Dataset Supporting: Effect of Mass Transfers of Enrollees from Liquidated Health Insurance Companies on the Financial Sustainability of Health Insurance in Colombia (2017–2024)

This dataset contains insurer-level panel data for Colombian health insurance companies (EPS) covering the period 2017–2024. The dataset was constructed to evaluate the financial and operational effects of administrative interventions and mass enrollee transfers resulting from insurer liquidations within the Colombian General System of Social Security in Health.

The unit of analysis is the EPS-year observation. The dataset integrates financial statements, regulatory reports, and administrative data from official national sources, including the Ministry of Health and Social Protection (MSPS), the National Health Superintendence (SNS), and ADRES.

The database includes the following categories of variables:

• Financial performance indicators (operating revenue, healthcare costs, gross margin, total expenditures)

• Annual variation indicators (year-over-year percentage changes)

• Technical reserve variables (liquidated pending payments, recognized but not liquidated obligations, reserve releases)

• Enrollee-level indicators and per-enrollee financial metrics

• Medical Loss Ratio (MLR) indicators calculated under MSPS and SNS methodologies

• Administrative intervention indicators

• Derived variables for econometric analysis, including difference-in-differences (DiD) treatment indicators

Per-enrollee variables were constructed using equivalent enrollee counts to ensure comparability across insurers of different sizes. Annual variation rates were computed as year-over-year percentage changes. Intervention indicators were coded based on official administrative resolutions.

This dataset supports panel regression analyses, including difference-in-differences and mixed-effects models, examining the impact of mass enrollee transfers and regulatory interventions on financial sustainability.

All data are aggregated at the insurer level and contain no personally identifiable information.

The dataset is released under a CC BY 4.0 license to promote transparency, reproducibility, and responsible reuse.