Building a sanitary vulnerability map from open source data in Argentina (2010-2018).

Area of Science:

• Public Health
• Geospatial Analysis
• Data Science

Background:

• Granular data is crucial for effective public health policy.
• Official health statistics often lack spatial detail.
• Sanitary Vulnerability metrics can identify areas needing attention.

Purpose of the Study:

• To develop a novel methodology for measuring Sanitary Vulnerability.
• To create a high-resolution Vulnerability Map.
• To identify areas with significant deficits in population health coverage.

Main Methods:

• Utilized Census, public health facility listings, and crowdsourced georeferenced data.
• Employed dimensionality reduction techniques like Autoencoders and Non-parametric PCA.
• Constructed a Sanitary Vulnerability Index.

Main Results:

• Generated a high-resolution map displaying the geographical distribution of the Sanitary Vulnerability Index.
• Successfully integrated official and crowdsourced open data.
• Overcame limitations of lacking official local-level health indicators.

Conclusions:

• The Vulnerability Map serves as a valuable tool for place-specific policymaking.
• The map's utility was validated by predicting local health coverage metrics.
• Future work includes exploring disease prevalence interactions and applying the methodology to other public services.