Contact tracing: moving from reactive to proactive

Telecoms.com periodically invites expert third parties to share their views on the industry’s most pressing issues. In this piece Richard Baker, CEO of GeoSpock, looks at the various technologies being considered to help with the fight against coronavirus.

Easing the lockdown, while avoiding a second wave, is one of the great challenges of our time. To slow the spread of the virus, we’ve put our lives on hold, with health services and supply chains pushed to breaking point. With new cases expected until at least 2024, a rapid, joined-up approach to controlling the virus and preventing another pandemic is crucial.

Until a vaccine is found, technology holds the key to navigating back to normality and contributing to a coordinated response. Bluetooth-based contact tracing, which uses signals between mobile devices to ascertain which users are close to each other, has already been widely adopted in Asian countries. This is arguably more efficient than traditional contact tracing methods that require large staffs to interview patients about their whereabouts or even knock on the doors of contacts. However, it also has several areas in need of improvement, such as offering more incentives to encourage more people to use the app or tapping on other areas of technology to expand its functionality and efficiency.

To bring rates of transmission under control, it is necessary to shift from a reactionary to a proactive approach to contact tracing, with data-driven decisions at its core.

Bluetooth: a half measure

The tracing apps currently being tried and tested are Bluetooth-based. These are useful in alerting people when those infected cannot recall whom they have been in close proximity with for an extended duration. However, their uptake globally has proven slow, with many raising concerns over data privacy. While such Bluetooth apps do not collect location data, this is exactly the area where contact tracing could be further improved.

Bluetooth-only contact tracing is a reactionary solution, not a proactive one. If you are close enough to catch a Bluetooth signal, you are close enough to have already caught the virus. Nor are they sufficiently accurate. Bluetooth on its own only tells people the ‘who’, and not so much of the ‘where’ – a person travelling on a train could infect many people at opposite ends of the country in a single day. In addition, signals could be received through the windows of an isolation booth and can propagate further than airborne viral spread, providing inaccurate data. Whilst helping reduce onward infections with post-exposure mitigation, it isn’t effective for pre-exposure prevention. This all suggests we should be looking for a better technology solution.

The prevalence of communications technology can provide the answer. When harnessed fully, widespread mobile phone adoption can provide sophisticated environmental sensing. When combined with additional datasets and the ability to carry out rapid large-scale location data analytics, it would provide full-situational contextual intelligence. In turn, this can enable true data-driven decision making and a more coordinated response to the pandemic.

Greater understanding through location intelligence

A step by step approach is at the heart of the government’s plans to reopening of the economy, with non-essential services resuming gradually and when conditions allow. An understanding of the progression of the virus enables low-risk or low-exposure areas and industries to reopen, minimising both short-term and long-term economic disruption. It also allows continuity planning with an early-warning system to alert businesses to regional supply-chain disruptions or unavailability of public transportation services.

Data is vital to be able to understand our invisible enemy’s activity. This can be in the form of close-contact data, such as Bluetooth signals, or fine-grained location data, such as GPS signals from mobile devices or social media check-ins. Another category is coarse-grained location data, which includes government place-of-residency records and network-based location trilateration from telecoms providers using mobile phone towers. Meanwhile, supplementary contextual data can be in the form of COVID-19 testing outcomes, like crowd monitoring via CCTV or even weather conditions.

Location intelligence enables us to understand when and where the virus is progressing in real-time. This is vital for continuous situation management and provides governments with the data necessary to act safely, efficiently. It can underpin everything from healthcare resource management and business continuity planning, to exposure-risk assessment for individuals and track-and-trace activity with granular containment, isolation and travel policies. However, it’s important to remember that flexibility is paramount as the pandemic evolves and as new datasets become available or new outcomes are desired. In addition to the complex nature of the data ecosystem, some of the datasets will be machine generated, which means very large amounts of data will need to be stored and processed for insights and analytics.

Identifying virus hotspots

A highly sophisticated and specialised database is required to cope with these large and complex datasets, while maintaining rapid speed of response. The ability to fuse different data together and extract new insights can guide the design of new policies and guidelines.

For example, while a contact tracing app could alert individuals who have encountered someone who has tested positive for COVID-19, they could also be informed if hyper-localised cases have occurred in their neighbourhood or at their place of work. This would enable them to stay at home or avoid visiting public places to minimise exposure to any risks, without the need for a nationwide lockdown.

Healthcare managers could be alerted at an early stage to any regional growth in infection rates, enabling them to shift medical staff and resources to where they are needed most. Crucially, this can prevent healthcare services from becoming overwhelmed, while supplying frontline workers with the right equipment to keep them safe.

In addition, any nationwide analytics could enable the government to identify virus hotspots and react proactively and appropriately by locking down a town or city as opposed to the whole country. This kind of data-driven decision making, along with real-time monitoring and management of physical world systems, is critical when it comes to improving real-world outcomes.

Speed is the critical. With coronavirus able to spread rapidly, insights must be available instantly for decisions to be effective. The road ahead is uncertain, with globalisation increasing the likelihood that we will face future pandemics. As we move forward, data-driven technology is crucial if we are to navigate a path back to normality for all of us.

Richard Baker serves as the Chief Executive Officer of GeoSpock. With 25 years of experience in the technology sector, he joined GeoSpock to build the next great business to emerge from the Cambridge technology cluster. Prior to this, Richard ran his own successful start-up and has experience across four different industries.