Inflammatory and demographic determinants of elevated white blood cell counts: insights from predictive modeling and NHANES analysis

Résumé fourni par la source

Hematologic markers such as white blood cell (WBC) count, red blood cell (RBC) count, and platelet count are essential indicators of inflammation and immune responses. In particular, increased WBC counts have been linked to systemic inflammation and chronic illnesses. Inflammatory markers like high-sensitivity C-reactive protein (hs-CRP) and alpha-1-acid glycoprotein (AGP) are associated with hematologic abnormalities, yet their ability to predict elevated WBC counts and their relationships with demographic factors are still not fully examined. Data from the National Health and Nutrition Examination Survey (NHANES) 2021–2023 were examined to investigate relationships between inflammatory markers, hematologic parameters, and demographic factors (age, gender, ethnicity). Descriptive statistics, correlation analysis, and ordinary least squares (OLS) regression were employed to determine the predictors of WBC, RBC, and platelet counts. A class-weighted Gradient Boosting model was created to forecast elevated WBC counts (top 25th percentile), with model performance evaluated through area under the receiver operating characteristic curve (AUC-ROC), precision, and recall metrics. AGP was a robust predictor of WBC count (β = 1.56, p < 0.001, 95% CI [1.28, 1.84]), with hs-CRP following (β = 0.058, p < 0.001, 95% CI [0.032, 0.084]). Younger people displayed more robust associations between AGP and WBC counts, whereas females demonstrated weaker associations with hs-CRP in comparison to males. Ethnicity was an important predictor, as Non-Hispanic Black participants had greater WBC counts (β = 0.67, p < 0.001) in comparison to Non-Hispanic White participants. The Gradient Boosting model achieved an AUC-ROC of 0.64 (95% CI [0.59, 0.69]), indicating poor discriminative ability. While the model demonstrated high recall (0.74), its precision was limited (0.42), highlighting the need for further optimization. Analysis of feature importance revealed that AGP, hs-CRP, and age were the key predictors. AGP and hs-CRP are powerful indicators of inflammation-related blood disorders, with demographic variables playing a significant role in these relationships. The predictive model showed promise for combining demographic and inflammatory data to pinpoint high-risk individuals, although it requires further refinement for clinical use. These results underscore the significance of tailored methods in hematologic and inflammatory evaluations, affecting the enhancement of diagnostics and the minimization of health inequalities. Not applicable.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.