Spatial clustering of average risks and risk trends in Bayesian disease mapping

Area of Science:

• Epidemiology
• Biostatistics
• Geographic Information Systems (GIS)

Background:

• Spatiotemporal disease mapping estimates disease risk across geographic areas over time.
• Identifying areas with high disease risk or increasing trends is crucial for public health interventions.
• Existing research primarily focuses on purely spatial clustering, with limited spatiotemporal approaches.

Purpose of the Study:

• To develop and present a new statistical modeling approach for clustering spatiotemporal disease risk data.
• To cluster areal units based on both average disease risk levels and temporal trend behaviors.
• To provide a tool for more effective identification of disease hotspots and emerging risks.

Main Methods:

• A novel modeling approach for spatiotemporal disease risk clustering is proposed.
• The method clusters areas by considering both mean risk levels and temporal trend patterns.
• Methodology efficacy is validated through a simulation study.

Main Results:

• The new methodology effectively clusters spatiotemporal disease risk data.
• Simulation studies confirm the efficacy of the proposed clustering approach.
• Application to respiratory disease risk in Glasgow demonstrates practical utility.

Conclusions:

• The developed spatiotemporal clustering model offers a significant advancement in disease mapping.
• This approach enables more precise identification of areas requiring public health attention.
• The methodology is applicable to various diseases and geographic settings for improved health surveillance.