Research on Dynamic Outpatient Respiratory Nosocomial Infection Control Methods Through Multi-Data Prediction.

Abstract

Objective: This study aimed to develop a dynamic prevention and control method for fluctuating respiratory nosocomial infections in outpatients.

Methods: Six sets of surveillance data such as influenza-like case counts and their predicted results were used in the autoregressive integrated moving average model (ARIMA) to forecast the onset and end time points of the epidemic peak. A Delphi process was then used to build consensus on hierarchical infection control measures for epidemic peaks and plateaus. The data, predicted results, and hierarchical infection control measures can assist dynamic prevention and control of respiratory nosocomial infections with changes in the infection risk.

Results: The ARIMA model produced exact estimates. The mean absolute percentage errors (MAPE) of the data selected to estimate the time range of the high-risk and low-risk periods were 15.8%, 9.2%, 15.4%, 16.8%, 25.6%. The hierarchical infection control measures included three categories and nine key points. A risk-period judgment matrix was also designed to connect the surveillance data and the hierarchical infection control measures.

Conclusion: Through a mathematical model, dynamic prevention and control of respiratory tract infections in outpatients was constructed based on the daily medical service monitoring data of hospitals. It is foreseeable that when applied in medical institutions, this method will provide accurate and low-cost infection prevention and control outcomes.