Development and Validation of a Predictive Model for Wheezing Illness Following Human Bocavirus 1 Infection in Children.

Human Bocavirus 1 (HBoV1) is one major pathogen that has been associated with wheezing illnesses. However, there is still a lack of effective clinical predictive indicators for wheezing illnesses in children infected with HBoV1. A retrospective cohort study was conducted among pediatric patients with single-HBoV1 infection from September 2016 to August 2023. Then, univariate logistic regression was used to screen potential predictors for wheezing illness, and Least Absolute Shrinkage and Selection Operator (LASSO) regression was applied to minimize overfitting and select key predictors. Finally, a multivariate logistic regression model was constructed in a training dataset comprising 80% of patients and validated in an independent test dataset comprising 20% of patients. Then, its performance was evaluated using the Area Under the Curve (AUC). A total of 330 pediatric patients were enrolled, including 228 in the wheezing-illness group and 102 in the non-wheezing group. Three independent predictors, including abnormal NK cell percentage (OR = 1.101, 95 %CI 1.03-1.27), preterm birth (OR = 1.65, 95 %CI 1.49-1.82) and personal history of allergy (OR = 1.25, 95 %CI 1.11-1.41), were identified. The model achieved AUCs of 0.904 and 0.876 in the training and test sets, respectively. Using a Youden-derived threshold (0.382), the high-risk group in the test set had an observed wheezing rate of 85.3%, compared with 18.7% in the low-risk group (p < 0.001). Calibration was satisfactory (Hosmer-Lemeshow p = 0.324 and 0.576). A validated predictive model incorporating abnormal NK cell percentage, preterm birth and personal history of allergy accurately stratifies the risk of wheezing illness after HBoV1 infection in children, facilitating early clinical intervention.