Estimating the optimal population upper bound for scan methods in retrospective disease surveillance

Area of Science:

• Epidemiology
• Public Health
• Biostatistics

Background:

• Effective disease surveillance relies on accurate identification of disease patterns and clusters.
• The circular scan method is a common technique for detecting disease outbreaks but its performance depends on the population upper bound.
• Current methods for setting this bound, including a default 50% and a Gini coefficient approach, have limitations.

Purpose of the Study:

• To introduce and evaluate the elbow method for selecting a population upper bound for the circular scan method.
• To improve the performance of disease cluster detection compared to default and Gini coefficient methods.
• To assess the sensitivity and positive predictive value of different population upper bound selection methods.

Main Methods:

• The study evaluated three methods for setting the population upper bound in the circular scan method: the default 50% value, the Gini coefficient method, and the proposed elbow method.
• Performance was assessed using publicly available benchmark data.
• Key performance metrics included sensitivity and positive predictive value.

Main Results:

• The elbow method demonstrated improved performance over the default population upper bound.
• The elbow method also showed advantages compared to the Gini coefficient method in certain aspects.
• Specific improvements in sensitivity and positive predictive value were observed with the elbow method.

Conclusions:

• The elbow method offers a novel and effective approach for determining the population upper bound in circular scan analysis.
• This method enhances the accuracy and effectiveness of disease outbreak and cluster detection.
• The findings suggest the elbow method is a valuable tool for public health surveillance and epidemiological studies.