The Covid-19 pandemic represents the clearest and most graphic example in our lifetimes of a rapidly emerging exponential threat. The rate at which the crisis doubles in scale outstrips many of our best efforts to understand where the events are leading and the type of response required. The challenge of exponential growth is not merely the eventual scale of the crisis, but, the fact that, at many early points, until the curve starts to bend upward, the direction and speed of the main trends themselves are not immediately apparent. And by then it may be too late.
Steering through the response to an exponential threat requires far closer attention to the details of real-time data than other crises, in part because, unlike disasters such as earthquakes and floods, effective mitigation of an exponential crisis demands careful attention and strategic action, even before the worst effects are visible. Public health actions need to be quick and decisive, more so than at other times, based upon the analysis of available data.
Delivering Medical Aid at Exponential Scale
As the novel coronavirus spreads at exponential speed throughout the world and across all areas of the United States, Direct Relief has had to scale and target its humanitarian medical supplies like no time before in its 72-year history.
Direct Relief began its response in Wuhan, China in the middle of January at the request of a small number of concerned Chinese medical school alumni. Since then, Direct Relief has extended its network of support from a handful of institutions in one city to thousands of institutions across dozens of countries in only a few short weeks. Starting with the goal of improving access to personal protective equipment (PPE) to keep frontline health workers safe, Direct Relief has expanded its scope of support. This support expansion includes essential medications for intensive care units, medical aid for health centers and clinics experiencing unprecedented medical pressures, and tens of millions of dollars in grant resources to keep the U.S. health care safety net functioning. Starting with only the basics of case data, Direct Relief has had to bring dozens of different datasets together to forecast constantly changing needs for medical support.
Building this massively scaled humanitarian aid flow at great speed has been facilitated by data analytics from Qlik. Whereas Direct Relief routinely runs core business analytics in QlikView for the operation of its humanitarian warehousing and distribution operations, there is now more need for connecting these operational views with epidemiological analysis of case spread, risk factors and the uneven impact on medical infrastructure. There is constant concern about where the outbreak could spread and the effects on what are likely to be vulnerable populations and already overburdened health systems.
The Direct Relief Qlik Sense Covid-19 App
In the early days of the U.S. outbreak, Qlik’s Julie Kae reached out to Direct Relief to understand what they could do to help. The answer was clear and clean data, in copious amounts.
Almost immediately, a team of data engineers began to work with Direct Relief’s Research and Analysis team to determine what questions needed to be answered and which datasets would yield the most valuable insights. The result was an invaluable data analysis application (shown below), updated in real time, which allows Direct Relief to constantly track the changing dynamics of Covid-19. These dynamics include the rate of growth of cases, the expansion and decline of testing, and the range of health infrastructure being impacted. Considerations also had to be made for risk factors like age and co-morbidities, such as heart disease and diabetes, which are known to be associated with higher chances of Covid-19 related hospitalization.
Early insights revealed that urban areas were Covid-19 hotbeds, as cities experienced explosive case growth, such as New York, Detroit and New Orleans.
But why were small but significant clusters emerging in places like Blaine County, Idaho and western Wyoming? The answer was that vacation travelers were spreading infectious contacts in areas with great natural beauty and famed resort facilities but relatively poor health infrastructure and older populations. Why were case clusters suddenly popping up outside places like Sioux Falls, South Dakota and in parts of Nebraska with low population densities? The answer involved concentrations of meat processing plant workers and shifting adherence to social distancing guidelines. Why did two counties in the middle of Ohio suddenly see spikes of new cases in the beginning of April? Quickly spreading infections in prison facilities proved to be a crucial emerging issue. Exponential curves are built from hundreds to thousands of smaller issues like these.
Accurate data, properly selected and transformed into insightful information, remains one of our best weapons against the pandemic. The types of insights which have helped to drive Direct Relief’s thinking are now being shared with Direct Relief’s broader community of supporters, as Qlik and Direct Relief make this Covid-19 data application open to all.
The world may still be only in the early phases of the response to Covid-19. To ensure that we can remain alert and aware of the many changes still to come, everyone may need to become better informed than ever about the data behind the exponential crises of our time. And, as we work to stop this crisis, we look forward to learning from these data sets and analysis to create tools to prevent a similar situation from happening again at this scale in the future.
View the Qlik and Direct relief Covid-19 data application by clicking here. In addition, click here to learn about Qlik’s Data Analytics Platform and here to sign up for a free trial of Qlik Sense.