Clinical Relevance of HALP Score in Predicting Hospital Stay Duration and Outcomes in Acute Heart Failure.

Background: Heart failure (HF) remains a global health burden characterized by frequent hospitalizations and high mortality. The Hemoglobin, Albumin, Lymphocyte, and Platelet (HALP) score has emerged as a composite marker reflecting nutritional and inflammatory status, but its prognostic utility in acute HF settings remains underexplored. Objective: To investigate the association between the HALP score and prolonged hospital stay and in-hospital mortality among patients admitted with acute heart failure. Methods: This retrospective study included 222 patients hospitalized with acute heart failure at a tertiary care center in Somalia between January 2024 and April 2025. Patients were stratified by hospital stay duration: short (≤7 days) vs long (>7 days). The HALP score was calculated using routine laboratory values, and a previously validated cut-off was used for stratification. Logistic regression analysis was used to identify predictors of long hospital stay. Model performance was assessed using Hosmer-Lemeshow test, Nagelkerke R2, classification statistics, and ROC curve analysis. Results: A total of 222 patients were analyzed; 86 (38.7%) had prolonged hospitalization. Patients with the HALP score above the prognostic threshold were significantly more likely to experience long hospital stay (p = 0.002). In multivariable analysis, The HALP score ≥ cut-off (OR: 10.19, 95% CI: 2.49-41.63, p = 0.002) and prior stroke (OR: 8.44, 95% CI: 1.15-61.88, p = 0.035) independently predicted prolonged hospital stay. Model fit was adequate (Hosmer-Lemeshow p = 0.105), and explanatory power was moderate (Nagelkerke R2 = 0.31). However, the HALP score's standalone discriminative ability was poor (AUC = 0.511). Conclusion: The HALP score is an independent predictor of prolonged hospitalization in acute heart failure patients. While its individual discriminative power is limited, its role within multivariable risk stratification models appears promising. Further prospective validation is warranted.