Real-Time Data Helps Leaders Forecast COVID-19 Tornado
Subscriber Benefit
As a subscriber you can listen to articles at work, in the car, or while you work out. Subscribe NowThe first tornado nationally was recorded in 1884 and since then, meteorologists have worked tirelessly to develop forecasting tools that give people more time to seek shelter. The COVID-19 pandemic reminds me of this herculean effort by forecasters over the decades to find the best, real-time data that could provide any sense of prediction and warning.
Prior to 1987, tornadoes could only be predicted with three-and-a-half minutes of warning time; that number has since increased to 14 minutes because of immediate access to real-time data and prediction models.
Similarly, SARS-CoV-2 spreads quickly and imparts devastating effects when left unchecked. To date, there are many tools available that share snapshots of infection rates across the country, but those data points haven’t historically been updated at the same speed at which the virus spreads.
Adding insult to injury, most snapshots of COVID-19 infection rates don’t drill down enough to give local government and health officials a clear picture of the storm’s trajectory. Without hyper-local data, it is impossible to properly predict and mitigate the spread before the pandemic leaves devastation in its wake.
The time to brace a community and proactively prepare is 5 to 21 days in advance of an acceleration of patient volumes with symptomatic COVID-19. This acceleration is a warning of an oncoming COVID-19 tornado. As the CEO of a bioinformatics company, our team calls this warning the Local Risk Index (LRI).
Presently, a majority of public sector websites reflect “hotspots” based on the total number of COVID-19 cases over the entire multi-month period, not the current week. In addition, the data flowing into public sector sites is often eight to 14 days old. The data needed to know when and where to act has to be forecasted quicker in order to keep our communities safe. Below, I share three ways that public access to granular, real-time testing data can help better prepare us for rapid detection of a resurgence.
Increasing infection rates
Like a potentially dangerous weather pattern, positive COVID-19 test results within a community indicate a storm is brewing and could wreak havoc in places where human interaction is most prevalent, such as workplaces, schools and public spaces. Hyper-local data indicating positive test results provides public health officials a clearer picture of where contact tracing, quarantining, and elevated testing should occur, and it sounds a siren for healthcare professionals to begin preparations in anticipation of rising hospitalizations for treatment for COVID-19 complications. Taking a proactive approach prevents the spread and protects the bandwidth healthcare officials need to respond appropriately not only to COVID-19 patients requiring their care, but also those seeking treatment for other ailments.
Infection rates hold steady
If local risk indices are flat, this indicates current demand for equipment, hospital beds and safety measures will remain level. Public officials will need to remain vigilant, ensuring mask mandates and social distancing measures in public spaces, schools and workplaces are being followed. Furthermore, residents in these communities will need to be notified that the threat is still prevalent and they should continue following proven mitigation strategies. When this data is available in real-time, we can safely begin to estimate how long it might take to flatten infection rates if residents follow best practices.
Decreasing infection rates
An LRI that shows a decrease in positive test results within a specific community is an indication that the pandemic is receding. The demands on healthcare will begin to decline in a Rolling Trend between 8-14 days, beginning with emergency rooms and finally in intensive care units and ventilator use. Calculated daily, the Rolling Trend in % Detected (7/7) is calculated by dividing the average detected rate for the past seven days by the average detected rate for the seven days prior to that period.
Unfortunately, we’ve also learned that this period is also the point in which communities see the largest amount of COVID-19 fatalities. Tragically, as many states have begun to understand the complexity of the virus, it is in this stage that we see heightened news coverage and reactive decision-making. At this point, we are working in a reactive mode.
Using localized data from the onset is much like weather forecasting. While we can’t completely eliminate the dangers of novel pathogens that threaten to become the next pandemics, with actionable insights disseminated in a timely manner, we can take cover and prepare to contain and minimize the damage.
Tying down lawn chairs when the wind is already 80 mph is not an effective way to protect property during a tornado. To lessen COVID-19’s effect in our community, it is imperative that we seek solutions that provide the most recent, relevant and accurate information about the infection’s spread in order to safely ride out the storm.