Antibiotic resistance of pathogenic bacteria has historically arisen in parallel with the development of new antibiotics; this race posing a major health problem worldwide where bacteria seem to be winning. A paradigmatic example is methicillin-resistant Staphylococcus aureus, which can cause severe infections in the bloodstream and the lung. Methicillin-resistant Staphylococcus aureus is a bacteria species that, after developing resistance against penicillin, has become resistant also against a second antibiotic, methicillin. Development of resistance against antibiotic can occur due to antibiotic pressure, where non adequate prescription policies play a fundamental role (for example, if broad-spectrum antibiotics (which are not specialised for fighting against the particular bacteria involved in the infection) are used instead of narrow-spectrum ones (which are specialised in the bacteria participating in the infection)). This is one of the reasons for drug resistant bacteria being a particular challenging problem in healthcare facilities, together with other reasons such as the presence of aged individuals with weaken immune systems. The problem of the presence of drug resistant bacteria in healthcare settings has taken the next step by their spread in the community (non-healthcare environments). This has yielded new strains which are able to cause severe infections in healthy individuals.

In order to avoid the emergence and spread of drug resistant bacteria in healthcare settings, different strategies are usually followed: appropriate antibiotic prescription policies, management of staffing levels, isolation of infected patients, compliance of hygiene procedures,... However, most of healthcare settings usually follow a combination of these procedures, and the individual efficacy of each of them is hard to measure. This quantification is important not only due to the scarcity of resources in these clinical environments, but also because some of these policies entail moral and ethical problems. Mathematical models have proven to be a robust tool for addressing the efficacy of these individual strategies, as well as for identifying the factors involved in the emergence and spread of drug resistant bacteria in healthcare settings.

In the research project that I am carrying out funded by the Medical Research Council, the aim is to contribute to the mathematical modeling in the area, in order to answer questions not solved yet. In particular: which is the importance of some factors, such as the contamination of the healthcare setting environment (for example, equipment), in the spread of resistant bacteria in healthcare settings? How does this spread occur in different healthcare settings (for example, in hospitals vs. nursing homes)? Which is the impact caused by the existing heterogeneities among individuals within the healthcare settings (healthy individuals, such as healthcare workers, vs. moderate or severe ill patients; adults vs. children; patients under antibiotic treatment,...)?

The emergence and spread of drug resistant bacteria is a major problem worldwide. However, due to financial reasons (for example, some antibiotics newly developed are only effective for a few years, with the subsequent development of new drug resistant bacteria strains) the number of pharmaceutical companies working in new antibiotics development is scarce, and governmental financial incentives are usually required. Moreover, it is worth noting that, in Europe, it has been estimated that infections with drug resistant bacteria cause around 25000 deaths per year, with an estimated cost of 7 billion pounds in direct medical costs. Thus, it is necessary to combine the development of new antibiotics with control intervention measures for avoiding the emergence and the spread of drug resistant bacteria among healthcare settings. Mathematical modelling will help to understand the dynamics by which drug resistant bacteria emerge and spread in healthcare settings, which is at the same time crucial to implement intervention strategies based in quantitative knowledge.