Antibiotics are one of the most important classes of medical drugs. Thanks to their discovery and widespread use, bacterial infections that used to be fatal are now treated in a few days. However, these advances came at a cost. When exposed to antibiotics, bacterial populations can quickly become resistant to them and infections caused by antibiotic resistant bacteria are an increasingly serious problem.

The best way to prevent the evolution of new resistances is to use antibiotics only when necessary.

Through an FCT grant, we are collaborating with the Portuguese Ministry of Health (through the SPMS) and analysing a large database of medical prescriptions to ask: 

  • What should be considered antibiotic overprescription?
  • And are medical doctors overprescribing antibiotics?
  • What would be the overall effect of reducing chronic illnesses related with antibiotic prescription?

We have characterized the distribution of antibiotics prescription by medical doctors and identified some of the factors underlying increased prescription rates. We proposed a behavioural intervention to reduce over-prescription that we hope to test in the near future. 

We are also using Natural Language Processing tools to identify groups of patients at a higher risk of antibiotic prescribing and simulate the impacts of reducing disease burden on overall antibiotics consumption.


Fundação para a Ciência e Tecnologia:

  • DSAIPA/AI/0087/2018ADD (Projeto Piloto em Ciência dos Dados e Inteligência Artificial na Administração Pública)


Starting Date: October 2017
Ending Date: March 2022 (predicted)