Global Monitoring of the Impact of COVID-19 Pandemic through Online Surveys Sampled from the Facebook User Base
Preprint
Abstract
Simultaneously tracking the global COVID-19 impact across multiple populations is challenging due to regional variation in resources and reporting. Leveraging self-reported survey outcomes via an existing international social media network has the potential to provide reliable and standardized data streams to support monitoring and decision-making world-wide, in real time, and with limited local resources. The University of Maryland Global COVID Trends and Impact Survey (UMD-CTIS), in partnership with Facebook, invites daily cross-sectional samples from the social media platform’s active users to participate in the survey since launch April 23, 2020. COVID-19 indicators through December 20, 2020, from N=31,142,582 responses representing N=114 countries, weighted for nonresponse and adjusted to basic demographics, were benchmarked with government data. COVID-19-related signals showed similar concordance with reported benchmark case and test positivity. Bonferroni significance and minimal Spearman correlation strength thresholds were met in the majority. Light Gradient Boost machine learning trained on national and pooled global data verified known symptom indicators, and predicted COVID-19 trends similar to other signals. Risk mitigation behavior trends are correlated with, but sometimes lag, risk perception trends. In regions with strained health infrastructure, but active social media users, we show it is possible to define suitable COVID-19 impact trajectories. This syndromic surveillance public health tool is the largest global health survey to date, and, with brief participant engagement, can provide meaningful, timely insights into the COVID-19 pandemic and response in regions under-represented in epidemiological analyses. Significance Statement The University of Maryland Global COVID Trends and Impact Survey (UMD-CTIS), launched April 23, 2020, is the largest remote global health monitoring system. This study includes about 30 million UMD-CTIS responses over 34 weeks (through December 2020) from N=114 countries with survey-weights to adjust for nonresponse and demographics. Using limited self-reported data, sampled daily from an international cohort of Facebook users, we demonstrate validity and utility for COVID-19 impacts trends, even in regions with scant or delayed government data. We predict COVID-19 cases in the absence of testing, and characterize perceived COVID-19 risk versus risk-lowering measures. The UMD-CTIS has the potential to support existing monitoring systems for the COVID-19 pandemic, as well as other new, as-yet-undefined global health threats.Keywords
All Related Versions
This publication has 35 references indexed in Scilit:
- Fair Allocation of Scarce Medical Resources in the Time of Covid-19New England Journal of Medicine, 2020
- The global community needs to swiftly ramp up the response to contain COVID-19The Lancet, 2020
- Feasibility of controlling COVID-19 outbreaks by isolation of cases and contactsThe Lancet Global Health, 2020
- Defining the Epidemiology of Covid-19 — Studies NeededNew England Journal of Medicine, 2020
- PHD1 controls muscle mTORC1 in a hydroxylation-independent manner by stabilizing leucyl tRNA synthetaseNature Communications, 2020
- Determinants of Participants’ Follow-Up and Characterization of Representativeness in Flu Near You, A Participatory Disease Surveillance SystemJMIR Public Health and Surveillance, 2017
- Flu Near You: Crowdsourced Symptom Reporting Spanning 2 Influenza SeasonsAmerican Journal of Public Health, 2015
- Economic Impact of Dengue: Multicenter Study across Four Brazilian RegionsPLOS Neglected Tropical Diseases, 2015
- Flutracking provides a measure of field influenza vaccine effectiveness, Australia, 2007–2009Vaccine, 2010
- Non-response and related factors in a nation-wide health surveyEuropean Journal of Epidemiology, 2001