Location Data and Analytics for COVID-19

Location Data Can Help Flatten the Curve

Location data and analytics are not new and have been used for some time. As location data has been available from various sources, advertisers have used location intelligence to target offers to certain demographics, display ads dynamically in billboards and many other such initiatives, even creating new categories as “hyper local marketing”. Location data is also frequently used by law enforcement or city services to keep neighborhoods safe and provide better public services. 

Data is collected from multiple sources including mobile phones, apps, wifi hotspots as inputs to create geo-fenced location analytics heatmaps that can be used to drive various use cases. Many urban areas have also installed extensive collection of CCTV equipment that can be used to feed into various Location analytics applications beyond monitoring and surveillance for crime. Contact less payments and credit card transaction data can also be used for Location based analytics. With the rise of applications such as Apple Pay, Android Pay, AliPay – these applications capture the location of the transaction and can be used as an input for various use cases. 

Why Location Tracking for COVID-19

As the spread of COVID-19 virus is mainly through airborne transmission from infected individuals, location tracking of the infected individuals and their containment can help reduce the spread. Location tracking can be use effectively to contain the spread of COVID-19 by:

1. Identifying the location of the COVID-19 positive cases and containing them with quarantine.
2. Identifying other individuals who may have come in contact with COVID-19 positive cases, providing them with testing and isolation. 

East vs. West

As we consider various approaches for Location tracking, it is important to consider the cases (and deaths) by country or region where location based technologies are being used or not used. The following chart shows daily deaths by Country by number of days since 3 daily deaths first recorded.

Financial Times has created visualizations that show some of the key patterns. The following chart shows the daily deaths by region. 

One key observation is that COVID-19 impact is now mostly in the Western countries – with only 15% of the world population. Asia and the East with the majority of the population has far less number of cases. One of the key reasons is that many of the countries in Asia have instituted aggressive lock downs, performed testing and are using location based tracking. Location based tracking is one of the key reasons for containment of cases. China and South Korea are two countries who have instituted extensive tracking and surveillance programs.

China has been in the forefront of tracking and surveillance to contain the spread of COVID-19. They have instituted a systematic approach of requiring everyone to use the mobile phone app Alipay which is used to  define a status of Green, Yellow and Red based on an individual’s current risk level. After an initial questionnaire, the color coded QR code called Alipay Health code is generated based on the individual’s  travel history, duration in an outbreak area, and relationships to hosts/carriers. Alipay Health Code is checked in public venues – apartments, office buildings, public transit and any other public facilities before letting anyone in. People with Yellow or Red status are not allowed and asked to isolate themselves. 

South Korea has been exemplary in their approach with wide spread testing and tracing past contacts of anyone who has tested positive. South Korea has one of the highest penetration of mobile phones. There are approximately 112 phones for every 100 people. Even though it is of the same area as the state of NY, it has a 50M population (vs. New York’s 19M). It has so far 10,000 cases vs. 220,000 in the state of NY. South Korea has leveraged data from its massive network of CCTV cameras, Contactless Payments from 20+ credit card issuers and Location data from cell phones to trace contacts for anyone who tests positive so that those contacts can also be tested and isolated to avoid further spread, reducing tracking time of potentially infected contacts to 10 minutes. All this data is made available allowing the public to see all the locations of a newly infected person before they were diagnosed.  

Several countries are tapping into data collected by the telecom operators to monitor the movement of individuals who have been quarantined due to travel or exposure to infected individuals. Location data is generated by the phones pinging cell towers and can be used effectively to generate geo-fences to monitor movement of individuals. Heatmaps of infected individuals can also be used to detect areas of hotspots.

In the US, most of the location data efforts for COVID-19 are based on aggregate data collection approach to detect the mobility in various counties/states after stay at home orders and social distancing guidelines have been announced. One example is a company called Unacast that is providing a Social Distancing Dashboard to counties and state governments so that they can see the impact of social distancing  and stay at home orders from the location data being collected from various sources. 

Data privacy is a big concern. Countries like China and South Korea have used the brute force approach with location tracking- not an option. Although most users have location data enabled for apps like Google Maps, Facebook, Instagram and others, when it comes to sharing location data augmented with health information, it is a non-starter in many western countries. Data must be anonymized before it can be used for such applications. 

Contact Tracing

As the United States considers reopening in a phased manner, Contact tracing along with widespread testing will be a key requirement for getting to the “new normal”. Contact tracing as CDC defines it is the ability to “trace and monitor contacts of infected people. Notify them of their exposure.”

Contact tracing when done manually needs an army of people to trace contacts of infected individuals. Disease detectives start by trying to locate the “patient zero” in a given hotspot area, interviewing this person to learn about their movements and identify all their close contacts they have interacted with in the past several days. Based on these answers, they may contact various public venues, review passenger manifests, reservation logs, etc. to compile a list of people who may have been exposed to patient zero. Public health workers call the list of contacts and prompt them for testing or ask them to monitor for symptoms of the infection. This is a tedious process, mostly happens over the phone and can have many gaps. A better option could be based on technology that is already available in our mobile phones.

The new contact tracing approach proposed by Apple and Google is based on an interoperable bluetooth wireless technology called Exposure Notification, enabling public health agencies to create apps to track which phones come in contact(near) with each other. The proposed approach tries to address data privacy by making it voluntary and anonymized. Each phone will generate an anonymized identifier beacon that will be exchanged when it is in close proximity of another phone. When a user tests positive, they can volunteer to report anonymously through the public health agency’s app. This would notify all the other users whose phones had been in close contact during a certain past period of time. Such users will be prompted to get tested or start quarantine. The identity of the user who tested positive will not be disclosed in the process. 

As the US considers reopening and going back to normalcy, having a robust mechanism for contact tracing along with widespread testing will be required for containing the spread of COVID-19. Privacy and transparency will be key for adoption of the proposed Exposure Notification bluetooth based contact tracing. A public health agency such as CDC has to take the lead for creating the app, making it easy to install and configure so that it has widespread adoption – which will be needed for it to work. Using data with location based technologies will help contain the spread of COVID-19.