LASSO regression and Boruta algorithm to explore the relationship between neutrophil percentage to albumin ratio and asthma: results from the NHANES 2001 to 2018.

Abstract

The present study aims to investigate the relationship between the neutrophil-percentage-to-albumin ratio (NPAR) and asthma using least absolute shrinkage and selection operator (LASSO) regression and Boruta algorithm. Based on the National Health and Nutrition Examination Survey database from 2001 to 2018, a total of 31,138 eligible participants were included in this study. The participants were randomly divided into a training cohort and a validation cohort in a 7:3 ratio. LASSO regression and Boruta algorithm were applied to the training cohort for assessment, selection of the optimal model, and identification of potential confounding factors. A nomogram prediction model, receiver operating characteristic curve, calibration curve, and decision curve analysis were constructed to evaluate the model's ability to predict the risk of asthma and its stability. These analyses aim to provide a reference for clinical diagnosis and treatment. The study demonstrated that after adjusting for potential confounding factors, the NPAR was positively correlated with asthma incidence (P < 0.01). The area under the curve for the training set was 0.66 for LASSO regression and 0.64 for Boruta algorithm, indicating that LASSO regression exhibited superior performance. Through LASSO regression, 10 variables were selected, including gender, race, smoking status, hypertension, diabetes, cancer, poverty-income ratio, BMI, cardiovascular disease, and age. A nomogram prediction model was constructed based on these predictors. The calibration curve showed good fit between the two groups. A higher NPAR is significantly positively correlated with an increased risk of asthma.