Predictive value of the Hemoglobin, Albumin, Lymphocyte, and Platelet score for mortality in asthma

Ann Allergy Asthma Immunol. 2026 Jun 3:S1081-1206(26)00259-0. doi: 10.1016/j.anai.2026.05.039. Online ahead of print.

Abstract

Background: Inflammation and nutritional status are critical factors in the progression of bronchial asthma. The hemoglobin, albumin, lymphocyte, and platelet (HALP) score has emerged as a promising integrated marker for assessing inflammatory and nutritional status, yet its prognostic value in asthma remains unexplored.

Objective: To explore the relationship between the HALP score and the prognosis of patients with asthma.

Methods: We analyzed data from 1,303 patients with asthma in NHANES(2005-2018). Survival associations were assessed using Kaplan-Meier curves, multivariable Cox regression, and restricted cubic splines (RCS). Predictive performance was evaluated via time-dependent ROC curves. An independent clinical dataset of hospitalized patients with asthma was used for external validation to ensure generalizability.

Results: In the NHANES cohort, the highest HALP tertile (Q3) showed the highest cumulative survival. Multivariable Cox analysis revealed that the Q3 group had a 53.1% lower mortality risk comparevd to the Q1 group (HR = 0.469, 95% CI: 0.259-0.847). RCS analysis identified a non-linear relationship with a threshold at 61.33; below this threshold, each 1-unit increase in HALP reduced mortality risk by 3% (HR = 0.97). The model demonstrated strong predictive performance (AUCs > 0.8 at 1, 3, 5, and 10 years). External validation dataset confirmed these findings, showing a 65.7% risk reduction in the Q3 group (HR = 0.343, 95% CI: 0.155-0.761).

Conclusion: This study reveals a significant nonlinear association between HALP score and all-cause mortality in adult patients with asthma. Low HALP score independently predicts poor prognosis, serving as a promising, readily available biomarker for identifying high-risk individuals.

Keywords: Asthma; HALP score; Mortality; NHANES; Non-linear association.