A pandemic is characterised by the development of severe unknown diseases in many people, spreading over several regions. The current COVID-19 pandemic has brought new challenges for the entire health sector. Health insurance companies did not have any cover expenses relative to pandemic diseases. They use protective clauses since the risk is too complex as well as high. The current pandemic forced insurers to study future coverages for that hand of risks. The goal of this project is to study and estimate the associated costs of medical treatments for future pandemics, using the current COVID19 pandemic as a good example. At the beginning of the pandemic, the own company decided to cover hospital admissions for COVID-19, tests performed to track the disease, and IPEs, which began to be used more frequently due to pandemic disease.

Therefore, to estimate the cost of a pandemic, we considered these three variables - hospitalisations, tests, and IPEs - using the data related to the COVID-19 pandemic. The first challenge of the project was to estimate the number of hospitalisations that the insurer would have to support, taking into account the incidences, the hospitalisation rate, and the limit of available beds. In this sense, we used a simulation procedure. It was possible to know how many beds were occupied each day by patients, and thus know how many of the hospitalisations were estimated to have is that the company would pay them. It is impossible to predict the characteristics of a future pandemic, so in addition to the characteristics of COVID-19, we have also carried out stress tests to evaluate more extreme situations. The time between pandemics has been decreasing. This is mainly due to three factors: climate change, urbanization, and globalisation. For that reason, the last part of the project was to infer the risk for future pandemics, focusing mainly on the impact of the duration of a pandemic and the time between pandemics in which we used two methodologies: Poisson processes and Extreme Value Theory.

Find the Q&A here: Q&A on 'Measuring Health Outcomes'