What is it about?
Colombia spends public money on medicines. Those contracts sit on SECOP-II, the country's public procurement portal, but the wording is messy: one purchase is hard to compare with another. This paper turns that noise into one comparable picture for 2020–2025: 161,830 pharmaceutical contracts. Two findings stand out. The top 3% of suppliers account for about 86% of the contracted value. And 664 contracts look unusual within their medicine group. That is a reason to review them first, not a verdict. A statistical alert is a starting point. It is not proof of fraud. Data and code are on Zenodo.
Featured Image
Photo by Kanchanara on Unsplash
Why is it important?
When a country buys medicines with public money, people should be able to see the pattern: who supplies most of the value, and which contracts look unusual. This paper turns Colombia's published contracts into that picture for 2020–2025. Readers can use it as a shared set of indicators — and as a shortlist of contracts that deserve a second review. An alert is not proof of irregularity. It is a reason to look again, a queue for another reviewer. The same readable picture is what you need before you digitize purchasing, inventory or payment. This article does not install that system. It makes the conversation possible, with evidence instead of slogans.
Perspectives
Ukraine's ProZorro showed that government purchasing can be made readable. This paper is that kind of photograph for Colombian medicines — not a live marketplace. The hard work today is untying contract language and reading three public sources together: purchasing, drug registration, and reference prices, always with the date of the data in view. Linking a registered product to a contract helps you read. It does not tell a hospital when to reorder stock. Tokenized inventories and automatic payment appear in the article only as a lab prototype, under simulated conditions. They are not running in the real world, and they did not produce savings we observed. The rule I will not bargain: a signal is not proof.
Andrés Soto
Read the Original
This page is a summary of: Computational surveillance of Colombian public pharmaceutical procurement using public administrative data: A reproducible analysis of a closed 2020–2025 cohort, PLOS One, September 2026, PLOS,
DOI: 10.1371/journal.pone.0350967.
You can read the full text:
Resources
Contributors
The following have contributed to this page
