Now, a mobile app predicts COVID-19 incidence days in advance

#GS3 #Science&Technology

The predictive power of the app highlights the potential utility of real-time symptom tracking

  • By repurposing an existing mobile application used for personal nutrition studies, a multinational collaboration — the COronavirus Pandemic Epidemiology (COPE) — has developed a COVID-19 Symptom Tracker mobile app. 
  • The COVID-19 Symptom Tracker enables people to self-report data related to novel coronavirus exposure, symptoms and infections. 
  • According to a paper published in Science, among the first 1.6 million users (mean age 41 and 75% females) in the U.K. over the first five days of use has confirmed the variability of symptoms in people with suspected infection. 
  • The most common symptoms were found to be fatigue and cough, followed by diarrhoea, fever and loss of smell. 
  • Shortness of breath was relatively rarely reported. 
  • Only 0.4% (1,176) of individuals reporting possible COVID-19 symptoms reported receiving a RT-PCR test for COVID-19.

Predicting incidence

  • The researchers subsequently developed a weighted prediction model based on the symptoms of more than two million individuals who used the app. 
  • And they found that the app can predict the dynamics of COVID-19 incidence a few days in advance of traditional measures, such as positive tests, hospitalisations or mortality.
  • Based on the predictive value of the app, the researchers are planning to validate the symptom-based modelling of COVID-19 incidence by carrying out additional studies using representative sample of individuals who will undergo uniform COVID-19 testing.
  • The predictive power of the app highlights the potential utility of real-time symptom tracking to help guide allocation of resources for testing and treatment as well as recommendations for lockdown or easement in specific areas.

Definitive symptoms

  • Among those who got themselves tested, it became possible to know which symptoms commonly lead to testing, and which symptoms better correlated with a positive test result. 
  • For instance, while cough and fatigue did lead people to get themselves tested, the symptoms do not seem to be particularly suggestive of an infection as revealed by the test results. 
  • But people who had cough and/or fatigue and at least one additional symptom — including diarrhoea and loss of smell — were more likely to test positive, the researchers found.
  • They found that loss of smell was a more sensitive symptom that even fever. 
  • But a greater number of people with a fever tested positive when fever was accompanied by other symptoms.

Need prioritising

  • These findings suggest that individuals with complex or multiple (three or more) symptomatic presentation perhaps should be prioritised for testing. 
  • Concerningly, 20% of individuals reported complex symptoms (cough and/or fatigue plus at least one of loss of smell, diarrhoea, or fever) but had not yet received testing, representing a substantial population who appear to be at greater risk for the disease.
Print Friendly and PDF
blog comments powered by Disqus