Development and validation of a clinical severity prediction model for hospitalized children with pertussis.

BACKGROUND

To explore the risk factors for severe pertussis and to develop a classification model for illness severity in hospitalized children with pertussis.

METHODS

A regression analysis was carried out on the cases diagnosed with pertussis in Wuhan Children's Hospital from January 2022 to October 2024. With April 2024 as the cut-off point, the patients were divided into a training group and a verification group. SPSS and R language were used to analyze the risk factors of severe pertussis in the training group and build a classification model for illness severity in hospitalized children with pertussis.

RESULTS

Among the relevant predictors, HR, RR, Moist rales, Hb, Alb and IgG are the independent risk factors for severe pertussis. In this study, a nomogram model for illness severity in hospitalized children with pertussis was constructed. Firstly, through training sets, it was confirmed that the prediction model had good discrimination (AUC = 0.935) and calibration (Mean squared error = 0.00062). Finally, the predictive value of the model was further confirmed in the internal validation sets as well (AUC = 0.93, Mean squared error = 0.00089).

CONCLUSION

The nomogram constructed in this study provides a clinical tool for assessing severity stratification in hospitalized children with pertussis, guiding the early identification and clinical management of severe cases.

CLINICAL TRIAL NUMBER: Not applicable.