COVID-19 Scenarios: evolution and adaptation of a projection tool for contamination scenarios to the Brazilian reality

Brazil currently has more than 3,400 confirmed cases of COVID-19 and an unknown number of unreported cases. With the rapid growth of the epidemic observed over the past month, the country is close to a critical point in which, if it does not offer a quick and incisive response to the crisis, it will no longer be able to suppress the spread of the disease and must proceed with a damage mitigation phase at the cost of many lives. The way the government will act in the coming weeks and how it will manage a possible mitigation scenario in the medium and long term will determine the severity of the epidemic and its impact on health systems. The rapid speed of propagation of COVID-19 imposes a very short window of action, which requires a high degree of coordination and planning by governments. The Brazilian president, refuses to adopt mitigation measures recommended by the World Health Organization, resorting to measures initially adopted in other countries whose COVID-19 scenario was devastating. It adopted a policy of minimizing the problem so reckless that it was cited in an editorial in the prestigious scientific journal Lancet as an example of a mistaken response to the crisis (The Lancet. March 28, 2020. COVID-19: learning from experience. Vol. 395, issue 10229, P1011). There are, however, formal and informal networks that bring together members of the technical staff of the Ministry of Health, PAHO, state health teams and researchers in the field of epidemiology and public health, This project proposes to improve an existing tool that would facilitate the communication of these specialists and that is capable of predicting some impacts of the epidemic on communities. The target application of the project presented here (COVID-19 Scenarios - https://neherlab.org/covid19/), developed by the University of Basel, Switzerland, adopts the epidemiological model SEIR to model mathematically and outline possible scenarios for the development of an epidemic and thus assist in planning to respond to SARS-CoV-2 outbreaks. COVID-19 Scenarios is a web application that is easy to deploy and adopt by potential stakeholders. The proposed objectives are (i) to feed COVID-19 Scenarios with data from Brazil at the municipality, state and metropolitan region level; (ii) improving the model implemented in the application; (iii) create tools for the sharing and simulation of epidemic scenarios by specialists; (iv) complete the translation into Portuguese and produce texts and interfaces suitable for more lay audiences. The development methodology is based on agile methods and distributed development techniques already established in open source communities. Thus, we hope to improve this existing platform to better serve as an interface for presenting the scenarios planned for Brazil for managers and the general public so that at the end, we have a simplified Portuguese interface for presenting the results of the model's predictions. In this way, the platform can serve as a basis for (i) state and municipal governments to make decisions and propose public policies to contain and mitigate the epidemic with the support of plausible scientific data and extrapolations and that (ii) the general public can be informed on the progress of the epidemic and the actions applicable in each region. The project's executive team consists of programmers, biologists, designers and epidemiologists. From the point of view of software development, the development team has members who dominate the main technologies adopted by the project, which are React / Typescript and Python. From the design, technical methodologies will be applied for better understanding and user usability, process mapping and user journey and use of user interface creation software with online communication between team members (Figma or Sketch + Zeplin). In addition, the team has epidemiologists who work with public policy and are therefore ideally trained to raise requirements and propose useful changes to the application's objectives. Finally, several members have simultaneous experience and knowledge in the areas of biology,