SOFIAbot: chatbot for expanding health services during the COVID-19 pandemic

Abstract

COVID-19 has proven to be a daunting challenge to healthcare systems and, in response, new strategies, such as telehealth, were implemented to expand healthcare services using technologies such as chatbots. The objective of this article is to describe access to the SOFIA Bot chatbot system for telescreening suspected cases of COVID-19. This is a Research, Development and Innovation (RD&I) study conducted at the Telehealth Center of the University Hospital of the Federal University of Maranhão. The SOFIA Bot was developed as part of an automated digital platform using algorithmically structured dialogues. Based on the symptoms reported by those individuals who access the tool, it provides guidelines and behaviors to be adopted by the service user, classifying them according to the risk of having COVID-19 and the severity of the symptoms. The resulting data were managed by the Teleconsultation Monitoring and Management System. SOFIA Bot recorded 2,519 hits with 27.9% classified as having symptoms of high, 30.6% medium, and 41.5% low severity. The risk of having COVID-19 was low at 52.8%, medium 35.1%, and hgh risk at 12.1%. The SOFIA Bot was accessed more than once by more than a quarter of the individuals, with the higher frequency coinciding at the peak of the pandemic, with a period of greater transmissibility.