimorbidity count (adjusted OR, 1.59; 95% CI, 1.26-2.00; P < .001) and the CCI (adjusted OR, 1.18; 95% CI, 1.01-1.37; P = .03) were independently associated with mortality. Conclusions: A simple multimorbidity count demonstrated prognostic performance comparable to that of the age-adjusted CCI. Its simplicity and ease of use may make it a practical tool for early risk stratification in heterogeneous ICU populations.