News Blog

COVID-19 Misinformation: The Flip Side of ‘Knowledge is Power’

A social media post with text saying Hashtag Quarantine

Since the COVID-19 pandemic began, the phrase “knowledge is power” has taken on significant meaning. While it’s typically associated with a force of good — the more someone knows about something they can do to make a positive difference in their lives or others’ — it’s becoming clear through recent ongoing research that many have underestimated the force of knowledge that doesn’t originate from the truth.

The term “infodemic” — which has its very own page on the World Health Organization’s website — became common parlance for the exponential way in which COVID-19 misinformation spreads. Every part of the pandemic seemed to have its own piece of mistruth or lie to go with it, ranging from the disease’s origins to treatments and the vaccines that have dulled COVID-19’s impact.

This misinformation has hurt people: An early study estimated that one rumor, which had to do with drinking highly-concentrated alcohol-based cleaning products as a “cure” for COVID-19, led to more than 5,800 people being hospitalized (with 800 dying) from January through March of 2020 alone.

Misinformation causing public health challenges is not a new phenomenon. One branch of COVID-19 misinformation, about vaccines, can be traced virtually back to the origin of inoculations themselves, and, more recently, to a debunked study with fake evidence that sought to link the common measles, mumps, and rubella (MMR) vaccine to diagnoses of autism.

But while we know where some of these myths and conspiracy theories come from, understanding how they spread, and who shares the misinformation, is vital. That can provide opportunities for heading misinformation off before it has a real-world effect. And that’s why the National Institutes of Health (NIH) is providing $3.8 million to a study led by Anish Agarwal, MD, an assistant professor of Emergency Medicine and the deputy director of the Center for Digital Health in the Perelman School of Medicine at the University of Pennsylvania, and Sharath Chandra Guntuku, PhD, an assistant professor (research) in Computer and Information Science in Penn Engineering and a senior fellow in the Leonard Davis Institute of Health Economics, to study the COVID-19-related social media habits of four different populations.

Agarwal and Guntuku hope that using machine learning to analyze social media posts will uncover actionable commonalities across different racial groups and across urban vs. rural environments — both in the spread of misinformation as well as the circulation of scientifically-backed or helpful information.

First, understanding that COVID-19 is not going away and remains a significant threat, it also does not seem to be at the forefront of most people’s minds now. Why is it important to continue studying it in this way? What can we gain?

Agarwal: I imagine the social media discourse around COVID-19 will continue related to boosters, surges, and vaccinations in the coming years and thus it will remain relevant. Moreover, the lessons we learn about how people are engaging and consuming social media content and misinformation will be key for future public health campaigns, health equity, and population health.

You’re planning to study four different population groups segmented by race and where they live. Given some of your past research, such as your analysis of Twitter posts related to specifically to COVID-19 vaccines, do you have a hypothesis about what you may find?

Guntuku: Our preliminary research found that attitudes and behaviors associated with COVID-19 varied widely across different population groups. For instance, at the community level, tweets about COVID-19 vaccinations from counties with predominantly Black Americans had themes about issues of trust, hesitancy, and history in the medical system. In comparison, tweets from other urban counties where the population was predominantly white mainly focused on in-clinic vaccinations near universities, herd immunity, and supporting public access efforts. We want to understand the drivers for these differences expressed on social media.

Why is social media a good lens for studying attitudes toward public health issues and crises?

Agarwal: Social media is an ever-expanding environment. It provides a fascinating lens to see how people share their lives, interact with others, and gather information related to themselves and their health. The ways in which we all interact with social media are important, and now more than ever it's important to investigate our attitudes toward it and how it does (or doesn't) impact what we do especially in regards to our health and well-being.

What are the challenges of studying something like this through social media?

Guntuku: Not everyone uses social media. However, there has been an increasing adoption across different demographic groups during COVID-19 — social media gives us news, health information, and influences our health decision making. Further, social desirability bias, the tendency for people to share information that they think people want to hear, plays a huge role in what information we choose to share on social media.

What will success look like in this study?

Agarwal: We hope it gives us an an extensive and multi-layered method for understanding how social media content drives health decisions and the impact of misinformation upon health. We also are going to conduct in-depth interviews with communities to gain more context about how and why they use social media and their health. Success, to me, would be to help identify key themes to inform future public health messaging for crises like another quick-spreading disease that maintains a clear focus on equity.

Guntuku: We want to create a social listening framework, a method to easily catalogue data on platforms like Twitter for analyzing later, that we could customize to monitor different health topics and make it scalable so that we can focus on specific audiences, like we are in this current study. We have an amazing team spread across Penn and are excited to develop human-centered AI methods — machine learning and natural language processing that could identify and combat misinformation and offer a powerful synergy with traditional approaches to public health campaigns.

The NIH grant funding this work is R01MD018340-01.

You Might Also Be Interested In...

About this Blog

This blog is written and produced by Penn Medicine’s Department of Communications. Subscribe to our mailing list to receive an e-mail notification when new content goes live!

Views expressed are those of the author or other attributed individual and do not necessarily represent the official opinion of the related Department(s), University of Pennsylvania Health System (Penn Medicine), or the University of Pennsylvania, unless explicitly stated with the authority to do so.

Health information is provided for educational purposes and should not be used as a source of personal medical advice.

Blog Archives


Author Archives

Share This Page: