Abstract: Objective To analyze the operational performance of the Guangzhou Student Health Monitoring System and evaluate its early warning effectiveness, thereby providing an evidentiary basis for the further optimization of the system's predictive capabilities. Methods An analysis was conducted on the characteristics of the alert signals generated by the Guangzhou Student Health Monitoring System and its efficacy in the early warning of infectious disease-related public health emergencies on school campuses. Furthermore, a cross-correlation analysis was employed to investigate the time-lag relationship between two data sources: the early warning signals produced by the student monitoring system and the corresponding case reports from the statutory infectious disease reporting system. Results From January 2023 to July 2024, the system generated a total of 61 555 early warning signals, achieving a timely response rate of 97.54% (60 041/61 555), a signal accuracy rate of 98.25% (60 480/61 555), and an overall positive predictive value of 28.15% (17 325/61 555). For campus-based infectious disease public health emergencies, the system demonstrated a sensitivity of 90.59% (77/85) and a timeliness rate of 87.06% (74/85). At a weekly aggregation level, the warning signals and the corresponding reported infectious disease data exhibited a significant positive correlation (P<0.01). The cross-correlation analysis revealed that alert signals for rash-related symptoms and Hand, Foot, and Mouth Disease demonstrated a temporal precedence, enabling the detection of fluctuations in the incidence of these target diseases 1 to 3 weeks in advance. Conclusion The Guangzhou Student Health Monitoring System demonstrates substantial concordance with the statutory notifiable infectious disease reporting system, providing a robust foundation for the early identification and warning of infectious disease outbreaks among the school-aged population. System performance may be further enhanced by strategically adjusting parameters such as warning thresholds and improving the quality of data reporting to increase the positive predictive value of the alerts.

Key words: Student, Symptom surveillance, Public health emergency, Early warning, Evaluation