[Study on early warning threshold values for 7 common communicable diseases in Gansu province, 2016]

Area of Science:

• Epidemiology
• Public Health Surveillance
• Biostatistics

Background:

• Effective communicable disease surveillance is crucial for public health.
• Gansu province faces challenges in optimizing early warning systems for various infectious diseases.
• Existing warning thresholds may not be optimally sensitive or specific.

Purpose of the Study:

• To optimize warning threshold values for common communicable diseases in Gansu province.
• To improve the early warning effectiveness and public health response.
• To evaluate different statistical methods for threshold determination.

Main Methods:

• Utilized moving percentile and cumulative sum methods to establish early warning models for influenza, scarlet fever, infectious diarrheal diseases, dysentery, typhoid/paratyphoid, viral hepatitis E, and hand foot and mouth disease (HFMD).
• Employed sensitivity, specificity, predictive values, Youden' index, and receiver-operating characteristic (ROC) curves to select optimal threshold values.
• Calculated and compared performance metrics for various threshold settings.

Main Results:

• Determined optimal early warning boundary values for influenza (P(90)), scarlet fever (P(80)), other infectious diarrheal diseases (P(95)), dysentery (P(90)), typhoid/paratyphoid (P(80)), and viral hepatitis E (P(90)).
• Identified optimal parameters for HFMD as k=1.2 and H=5σ.
• Achieved high sensitivities (86.67%-100%) and specificities (62.22%-100%) with the optimized thresholds, indicating improved early detection capabilities.

Conclusions:

• Recommended adjusting early warning thresholds for influenza, other infectious diarrheal diseases, dysentery, and hepatitis E in Gansu.
• Suggested specific parameter adjustments for HFMD early warning systems.
• The optimized thresholds and parameters significantly enhance the effectiveness of communicable disease surveillance and early warning systems in the region.