Online symptom checkers are digital health solutions that provide a differential diagnosis based on a user’s symptoms. During the coronavirus disease 2019 (COVID-19) pandemic, symptom checkers have become increasingly important due to physical distance constraints and reduced access to in-person medical consultations. Furthermore, various symptom checkers specialised in the assessment of COVID-19 infection have been produced.
Assess the correlation between COVID-19 risk assessments from an online symptom checker and current trends in COVID-19 infections. Analyse whether those correlations are reflective of various country-wise quality of life measures. Lastly, determine whether the trends found in symptom checker assessments predict or lag relative to those of the COVID-19 infections.
In this study, we compile the outcomes of COVID-19 risk assessments provided by the symptom checker Symptoma (www.symptoma.com) in 18 countries with suitably large user bases. We analyse this dataset's spatial and temporal features compared to the number of newly confirmed COVID-19 cases published by the respective countries.
We find an average correlation of 0.342 between the number of Symptoma users assessed to have a high risk of a COVID-19 infection and the official COVID-19 infection numbers. Further, we show a significant relationship between that correlation and the self-reported health of a country. Lastly, we find that the symptom checker is, on average, ahead (median +3 days) of the official infection numbers for most countries.
We show that online symptom checkers can capture the national-level trends in coronavirus infections. As such, they provide a valuable and unique information source in policymaking against pandemics, unrestricted by conventional resources.
Full manuscript: [TBA]