Behavioral data to map COVID-19 risk
The Covid-19 pandemic has thrust the world into the depths of an unprecedented crisis. As the virus continues to ravage human lives, it is also destroying livelihoods. The magnitude of the public health and economic challenges facing governments worldwide is immense. Given the speed at which the virus reproduces, policy makers urgently need more data to help them track and understand how citizens are exacerbating or taming its spread.
The only way countries will be able to navigate the difficult reopening process is if they know how their policy decisions are actually affecting the risk of spread. This boils down to simple human behavior: how much people interact with each other and protect themselves. Today, there’s too little data on this and governments are often in the dark. That’s why we decided to use our platform which enables us to reach millions of people around the world to ask them about their daily habits, and to start creating a real-time dataset that sheds light on contagion risks as countries attempt to reopen step by step.
In a first run, we surveyed citizens of ten countries to understand how their behaviour may be putting them at heightened Covid risk. Our initial results show that Russia, China and the United States face the highest risk, followed by Germany, France, the UK, Canada, India, Mexico and Spain.
We calculate the COVID risk in each country as a score on a scale of 0-100, where 100 represents the highest possible risk. It is compiled based on the results of seventeen survey questions, falling within four broad categories. The metrics are detailed below.
1. Interaction: How much contact people are having with others?
Measured by the extent of ‘face-to-face conversations,’ ‘extent of close contact,’ ‘household size,’ and ‘time in crowded places.)
China scores the highest on the ‘Interaction’ metric. This is likely due the easing of lockdown and social distancing measures as well as larger household sizes. It will be important to track if this loosening social interaction leads to a resurgence of the virus. Ideally, this metric could be expanded to encompass regional granularity. As lockdown is eased in countries like China, this would shed light on any potential second waves, identifying where they might be brewing based on human contact indicators.
2. Vulnerability: Are people protecting themselves?
Measured by how often people ‘wash their hands,’ ‘cover their mouths,’ ‘avoid touching their face,’ or ‘wear a mask’ when in public.
The United Kingdom scores the highest in terms of ‘Vulnerability.’ In other words, UK citizens are least likely to take protective measures like washing their hands after being in public. With the second-highest mortality rate in the world after Spain (891 per million), the expansion of this research would elucidate to what extent this public behaviour is causing the higher death rate. Additional data would allow us to map how this relates to certain demographic groups which have higher death rates (in the UK or anywhere in the world.)
3. Non-Adherence: Are people following COVID government guidelines?
Measured by adherence for government measures including ‘bans on social gatherings of 5 or more,’ ‘closure of non-essential shops,’ and general ‘government Covid guidelines’ (ie. social distancing.)
Russia scores highest on the ‘Non-Adherence’ metric, followed by Germany, France and the United States. The expansion of this dataset to measure to what extent public trust or disillusionment in government is leading citizens to take undue risks, such as by flouting lockdown. This is another metric that would be interesting to examine regionally to see if it correlates to any partisan rhetoric on the pandemic, for example in the Red versus Blue states of the United States.
4. Misinformation: Can people distinguish Covid fact from fiction?
Measured by whether people believe that ‘sun exposure,’ and ‘masks’ protect from the virus; if the ‘the incubation period lasts 2 weeks,’ that ‘holding one’s breath is a sign of being virus free,’ or that ‘mosquitoes transmit the virus.’
The countries that rank highest on misinformation are India, Russia, and the United States. Perhaps this is no surprise given that communications in all three countries have been muddled from the top. In India, members of the ruling BJP have claimed that ‘cow’s urine’ can cure the virus. In Russia, President Vladimir Putin has claimed the country is ‘past its peak,’ despite documenting its highest daily death toll. In the United States, President Donald Trump suggested that ‘injecting disinfectant’ may cure the virus, and that it would ‘disappear one day’ like a miracle. Given the dangerous information ecosystem around Covid-19 (it literally kills), it is interesting to note that these countries rank highly on this metric. The expansion of this research would allow us to track how and where mis- and disinformation is costing lives.
Next steps: A global early-warning system?
This study provides a mere snapshot of how public behaviour might be exacerbating Covid risk in a small selection of countries. However, given the universal nature of this pandemic, we see urgent value in building a behavioural dataset to track risk globally. It should also have a sufficiently large sample size to be able to drill down into regional and demographic differences within countries. This data could be then correlated to actual Covid-19 infection rates to help develop an early warning system.
These data are becoming increasingly important given that the current policy responses – composed of blanket lockdowns and drastic social distancing measures — cannot continue indefinitely. First, because the public may start to lose faith in them (as seen in the violent anti-lockdown protests in the United States and elsewhere), and secondly, because the economic consequences may spiral even further out of control. Ultimately, these data are vital as this pandemic may last a long time yet and we need to build more agile and tailored responses to the challenges ahead.
For more information contact:
Insights Innovation Lead
christoph.doelitzsch(at)daliaresearch(dot)com or covid-19(at)daliaresearch(dot)com