A Patient Self-Checkup App for COVID-19: Development and Usage Pattern Analysis (Preprint)

Abstract

BACKGROUND Patients around the world do not have clear guidelines on the steps to be taken if COVID-19 infection is suspected. Many countries rely on self-assessment via the phone, but this only adds to the burden on the already overwhelmed healthcare system. In this study, we develop an algorithm that will help with the screening and provide patients with guidance. OBJECTIVE The aim of this study is to make decision making easier for the general public by developing a mobile application that will enable them to decide when to seek timely medical care. METHODS The algorithm was developed by consulting six physicians who are directly involved in the process of screening, diagnosis, and/or treatment of COVID-19 patients. The main focus in developing the algorithm was when to test the patient, under the limitation of laboratory capacity. The application was deployed on the web and designed to be mobile-friendly. Google Analytics was embedded and to collect usage data from March 1, 2020, to March 27, and the data were correlated with COVID-19 confirmed cases, screened cases, and death counts by the access location. RESULTS Epidemiological factors, fever, and symptoms were used in the algorithm. The application is deployed on the web, https://docl.org/ncov/. Total of 96,972 users assessed the application 128,673 times during the study period. Without any advertisement, almost half of the access was from outside of Korea. The number of people confirmed was highly correlated to the number of users (rs = 0.82, p-value < 0.0001). And the number of people deceased due to COVID19 was moderately correlated to the number of users. (rs = 0.77, p-value 0.0001). Even though the digital literacy of the 60s were half of that of the 50s, the user count was similar in our application. CONCLUSIONS Expert opinion-based algorithm and mobile application for patient screening and guidance can be beneficial in a circumstance where there is not enough information yet on the novel disease, and medical resource allocation is crucial.