Explaining COVID-19 dynamics through user activity data from digital platforms with Yandex's self-isolation index as...

Area of Science:

• Epidemiology
• Data Science
• Public Health Policy

Background:

• Social distancing is crucial for pandemic control but has socio-economic costs and variable effectiveness.
• Accurate measurement of people-to-people contact and policy compliance is needed for effective pandemic response.
• Digital platforms offer potential for real-time monitoring of population behavior during health crises.

Purpose of the Study:

• To explore the utility of composite indices from digital platforms as proxies for people-to-people contact intensity.
• To evaluate the Yandex self-isolation index as a measure of social distancing compliance.
• To test the cointegration between self-isolation indices and COVID-19 case/death data.

Main Methods:

• Utilized Yandex's self-isolation index, aggregating multiple digital activity data sources.
• Developed two error-corrected autoregressive distributed-lag models, inspired by the SIR model.
• Tested for cointegration between the self-isolation index and official COVID-19 data for Moscow and St. Petersburg.

Main Results:

• Found evidence of cointegration between the self-isolation index and COVID-19 case and death numbers.
• Confirmed that the COVID-19 epidemic curve is explainable by people-to-people contact intensity measured by the index.
• Demonstrated the self-isolation index as a reliable real-time indicator of social distancing compliance.

Conclusions:

• The self-isolation index serves as a valuable real-time proxy for people-to-people contact intensity.
• This index can effectively inform policymakers by indicating compliance with social distancing measures.
• Findings support the use of digital platform data for calibrating public health interventions during pandemics.