Modifiable Areal Unit Problems for Infectious Disease Cases Described in Medicare and Medicaid Claims, 2016-2019

Area of Science:

• Spatial analysis
• Epidemiology
• Geographic Information Systems (GIS)

Background:

• Modifiable Areal Unit Problems (MAUP) introduce spatial uncertainty.
• The effect of MAUP on infectious disease dynamics and epidemic detection remains unclear.

Purpose of the Study:

• To investigate the impact of different geographic units on infectious disease surveillance.
• To compare epidemic peak detection using states versus mega-regions.

Main Methods:

• Extracted CMS claims data (2016-2019) with infectious disease codes (SNOMED CT).
• Analyzed data at state and mega-region levels using per member per month metrics.
• Employed rolling averages above the series median for peak detection.
• Utilized spatial random forest for agent-location discrimination and region segmentation.

Main Results:

• Mega-regions demonstrated superior peak discovery for most infectious agents compared to states.
• Spatial random forest analysis revealed significant differences in agent-location discrimination between geographic units.

Conclusions:

• The choice of geographic unit significantly influences infectious disease surveillance outcomes.
• Researchers must justify their selected geographic unit of analysis on an agent-by-agent basis in publications.