Visualization of the spatial scan statistic using nested circles.

Area of Science:

• Spatial statistics
• Geographic information systems (GIS)
• Public health surveillance

Background:

• Traditional cluster detection methods may obscure localized patterns within broader regions.
• Visualizing spatial scan statistic results effectively is crucial for public health analysis.
• Prostate cancer exhibits geographic variations in mortality rates.

Purpose of the Study:

• To propose and demonstrate a novel visualization technique for spatial cluster detection.
• To enhance the informational content of cluster detection outputs by integrating multiple risk measures.
• To identify both broad and localized patterns of prostate cancer mortality in the contiguous United States.

Main Methods:

• Application of Kulldorff's spatial scan statistic.
• Simultaneous consideration of likelihood ratio and relative risk.
• Development of a nested or contoured mapping approach.

Main Results:

• The proposed technique generates maps displaying both regional and focal patterns of excess and deficit.
• Application to US prostate cancer mortality data (1970-1994) revealed complex spatial distributions.
• The contoured maps provide complementary information to traditional choropleth maps.

Conclusions:

• The new visualization method offers a more comprehensive understanding of spatial disease patterns.
• This technique can improve the identification of high-risk sub-clusters for targeted interventions.
• Enhanced spatial data visualization is vital for effective epidemiological research and public health decision-making.