How Big Data Analytics Will Transform
Covid-19 Vaccination Campaigns
ByDale Dauten, Syndicated Columnist & Bill Davenhall, STC Geomedicine
Let’s assume there’s a new covid-19 vaccine being rolled out and you and your team are in charge of deciding on a mass immunization program. You’ve already immunized health care workers and first responders but now you’re getting doses for the general public. Picture this: You are about to receive a million doses. What are you going to do with them?
“There are two ways to plan a mass immunization campaign,” says Mike Popovich, CEO of STChealth: “the old way, which is logistics, and the new way, which is analytics. You can focus on efficiency or on effectiveness – you can simply count doses administered, or you can maximize the good they will do.”
The world is one big data problem.
That takes us to the topic of the day, which is how can big data allows us to make the best use of your shipment of covid vaccine.
We asked Dr. Scott Hamstra, Medical Director at STChealth and someone who had led immunization campaigns, how he’d use data to guide his immunization program.
“You want the most bang for your buck. It just makes sense to use analytics to figure out where it’s most likely to spread quickly. I think of this as being like controlled burns to prevent forest fires. You start your covid-19 immunization campaign by thinking about where it may have the fastest spread, where could it ‘spread like wildfire’ and prepare/prevent these hotspots from ever catching fire.
“There are obvious choices,” he added, “like workplaces where employees are working in confined areas, especially places like meat packing plants with lower temperatures – that’s great for the virus. But then, keep asking,
“Where will it break out and spread?
“For one thing, we know that the virus spreads within a household at over 80%. I’ve seen homes in my work with as many as 18 people living in a four-bedroom house. Most of these homes are multi-generational, with different social circles adding likelihood of exposure. One danger comes from children who bring the virus home to their grandparents. To address this risk, we can flip this equation to an advantage by creating a safe zone for this entire group by creatively offering timely convenient vaccination services. It just makes sense in 2020 to utilize public data to geolocate clusters of households with 5+ people, especially those with ages both under-20 and over-60.”
This is just one example of the sort of specific information that historically has been difficult and time-consuming to obtain. Today with 2020 big data analytics, it’s matter of minutes to take a list of possible clinics sites, define the area around the site (say, a ten-minute drive) and get a demographic profile of each potential site.
“With data collection, ‘the sooner the better’ is always the best answer.”
Would such precision targeting add time to site selection? No. Thanks to Accelerated Analytics. That’s the name Bill Davenhall has given to the ability to almost instantaneously profile any given site location. He says,
“Give me a set of addresses or intersections, and tell me what sort of people you want to focus on (whether it’s by age, race, income, employment status or so on) and how you want to define your area (say, five minutes of walking time or ten minutes of driving) and that same day I can give you the information you need to identify the right population in the right place.”
“What matters with data,” he added, “is not just volume but speed. The first access to information conveys the advantage to decision-making.”
“If the statistics are boring, you’ve got the wrong numbers.”
But using analytics to determine who and where to do immunization clinics is only the first half of the big data equation. Mike Popovich tells us, “There the front-side need for analytics, to figure out who gets the vaccine, but then there’s much more to be done on the back-end, as the shots are administered:
“As the doses are uploaded into the state registries, we can track usage, and direct doses to where they’re needed. We can eliminate most wasted vaccines.
“We can also provide real-time data on uptake and impact. This is where the analytics can turn data into stories, where we start to understand just what is working and how to maximize the effectiveness of the immunization campaign.”
“Maybe stories are just data with a soul.”
Put this all together and we can see why Dr. Scott Hamstra comes to this conclusion: “When I look at the power of data to prevent outbreaks and combat hot spots, I think that those states or agencies who don’t employ analytics will end up being embarrassed by those who do.”
We aren’t the only ones to recognize the enormous role big data can play in taking on covid-19. Here is an excerpt from a recent letter to CDC Director Redfield from a pair of Senators, one Republican and one Democrat, Mitt Romney and Kyrsten Sinema:
Despite having a world-class public health system, COVID-19 exposed many vulnerabilities in the United States’ ability to proactively monitor and mitigate the spread of a novel infectious disease outbreak. Our public health biosurveillance and data collection should be able to provide comprehensive, near real-time reporting. Currently, fewer than 10 percent of nationally notifiable disease reports are submitted in a format that allows for near real-time analysis.
It is critical CDC support the modernization of state public health data systems to ensure they have the data to inform public health actions and work with states and other stakeholders to implement a more robust data gathering system to provide near real-time visibility into the size and scope of the COVID-19 outbreak on a national level; track, triage, and quickly respond to hotspots; and prepare our nation for the next pandemic.