Application of ARIMA model in predicting carbapenem-resistant Klebsiella pneumoniae healthcare-associated infection

Abstract:

Objective To explore the application of autoregressive integrated moving average (ARIMA) model in the prediction of carbapenem-resistant Klebsiella pneumoniae (CRKP) healthcare-associated infection (HAI). Methods The monthly incidence of CRKP HAI in a hospital from January 2018 to June 2022 was selected. ARIMA model was established by SPSS19.0, and the fitting of the model were analyzed. CRKP HAI data from July 2021 to June 2022 was used to verify the model and evaluate its predictive value. Results Monthly CRKP HAI incidence of hospitalized patients from January 2018 to June 2022 was used to establish the model and conduct fitting process, finally acquired the optimal model ARIMA (0, 1, 9). With an average relative error of 7.76%, the model-predicted value for CRKP HAI incidence was consistent with the actual value, thus fitting well with actual data. Conclusion ARIMA model can effectively fit and predict the incidence of CRKP HAI, and provide scientific guidance for CRKP HAI prevention and control.