How the CDC’s Center for Forecasting and Outbreak Analytics is ramping up efforts to better plan for infectious disease outbreaks.
We live in a globalized, interconnected world. While this has created enormous opportunities for travel and trade, it’s also increased mankind’s vulnerability to zoonotic diseases through contact with animals. And most infectious diseases come from animals in some capacity, said Dylan George, PhD, director for the Center for Forecasting and Outbreak Analytics (CFA) at the Centers for Disease Control and Prevention (CDC).
“What we learned during the COVID pandemic is that the next infectious disease outbreak is only a plane ride away,” Dr. George said.
Indeed, as the pandemic taught us, we are just one incident away from one individual infecting another and causing an outbreak. Having better public health systems and capabilities will make sure that U.S. healthcare stakeholders can better prepare for and ultimately combat those infectious disease outbreaks.
Dr. George said we are at a transformative moment in history. It is an era of pandemics and epidemics throughout our interconnected world. We are also in the middle of a technological renaissance, with new inventions, systems and methods that can enhance our response to healthcare crises.
CFA’s team is tasked with developing the capabilities that will help address those threats in a much more robust way. CFA works with private industry partners, academia, local public health organizations, and other agencies in the federal government to create forecasting tools and other resources to respond to public health emergencies.
In the fight against infectious diseases, information is power, said Dr. George. “Our main goal is to support public health decision makers in making decisions associated with infectious disease outbreaks,” he said. “The way that we do that, our kind of secret sauce in how we’re trying to achieve that goal, is by developing modeling analytics and forecasts to help us understand what’s coming next in an outbreak.”
CFA makes models, tests and adjusts them so that healthcare stakeholders and government agencies can better understand how interventions are used in real time during an outbreak.
In addition to COVID-19, CFA has supported the development of new tools and techniques to forecast emerging disease threats such as Mpox, Polio, and Acute Pediatric Hepatitis. Since 2022, CFA has awarded over $122 million in funding to academic, private and public institutions to advance modeling and forecasting strategies. In 2023, CFA launched Insight Net, the first national network for disease outbreak modeling and analytics, which includes more than 100 partners.
Partnerships
As potential outbreaks develop, the CFA works closely with state and local jurisdictions, as well as academic groups, to analyze the data available and create better methods of monitoring. For instance, in the early stages of H5N1, the movement of dairy cattle being sold or transported across the United States was a critical concept to understand because it was leading to the spread of the disease. CFA’s partners at Northeastern University worked to come up with predictive models of how cattle movement was happening in the United States based on how they were being moved around from farm to farm across the interstate. This helped stakeholders target surveillance in different states in a much more effective way in the early stages.
Meanwhile, CFA’s partners at Johns Hopkins University created a risk assessment for H5N1. Although there was relatively little data to use, researchers were still able to apply a model for informed thinking towards the risk. The risk assessment went up relatively quickly to help stakeholders to start thinking through what the challenges and risks involved.
Partnerships with academic groups like Northeastern and Johns Hopkins are critically important toward developing new methodologies, modeling, and forecasting capabilities. State and local partners are also central to CFA’s mission. “In the United States, most of the action in public health happens at the state and local level,” said Dr. George. “The federal government is here to support the state and local level government in what they’re doing. They are a key partner in how we’re moving things forward. That’s why we want to work much more closely with our state and local partners and their public health organizations, to make sure they have better resources and better modeling capabilities to help the people that live in their jurisdictions make better decisions about ongoing outbreaks. Both academics and state and local [officials] are critically important for us to execute on our mission and meet our goals.”
Because of the way the nation’s public health system is set up, there is diversity across various groups on how they’re reporting data, and the frequency with the reporting data, Dr. George said. Efforts are ongoing to standardize what’s reported. There’s also a data modernization initiative underway to improve the technology used to enable data to be moved more effectively from electronic health and laboratory records, immunization services, death and vital record services to public health in a much more efficient and timely way.
“Not only are standards being developed, but also technology is being deployed to improve those processes to make sure that we have much more improved quantity and quality of data that will help us do the modeling that we need to moving forward.”
CFA and respiratory season
The big three respiratory diseases – influenza, COVID, and RSV – create the most hospital burden during respiratory season. Last year was the first year that CDC put out a Respiratory Disease Seasonal Outlook. Over the course of the season, CFA wanted to monitor how much hospital burden should be expected within various jurisdictions. This was a pivotal effort, because public health has historically looked retrospectively at the data to try to understand the real-time risks.
The pandemic revealed the need to do that much faster, and be forward leaning, said Dr. George “We need to be prospective, we need to be thinking forward into the future as to what’s coming at us, because it’s happening so fast,” he said. “CFA was born to create analytical capabilities that will help us anticipate the challenges we’ll see in an outbreak.”
The seasonal outlook was a major step forward in looking prospectively across the season, Dr. George said. “That’s really hard to do. I’m going to be the first one to admit that it is fraught with challenges going forward, but we just actually graded ourselves recently on that, and we did pretty well this first season. Now, we need to turn the crank and do it much better.”
The goal is for agencies and jurisdictions to be able to use the information to better anticipate and plan for the level of burden hospitals can expect for COVID, influenza and RVS. With an accurate forecast, state health officers can ask hospital leaders if they’re prepared for a surge, and if not, determine what steps they need to take before the numbers begin to rise.
“A couple different states were talking to their health hospital care, their hospital systems to actually do that sort of planning ahead of the time,” Dr. George said. “That’s how they used it last season, and they’re going to try to use it again in a similar way this upcoming season. We’re really excited about how it’s being used for preparation and preparedness in the face of this respiratory season.”
At the federal level, CFA works closely with the Administration for Strategic Preparedness and Response (ASPR) to anticipate the demand, and whether there could be supply issues in a time of crisis. “We work very, very closely with our colleagues in ASPR to make sure that they understand what the kind of epidemiological predictions are, and the anticipated forecasts, so that they can then figure out the analytics for the demand that would be needed for vaccines, therapeutics, and diagnostics in different capacity for important various medical countermeasures.”
Future response
Before joining CDC, Dr. George worked in the private sector as a vice president at Ginkgo Bioworks, where he helped develop improved real-time infectious disease monitoring capabilities and analytics for pandemic response. Prior to Ginkgo, Dr. George was a vice president at In-Q-Tel (IQT), where he vetted life science and healthcare deals, and developed science and technical strategies to strengthen capacity within the U.S. to counter biological threats from infectious disease.
Dr. George served on the Biden-Harris transition team working on national security policy for the COVID-19 response, and on the agency review team for the Department of Health and Human Services.
One of the reasons he left the private sector to help build CFA was because he fundamentally believes that we are at a pivotal point in history where we can actually bring to bear new technologies, improved data and improved analytical capabilities to generate better information so that people can make better decisions about their own health and their family’s health.
“I know that we saw lots of failings in how we were using data across COVID, across the pandemic,” he said. “I am confident that the team, the tools, the processes, and the capabilities we’re building are going to be helpful to keep Americans safe in the next pandemic. That’s why I get up in the morning. That’s why I can look my kiddo in the eyes and can smile at him, knowing that he’s going to have a brighter future.”
Dr. George likens CFA to a startup company within the government. They began as a five-person team, and have grown to more than 75. “I’m very proud of the opportunity to work with the team that we’ve built at CFA,” he said. “They are some of the best in the business, and we are really shooting for becoming the world leader in how to use analytics to guide decision-making during an outbreak or a response.”
CFA has come a long way in its capabilities in a short amount of time. In the early stages of building the organization, if there was an outbreak or a surge in infectious disease cases, they had to choose whether to help in the response efforts or continue to build out the infrastructure. “Of course, we were conceived of and built to respond, so we always picked response, but we had to put all the building the organization on hold while we were helping out.”
For example, during Omicron, there were only eight team members. While CFA helped with the Omicron response, it took weeks of involvement, and the team had to pause its build-out plans.
That’s no longer the case. For instance, CFA was involved with the measles response in Chicago in June and July. They’ve been involved in the H5N1 outbreak in cattle from the start. And they’ve been preparing for the fall respiratory season. “We’ve been doing high quality work in all of those responses and preparatory work, while we’ve still been able to continue to build our organization,” Dr. George said. “We can walk and chew gum at the same time. We’re meeting our mission, and we’re improving on how we’re doing that at the same time, which is a really great place to be. I’m super excited about where we’re going.”
AI and outbreak analytics
Technology is increasing the ability to better monitor and forecast infectious disease outbreaks as researchers are testing advanced analytics in different capacities to learn more about an infectious disease going forward. That includes the potential born from artificial intelligence. One of the challenges that researchers are finding currently with artificial intelligence, though, is that there needs to be a very big body of data to actually use it at scale.
“The challenge with an outbreak is that in its early stages, you have relatively few data,” said Dr. George. “There’s a mismatch of applying a lot of artificial intelligence capabilities to an emerging outbreak. We need to find tools that will help us work in a data sparse environment. That’s one of the challenges that we’re working on and trying to apply the advances in artificial intelligence and other advanced analytics to this particular problem.”
CFA Goals
CFA supports innovation to improve the science of disease outbreak analytics and modeling by:
- Producing data-driven models and forecasts to predict the course of disease outbreaks.
- Informing decision-makers of potential consequences of outbreaks and public health responses.
- Ensuring effective decision-making, using real-time data, advanced scientific analytics and modeling.
CFA staff work with partners in public, private, and academic sectors to advance their research and achieve their goals.
Predict
Deliver actionable analysis and response-ready modeling tools.
- Build an array of models and tools for outbreak analytics.
- Collaborate with public health programs to apply CFA’s models and community tools.
- Test, evaluate, and improve CFA’s modeling and data sources.
- Provide analysis for urgent response needs.
Inform
Generate practical decision support products.
- Share technical reports with actionable insights for decision-makers.
- Translate and circulate CFA-developed analytics.
- Establish and maintain collaborations to improve CFA techniques.
Innovate
Drive technological and analytic innovation.
- Transform forecasting, analytics, and communication.
- Advance analytic technology architectures for CFA and it’s partners.
- Establish technical specifications to support the flow of information.
Advance
Build a world-class forecasting and outbreak analytics organization.
- Attract and retain an ambitious and diverse workforce.
- Drive CFA talent growth through robust career development initiatives.
- Foster a culture of collaboration, innovation, and transparency.
- Plan and execute for sustained success
Insight Net
Established in 2023 by the Centers for Disease Control and Prevention (CDC) Center for Forecasting and Outbreak Analytics (CFA), Insight Net focuses on training, analytical tool development, and advancing the analysis and use of data about infectious disease spread. Through this, CFA aims to improve anticipation of infectious disease outbreaks, generate information to empower public health decision-makers, and guide actions for protecting communities in public health crises.
The network brings together 13 centers, including 5 centers of innovation, 5 centers of integration, and 3 centers of implementation.
Innovators
- Develop advanced methods and tools.
- Train the next generation of specialists in ID modeling and analytics.
- Focus on supporting public health decision-making.
Integrators
- Bring together methods and tools that are closer to real-world use.
- Establish response-ready data collection mechanisms.
- Evaluate integration efforts.
Implementers
- Help scale up promising outbreak tools after pilot testing.
- Train public health professionals in forecasting.
“Together, our mission is to support state, tribal, local, and territorial (STLT) public health departments by establishing connections with STLT jurisdictions, creating infrastructure for response activities, and identifying key areas of focus in public health emergencies,” CFA says on its website. “All centers stand ready to activate in an infectious disease emergency.”
Source: https://insightnet.us