The countries that fought COVID-19 with data and science
Around the world, right now, governments are optimistically gearing up for the implementation of the COVID-19 vaccine, creating campaigns and setting up vaccination centres.
With an incredible medical intervention like this on the horizon, the future of our personal lives and economies is looking the brightest it has all year. But as game changing as a vaccine is, it cannot work in isolation – it needs to be complemented with rapid, regular testing and, crucially, with data gathering and AI-driven analysis that keeps us one step ahead of the virus.
The power of numbers
This Summer, The New York Times published a story about William Farr, a 19th Century British doctor and statistician. Farr fought a deadly cholera outbreak in London by going door to door, tracking and tracing cases, and using this information to identify the source of contamination.
Technology has advanced in unimaginable ways since Farr’s day – we can watch global COVID-19 case updates rise and fall on our desktops – but the field that he pioneered, now known as epidemiology, is – at its core – the same.
It means fighting viruses with numbers and science: counting cases. discerning patterns, and making predictions.
With those predictions, we can make life saving choices. In the absence of a vaccine or miracle drug, data has – so far – been our most powerful tool in tackling COVID-19, especially when coupled with scientific, evidenced-based approaches. Data lets us see where the disease is spreading and how quickly. It allows us to identify hot spots, and with technology, warn people where those are and how they are changing in real time. It means that we can predict where healthcare resources are vitally needed, and get prepared.
The countries that got it right
The better we can forecast with data, the better equipped we are. Just take a look at the countries that sat up and paid attention to the numbers in the early days of the outbreak.
New Zealand is a great example, says Amir Banifatemi, Chief Innovation and Growth Officer at XPRIZE, praising the success of the country’s systematic contact tracing efforts and public health decisions based on evidence. As coronavirus infections reached the country earlier this year, the government quickly turned to non-pharmaceutical interventions, imposed a strict lockdown and radically upped testing capacity. Along with clear leadership, this meant New Zealand was able to become virus free in just three months.
New Zealand offers an impressive model for how the rest of the world could do things, but its strong economy, small population, and the fact that it’s an island were on its side. Taiwan, in a similar situation, dealt with the virus similarly successfully.
The outbreak was expected to be bad, due to proximity and go-between flights with China, but Taiwan learned from the 2003 SARS outbreak and married this knowledge with data. Their strategy involved integrating data from immigration and customs databases with national health insurance databases to get real-time information on potential infections based on travel history and symptoms. Their approach raised some concerns about privacy, but it also worked: Taiwan has been lauded as having one of the best global responses, with seven deaths to date.
In the UK, while the government has been widely criticised for poor handling of the virus, there was one excellent example of a strong, data-backed response: the NHS track and trace app.
“Oxford University’s Big Data Institute worked with government officials to explain the benefits of a mobile app that could provide valuable data for an integrated coronavirus control strategy,” explains Forbes. “Since nearly half of all coronavirus transmissions occur before symptoms occur, speed and effectiveness to alert people that may have been exposed are paramount during a pandemic such as coronavirus.”
Now, UK citizens can download the app, opt in to sharing location, and receive notifications when they’ve come into contact with the virus.
As for America, back in January, when President Trump said he was “not worried” about coronavirus, epidemiologists in San Francisco were diligently watching rising cases elsewhere in the world. They drew on their experience of dealing with HIV and, as described in Wired, predicted how the virus would explode, urging city officials to close public schools and lock down the city, which they did. “Tech's money and power depend on trust in data and respect for science,” writes the journalist, drawing a parallel to the approach the city took to coronavirus. Compared to New York City, San Francisco’s COVID-19 rates have been incredibly low.
Huge room for improvement
Things have moved on since these countries and cities implemented their initial COVID-19 responses. Many nations that got a hold on the virus are reopening business, doing all that they can to avoid a resurgence of the virus. Elsewhere in the world, second lockdowns are underway – at great cost to people’s mental health, and to economies.
Whatever the status of a country right now, Banifatemi expresses that data is still crucial in continually tracking and predicting what will happen next. It is the key to navigating a smooth path to reopening. However, when it comes to how we process and actually use this data, there is still huge, huge room for improvement.
Enter XPRIZE’s Pandemic Response Challenge. Launched with Cognizant, a world leader in data and AI technologies and services, the challenge calls for teams to create a groundbreaking AI system that will allow regional governments, communities, and organizations to minimize harm when reopening their economies.
It is based on our belief that technology can be used to advance data-backed approaches to be even more accurate and even more efficient. The idea that, although it may not exist yet, there is a system out there which can help us overcome this public health crisis.
In four months time, the $500K prize will be awarded to the team of data scientists, engineers, developers, or designers whose AI stem can successfully test “what if” intervention plans. An AI system that optimizes COVID-19 mitigation strategies while reducing the economic and social impacts, allowing us all to be safer as we get our lives back on track.
“No-one is exactly doing this on a unified and global scale,” Banifatemi explains. “We are creating a challenge for it so that all communities may have access to predictive models based on factual and evidenced-based data. So that we can rely on data science and AI to create models that can be rapidly tested and correlated to the reality of implemented policies.”
These non-pharmaceutical interventions are critical for reopening plans, he concludes. This is the urgent tech we need to move forwards.
A safer, smarter future
An AI system that could be rolled out on a mass scale would help more local governments use what we already know about COVID-19 to work out their next moves. “With many countries facing epidemic resurgence, evaluating the impact of different strategies implemented in the early phases of the pandemic is crucial for developing an effective long-term response,” The Lancet urged, in a review of what New Zealand got right.
A more accurate prediction model has the powerful potential to not just ensure our safety right now, but with other, future pandemics, too. As The Johns Hopkins University epidemiologist Caitlin Rivers put it to The New York Times this Summer, we have weather forecasting, so why don’t we have epidemic forecasting?
By using data in the most effective way possible, we can control pandemics, rather than letting them control us. The most effective way is through AI. While humans might not be able to apply and analyze data at the speed of computers, we as humans do have the ingenuity and the innovation to create the system that can do this.
So far, COVID-19 has seen 55 million documented cases of the virus and 1.3 million recorded deaths. What if in the near future, this could be prevented, and we could go about our lives more freely and healthily? What if a powerful AI system could be implemented far and wide in order to flatten curves before they even happen? What if “what if” wasn’t a question anymore, because we knew the answer?