Spatial Scan Statistics in Epidemiological Analysis

Spatial scan statistics constitute a robust framework for detecting and evaluating disease clusters in geographical and spatiotemporal data without prior assumptions on their location, size or shape. By moving a scanning window across a study region and comparing observed case counts within the window to those expected under a null model, this approach identifies areas exhibiting unusually high or low disease incidence. Extensions accommodate space–time dynamics, enabling early outbreak detection and monitoring changes in cluster geometry over time. Applications span infectious diseases, chronic conditions and environmental exposures, informing resource allocation, guiding targeted interventions and elucidating patterns of health inequity. Recent methodological refinements have addressed sensitivity to scanning parameters, irregular cluster shapes and the reporting of multiple clusters, thereby enhancing both the accuracy and interpretability of results in public health practice.

Research from Nature Portfolio

Recent studies have applied spatial scan statistics to uncover health inequities and inform intervention strategies. In one investigation of mental health care access in a middle-income country, travel-time modelling combined with spatial scan analysis revealed distinct hotspots of severe psychiatric cases, highlighting distance-driven disparities in outpatient and inpatient treatment. Another study at the township level examined influenza incidence over a decade, deploying spatiotemporal scan statistics to detect shifting clusters and temporal turning points of infection. This work demonstrated annual variability in hotspot locations, underscoring the value of fine-scale surveillance for optimising vaccination campaigns and resource deployment.

Research from all publishers

Comparative evaluations have advanced understanding of how scanning window shape and parameter choice affect cluster detection. A recent analysis contrasted circular versus flexibly-shaped scan statistics, demonstrating that irregularly-shaped methods often yield larger, higher-likelihood clusters, whereas circular windows offer more regular delineations. A comprehensive review of scan statistics synthesised methodological developments, including alternative probability models for continuous data and enhancements in statistical power for various cluster configurations. Foundational work on reporting strategies introduced the Gini coefficient to select non-overlapping clusters, refining the balance between sensitivity to small hotspots and the summarisation of overlapping cluster outputs.

Spatial Scan Statistics in Epidemiological Analysis publication trend

The graph below shows the total number of articles in spatial scan statistics in epidemiological analysis across all publications each year (not limited to Nature Index journals).

Technical terms

Spatial scan statistic: A statistical method that moves a variable‐size window across space to identify regions with unusually high or low counts relative to expected values.

Scanning window: The geometric shape (e.g., circular, elliptical or flexible) used to define candidate clusters when applying scan statistics.

Log‐likelihood ratio (LLR): A measure comparing the likelihood of observed data under the cluster versus the null hypothesis, used to rank and test candidate clusters.

Space–time scan statistic: An extension of the spatial scan statistic that uses a cylindrical or irregular window to detect clusters in both space and time.

Gini coefficient (in scan statistics): A criterion for selecting the most informative set of non‐overlapping clusters by quantifying heterogeneity among candidate clusters.

References

1. Geospatial investigations in Colombia reveal variations in the distribution of mood and psychotic disorders. Communications Medicine (2024).
2. Heterogeneity of influenza infection at precise scale in Yinchuan, Northwest China, 2012–2022: evidence from Joinpoint regression and spatiotemporal analysis. Scientific Reports (2024).
3. Comparing circular and flexibly-shaped scan statistics for disease clustering detection. Frontiers in Public Health (2025).
4. Using Gini coefficient to determining optimal cluster reporting sizes for spatial scan statistics. International Journal of Health Geographics (2016).
5. An up-to-date review of scan statistics. Statistics Surveys (2021).
6. A flexibly shaped space-time scan statistic for disease outbreak detection and monitoring. International Journal of Health Geographics (2008).