The 2025-26 flu season has been one of the most severe in recent years, with hospitalization rates reaching their second-highest level in 15 years of CDC tracking. A team at the University of Arizona was able to spot the surge early, weeks before federal lab data confirmed it. They did it with a digital platform that collects flu symptoms directly from people across the country. 

Based at the Mel and Enid Zuckerman College of Public Health, the digital platform Global Flu View brings together self-reported flu symptom data from partner programs in 11 countries. What began as a way to compare a handful of national surveillance systems grew into a shared digital infrastructure, with a common map, standardized data and AI-driven forecasts that project flu activity weeks ahead. The platform is part of the Global Health Institute in the College of Public Health. 

Image

Global Flu View is one of the very few functional digital epidemiology tools available worldwide – a field that uses non-traditional data sources such as self-reported health surveys to track and predict disease patterns.

"We wanted to see if connecting self-reporting systems into a single platform could help answer our fundamental questions – where flu starts and how it spreads globally. That was the genesis of this platform," said Mark Smolinski, who helped create Global Flu View, leads the Ending Pandemics Academy at the College of Public Health and serves as the Jeff Skoll Endowed Chair. 

From Flu Near You to Global Flu View

Global Flu View grew out of an earlier project called Flu Near You. Smolinski and his team built Flu Near You in 2011 in partnership with HealthMap, a disease surveillance project at Harvard, to track self-reported flu symptoms across the United States.

"We knew that any pandemic threat would most likely be a respiratory disease, because what's what allows it to spread globally. So, we built a tool around that." Smolinski said.

From the start, the team designed Flu Near You to be compatible with similar systems running in Australia and Europe. As more countries launched their own similar systems, the teams behind them began collaborating. It became clear that combining their data into a single platform could reveal patterns that no single system could see on its own. That's how Global Flu View came into existence. 

Beyond the map

Since its public launch in late 2022, Global Flu View has grown well beyond a simple map. The platform now pulls in more than 22 million individual symptom reports from 11 countries, said Onicio Leal-Neto, leader of Global Flu View and an assistant research professor at the College of Public Health.

Image

In fact, researchers and public health officials can access AI models that forecast flu activity four weeks out, share analytical tools with peers and browse a library of health data collected directly from the public.

But the feature that may matter the most is a toolkit that lets any institution in the world build its own symptom-reporting program from scratch and plug directly into the platform. The team has tested it in Nigeria and the Philippines, refining the tool based on what public health officials on the ground needed. 

"It's available for free for any institution in any part of the globe," Leal-Neto said. 

Members get a weekly notification by phone or email asking how they're feeling. They can either tap a button saying they are fine or check off relevant symptoms from a list. 

They are not diagnosing themselves with the flu. Rather, the system does that on the back end, matching reported symptoms against the CDC's definition of influenza-like illness, a type of illness characterized by a group of symptoms that are similar to those caused by the flu virus. 

The platform also creates opportunities for the next generation of public health researchers. Global Flu View's Spark program, offers students at the College of Public Health the chance to work directly with the platform on projects ranging from AI-driven forecasting to expanding surveillance into new regions.

Filling a gap

What makes this data valuable is what it captures that the traditional approach cannot. Official flu tracking relies on people visiting a doctor and getting a lab test. But an estimated 80% of people who get the flu never see a doctor, Smolinski said, which means the official numbers only reflect a fraction of actual cases.

The goal is not to replace the official systems, but to fill the gap between when people get sick and when they show up in the data, Leal-Neto said. 

Participants aren't just feeding into a global system. Through the platform, users can explore trend charts with historical data dating back to 2009 and maps displaying influenza-like illness hotspots across different global regions. That two-way exchange is central to what researchers call participatory surveillance.

"The platform really relies on the altruistic nature of people who care about the health of their communities," Smolinski said.

People in the United States can contribute by signing up for Outbreaks Near Me (previously Flu Near You). It takes about five seconds each Monday to complete a survey. Reporting when you are healthy matters just as much as reporting when you are sick, Smolinski said. 

"Join a system," Smolinski said. "That's the best thing you could do for the health of your community."