A Generalizable Evaluated Approach, Applying Advanced Geospatial Statistical Methods, to Identify High Lead Exposure...

Area of Science:

• Environmental Health
• Geospatial Analysis
• Public Health

Background:

• Despite progress, childhood lead (Pb) exposure persists in the United States.
• Identifying high-prevalence areas for elevated blood lead levels (EBLLs) is crucial for public health interventions.

Purpose of the Study:

• To develop a generalizable geospatial approach for identifying and analyzing locations with high prevalence of children's EBLLs.
• To assess existing lead exposure models as surrogates for blood lead data using a Michigan case study.

Main Methods:

• Utilized blood lead level (BLL) data from Michigan children (2006-2016).
• Employed geospatial cluster analysis (Getis-Ord Gi*) and a top 20 percentile method to identify high %EBLL census tracts.
• Evaluated convergence of identified hotspots with existing lead exposure models and sociodemographic data.

Main Results:

• Identified 11 primary EBLL hotspots via cluster analysis and 80 additional locations using the percentile method.
• Demonstrated data representativeness (0.93 correlation) and consistency (Kappa score) of EBLL rates.
• Found convergence between many hotspots and current Pb models, but also identified areas with divergence, indicating additional Pb sources.

Conclusions:

• Confirmed known and revealed new, finer-resolution childhood lead exposure hotspots in Michigan.
• Validated the utility of geospatial methods and surrogate models for identifying at-risk areas lacking direct blood lead data.
• The developed approach is applicable to other states for informing lead mitigation and prevention strategies.