Optimizing Public Healthcare Cost Recovery with R

A Use Case from Argentina

Ariana Bardauil

Buenos Aires City Government

Dacio Martinez

Buenos Aires City Government

Information and Health Statistics Management Office

Ministry of Health of the City of Buenos Aires

Information and Health Statistics Management Office

Ministry of Health of the City of Buenos Aires

Argentina’s Healthcare System

Argentina’s Healthcare System

SUMAR Program

A Result-Based Health Policy

SUMAR is a national policy in Argentina aimed at improving financing for the uninsured population through a result-based funding model.

Financed with World Bank support, it transfers funds to provinces based on:

  1. Identifying people with public coverage alone
  2. Providing essential health services (Basic Effective Coverage – CEB)
  3. Achieving health outcomes tracked by tracer indicators

Infrastructure and Automation in Buenos Aires City

Our Rol in GOGIES

In Buenos Aires City, our team plays a central role in automating SUMAR processes.

We focus on:

  • Identifying patients with priority conditions
  • Verifying services through the Health Information System (HIS)

Our main outputs:

  • PDF invoices (“cuasifacturas”)

  • Weekly Excel reports - Performance indicator summaries

The process involves three key stages:

  1. Enrollment
  2. Detection of Basic Effective Coverage services
  3. Analysis of performance indicators

Tools

R-based, open-source ecosystem for scalable health data processing

agiseR

is an internal R package developed by our team to facilitate access to local databases and automate reporting workflows.

Automated Detection and Reporting

of Health Services for the SUMAR Program

Beneficiaries

The identification of beneficiaries is based on two main national sources: (1) PUCO (Federal Unified Register of Beneficiaries) (2) SUMAR program registry provided by the National Ministry of Health

Impact

Time series of detected services, highlighting the scalability and increasing coverage of the automation process.

Thanks