top of page
pexels-crazy-motions-12920835.jpg

Virtual Town Hall

In 2023, the PILOT team convened three asynchronous virtual town halls to answer questions about three themes:

​

1. PANDEMIC POTENTIAL: Estimating the pandemic potential of a (re-)emerging pathogen;

2. DISEASE TRANSMISSION: Modeling the impact of public health interventions on disease transmission (with a special focus on using AI); and

3. INTERVENTIONS: Monitoring (mis)information about interventions to better mitigate infectious disease crises

​

As many questions as possible regarding Themes #1, #2, & #3 were answered by the PILOT team over three 90-minute periods on March 20th, May 2nd, and July 19th, 2023. Our answers appear below, and our question box remains open for ad-hoc questions from the public.

 

Thanks for submitting!

Theme #1: Pandemic Potential 

Questions and Answers 

​

Question: How important are routine surveillance efforts in estimating pandemic potential? What are some institutional/political/societal changes that would facilitate estimating the pandemic potential of a pathogen? What are some institutional/political/societal changes that would hinder estimating the pandemic potential of a pathogen? Could you outline more recent developments in serotyping/genotyping/rapid diagnostics that could assist in estimating pandemic potential? Thanks!

Answer: Thank you for your question! Routine surveillance is essential to being able to estimate parameters such as reproduction numbers that ultimately contribute to our understanding of the pandemic potential of a particular pathogen. While advancements in diagnostic technologies (including rapid diagnostics, at-home testing, multi-analyte molecular diagnostic platforms), genomic surveillance and phylogenetic analysis, and regional/international reporting mechanisms have improved our ability to conduct global disease surveillance, many gaps remain. Resource inequities, regulatory barriers, financial constraints, and lack of significant global, standardized reporting and data sharing are just some of the challenges that hinder rapid and accurate disease surveillance worldwide.

 

Question: There is a global gap in surveillance of companion animals and wildlife. Traditional veterinary surveillance is focused on livestock diseases, primarily in support of trade. We agree that 60-75% of emerging diseases of humans come from animal species, and in the last 40 years it’s been commonly direct from wildlife reservoirs, with occasional indirect amplification in livestock before spillover to humans. It stands to reason that veterinary surveillance must be enhanced. The so called “shift in surveillance to the left” will identify animal to animal transmission, and further to the left can identify vector shifts, all in an effort to predict spillover to humans. None of this is new. And yet very little funding/conversation/attention has been paid to enhancing all veterinary surveillance. Many health professionals dismiss inclusion of veterinary teams in pandemic discussions. Perhaps they wrongly imagine vets only from the perspective of cute puppies and kittens. And yet as the profession who has license to diagnose, treat and care for every animal species on the planet bar one, the critical need for inclusion of their expertise is urgent. I would be interested to hear your panels perceptions about enhancing veterinary surveillance by increasing veterinary workforces, and veterinary diagnostic capacity across species traditionally ignored.

Answer: Thank you for this comment! Prior studies have demonstrated that zoonoses comprise an increasing number of emerging infectious disease events worldwide.  Expanding disease surveillance efforts to include the ‘One Health’ approach you have outlined above is an important component to improving global detection of emerging infectious disease events and potential outbreaks. More recent research suggests that “spillover” events are much more common than previously believed; however, lack of ongoing surveillance as well as diagnostic limitations likely underestimate their contribution to human disease. Veterinary expertise and enhanced veterinary surveillance are critical to include in pandemic preparedness efforts for these reasons. However considerable debate remains regarding the extent to which extensive veterinary surveillance can inform pandemic risk, given the likely numerous spillover events that take place worldwide (and conversely, the many known animal pathogens that do not affect humans), the time (and probability) it takes for a pathogen to evolve to develop efficient transmission to humans and then between humans, and consideration of financial constraints. Improving diagnostic capabilities to detect spillover events in humans at near point of contact, improving veterinary surveillance at high risk interfaces such as poultry farms,   and addressing the underlying drivers of spillover events (including but not limited to increased human-wildlife interaction due to land use changes, urbanization, and climate change) are some possibilities that have been discussed to try and mitigate some of these risks. But we agree, more can and should be done!

 

Question: What is the pandemic or epidemic potential of Lloviu virus? Spillover of this filovirus hasn't been observed as of yet although there hasn't been detailed human serology studies in regions where the virus is endemic. Is there likely to be transmission from bats to other wildlife (mammalian) hosts?

Answer: Interesting question! As you have suggested, more research (including serologic studies in humans) is needed to characterize the virus itself, as well as its risk factors for cross-species transmission. Other high-risk filoviruses as you mention have been included in the WHO's R&D Blueprint. At least one recent study has suggested human cell lines being susceptible to LLOV introduction, however no human infections have yet been observed. Its potential for human pathogenicity remains to be seen, and more research is needed.

 

Question: What is the pandemic potential of H5N1? It seems to have had many chances in the past but never really took off. Do we have an idea of what specific mutations are needed for widespread human transmission?

Answer: Great question! This is a complicated issue. In the past, sporadic cases of H5N1 had been linked to exposures to infected birds. This has been thought to be due, in part, to receptors the virus preferentially binds to in the upper airway of birds, which are less common in mammals. In 2020-2021, a new variant (2.3.4.4b) spread geographically further and more rapidly than prior cases among birds. This current outbreak has also been unique in its isolation among a variety of mammals. An early paper documented H5N1 among mink at a mink farm; sequenced isolates from some of the mink demonstrated mutations that may have allowed the virus to replicate more easily in mammalian tissue (see de Vries and de Haan 2023). While it is difficult to predict what comes next, improved surveillance at high risk interfaces and data sharing are important to better track its potential moving forward.

​

Question: Considering the antigenic distance between the XBB variant and WT SARS-CoV-2, how likely is it that we could see another "Omicron-like" event in the near future? 

Answer: Thanks for this question! Difficult to predict as there are many factors which can lead to the emergence of new variants. Genomic surveillance (as well as potentially wastewater and other tools) will be important to help track its trajectory.

​

Question: Do we expect a pandemic driven by bacterial pathogens in the near future? Since the advent of antibiotics, we haven't seen anything near the scale of the Plague but is that due to change because of AMR?

Answer: Good question! Since the advent of broad-spectrum antimicrobials, bacterial pathogens of pandemic potential have been less common. However, as you suggest, rising rates of antimicrobial resistance and the emergence of multidrug resistant organisms such as carbapenem-resistant Enterobacteriaceae are certainly of public health concern. These organisms tend to, however, be present among long-term hospitalized patients or immunocompromised patients, somewhat limiting their pandemic potential for now. We do continue to see large-scale outbreaks of pathogens such as cholera and foodborne disease, though this is often in the context of extreme weather events, conflict, poor sanitation and hygiene, or limited public health infrastructure. See the JHU Center for Health Security publication, “The Characteristics of Pandemic Pathogens,” for an excellent in-depth discussion of this and other related topics.

​

Theme #2: Disease Transmission

Questions and Answers 

 

Question: What is the difference between an explainable and an interpretable AI model? Are these better than black box models for tracking disease spread? Or does explainability/interpretability make humans more likely to over-trust a model over their own judgment?

Answer: In many contexts, “explainable AI” and “interpretable AI” are used interchangeably. There are different ways to achieve explainable AI. Some existing work uses simple self-explainable models such as a shallow decision tree, which is essentially a simple rule-based model, to make predictions or suggest actions. Human experts can verify these models before using them. In this way, over-trust is less of a concern as the rule-based model is carefully checked and approved by human experts. Some other works in explainable AI try to use self-explainable models to explain the predictions made by complex black-box models such as deep neural networks. In this case, how the model actually works is different from what the explanation says. If the user does not understand this subtle difference, over-trust could be an issue.

 

Question: What does ChatGPT say we should do to control COVID? (And should what it says be trusted?)

Answer: With the prompt “What should we do to control COVID?”, we get the following answer from ChatGPT (based on GPT-3.5 model):

 

“As an AI language model, I can suggest some measures that have been recommended by health organizations such as the World Health Organization (WHO) and the Centers for Disease Control and Prevention (CDC) to help control the spread of COVID-19:

​

  • Get vaccinated: Vaccination is the most effective way to protect yourself and others from COVID-19.

  • Wear a mask: Wear a mask when you are in public settings or around others who are not in your household, particularly when social distancing measures are difficult to maintain.

  • Wash your hands frequently: Wash your hands often with soap and water for at least 20 seconds, especially after being in a public place, or after blowing your nose, coughing, or sneezing.

  • Practice social distancing: Keep a distance of at least 6 feet from other people, particularly those who are sick or may have been exposed to the virus.

  • Stay home if you're feeling unwell: If you have symptoms of COVID-19, such as fever, cough, or difficulty breathing, stay home and isolate yourself from others.

  • Avoid large gatherings: Avoid large gatherings, particularly in enclosed spaces where social distancing may be difficult.

  • Follow local guidelines: Follow guidelines and regulations issued by your local health department, as well as any travel restrictions or quarantine requirements that may be in place.

​

It is important to note that while these measures can help reduce the spread of COVID-19, there is no guarantee that you will not contract the virus. If you develop symptoms or have been in close contact with someone who has tested positive for COVID-19, it is important to get tested and follow the advice of healthcare professionals.”

​

It is known that ChatGPT can make mistakes and provide factually incorrect information. We should not trust everything ChatGPT says. But for this particular question, ChatGPT’s answer aligns with what we have learned from public health experts.

 

Question: In the context of respiratory pathogen pandemics, is it more feasible (financially etc.) to focus on air filtration or vaccines aimed at stopping transmission?

Answer: Air filtration and vaccines both are important and effective measures. In addition to these measures, there are other effective and feasible measures that should be considered such as social distancing, testing, etc. Based on what we have witnessed in the past few years, a combination of many different measures could be a good way to go (i.e., the Swiss cheese model of pandemic defense). Though we can't provide an estimate of the cost for different measures of the top of our heads, in the future, we hope to build mathematical models and run various simulations to compare the cost-effectiveness of these measures within the context of respiratory pathogen pandemics.

 

Question: The idea of using AI to monitor intervention effectiveness is super cool! Although I admit that my ideas around how that would look are limited other than thinking of a time series analysis of cases pre- and post-intervention or a regression analysis/diff-in-diff between areas that do and do not have the intervention. What are some potential novel uses of AI in this space? Do you have any good representative papers of what is possible?

Answer: Yes, there are many potential ways of using AI in this space. There has been work on predictive modeling, which forecasts the impacts of interventions on the spread of the disease. A typical way is to use machine learning models for predictive modeling. Beyond this, there are a number of works on agent-based modeling, Bayesian inference, game-theoretic multi-agent models, and other models that have been used to analyze how individuals, communities, or states might act under different scenarios or interventions, which can also help decision-makers analyze the potential impact of interventions. We also see the potential of using reinforcement learning to choose a sequence of interventions adaptive to the progress of an infectious disease crisis. Here we list a few references including some of the ones that we have worked on in our own research (and there are plenty more!):

 

  1. Wilder, Bryan, Marie Charpignon, Jackson A. Killian, Han-Ching Ou, Aditya Mate, Shahin Jabbari, Andrew Perrault, Angel N. Desai, Milind Tambe, and Maimuna S. Majumder. "Modeling between-population variation in COVID-19 dynamics in Hubei, Lombardy, and New York City." Proceedings of the National Academy of Sciences 117, no. 41 (2020): 25904-25910.

  2. Wilder, Bryan, Mina, Michael, and Tambe, Milind Tracking disease outbreaks from sparse data with Bayesian inference In AAAI conference on Artificial Intelligence (AAAI), February, 2021.

  3. Wilder, Bryan, Suen, Sze, Tambe, Milind Allocating Outreach Resources for Disease Control in a Dynamic Population with Information Spread In IISE transactions, 53(6): 629-642, 2021.

  4. Wilder,  Bryan, Suen, Sze, Tambe, Milind Preventing infectious disease in dynamic populations under uncertainty In AAAI Conference on Artificial Intelligence (AAAI), February, 2018.

  5. Xuan, Yao, Robert Balkin, Jiequn Han, Ruimeng Hu, and Hector D. Ceniceros. "Optimal policies for a pandemic: A stochastic game approach and a deep learning algorithm." In Mathematical and Scientific Machine Learning, pp. 987-1012. PMLR, 2022.

  6. Liu, Changliu. "A microscopic epidemic model and pandemic prediction using multi-agent reinforcement learning." arXiv preprint arXiv:2004.12959 (2020).

  7. Liang, Yu, and Amulya Yadav. "Let the DOCTOR decide whom to test: Adaptive testing strategies to tackle the COVID-19 pandemic." medRxiv (2020): 2020-12.

 

As a more specific example, we'd like to highlight this study (Larremore et al. 2021), which uses AI (i.e., agent-based modeling) to compare rapid tests against “gold standard” PCR tests. This paper came out in the early days of the pandemic (first as a pre-print and then in Science Advances) when rapid tests were not easily available and had no FDA approval. It provided some of the first quantitative evidence to show the benefit of rapid tests.

​

Question: We currently do not have well established correlates of protection against asymptomatic or symptomatic infection for SARS-CoV-2. Many mucosal vaccines aimed to reduce transmission are in the works. Do we expect these to be widely deployed for the next respiratory pandemic? (China/India currently have these but data are scare on how effective they really are) 

Answer: Great question. Remains to be seen! As you mention, there are some mucosal vaccines that are currently being studied in other contexts, and certainly may offer some immunologic benefits as well as potential ease of administration among other advantages. However, this has not always borne out in practice, as many mucosal vaccines which have been successful in animal models have not succeeded in subsequent clinical trials. Advancing our understanding of mucosal immunity and adjuvants to optimize mucosal vaccine uptake and delivery, as well as potential unique characteristics associated with a particular pathogen will be important factors to consider for the development and deployment of mucosal vaccines in the next pandemic (see Lavelle & Ward 2022 and Hameed et al. 2022). 

​

Question: What proportion of the national population should be covered by a strategic stockpile of masks, intensive care beds, emergency drugs, first line drugs, anti-infectives, antivirals, etc. to deal with a microbiological threat that is unknown at the time the strategic stockpile is created? What specific drugs should be included in the stockpile? How do you "scientifically" determine the quantity of each drug or class of drugs to plan for?

Answer: This is a very good question. There has been some work in AI analyzing the strategic allocation of resources while taking into account contagious phenomena [1], or more generally in strategic allocation and reallocation of resources [2], which can shed some light on this problem. However, a key challenge that’s emerged in scientific endeavors to date in this area is the uncertainty of available resources and the complex combination of different types of resources. In the future, we hope to build mathematical models that will address this particular challenge. Whereas the following papers are not all focused on allocation of the resources mentioned in this question (i.e.., strategic stockpiles of masks, drugs, and antivirals), the techniques outlined in them may help in making such allocations:

​

  1. Ou, Han-Ching,  Sibenbrunner, Christoph,  Killian, Jackson,  Brooks, Meredith,  Kempe, David,  Vorobeychik, Yevgeniv,  Tambe, Milind Networked Restless Multi-Armed Bandits for Mobile Interventions In International Conference on Autonomous Agents and Multiagent Systems (AAMAS), May, 2022.

  2.  Ou, Han-Ching,  Chen, Haipeng,   Jabbari, Shahin,  Tambe, Milind Active Screening for Recurrent Diseases: A Reinforcement Learning Approach In International Conference on Autonomous Agents and Multiagent Systems (AAMAS), May, 2021.

  3. Ou, Han-Ching,  Sinha, Arunesh,   Suen, Sze,  Perrault, Andrew,   Raval, Alpan, Tambe, Milind Who and When to Screen: MultiRound Active Screening for Network Recurrent Infectious Diseases Under Uncertainty In International Conference on Autonomous Agents and Multiagent Systems (AAMAS), May, 2020.

  4. Bai, Rufan, Haoxing Lin, Xinyu Yang, Xiaowei Wu, Minming Li, and Weijia Jia. "Defending against contagious attacks on a network with resource reallocation." In Proceedings of the AAAI Conference on Artificial Intelligence, vol. 35, no. 6, pp. 5135-5142. 2021.

  5. Bondi, Elizabeth, Hoon Oh, Haifeng Xu, Fei Fang, Bistra Dilkina, and Milind Tambe. "To signal or not to signal: Exploiting uncertain real-time information in signaling games for security and sustainability." In Proceedings of the AAAI Conference on Artificial Intelligence, vol. 34, no. 02, pp. 1369-1377. 2020.

 

Beyond figuring out where to have strategic stockpiles of certain goods like vaccines, there is also the question of getting those vaccines into arms. The challenge is often that in under-served communities, there may be a lack of awareness regarding the importance of vaccination or a lack of adequate transportation, and so on; thus, we need to plan specific interventions for specific subpopulations to enable vaccination. This paper (Nair et al. 2022) shows such an approach with results in simulations.

Theme #3: Interventions

Questions and Answers 

 

Question: Are there interventions developed to combat misinformation for future pandemics? There needs to be a way to address public concerns with scientific experts while tampering down on widespread speculation on social media. How do we share relatively advanced scientific knowledge while maintaining nuance inherent to scientific debate?

Answer: Thanks for this question! The good news is that many researchers specialize in studying how we can stop or slow the spread of misinformation and they have identified a lot of strategies that work. Some strategies focus on stopping misinformation from spreading to begin with. Social media platforms like Facebook or Twitter can remove posts or accounts that are spreading misinformation or make it harder to find posts from accounts known to spread false information (sometimes called “shadow banning”). These interventions have been shown to reduce the overall volume of misinformation on platforms, at least for a little while. 

 

There are also things we can do to keep people from believing and spreading misinformation. Simply warning people that there is a great deal of uncertainty in the early days of a pandemic and any claims may be false can make them more receptive to corrections later on. Specific warnings about false content that is circulating – a technique called “prebunking” – can help people resist falling for false claims by preemptively refuting them. And content warnings such as the fact checks or links to counter arguments that some platforms have implemented have also proven to be quite effective for stopping the spread of misinformation. 

 

Increasing media literacy is our most effective tool for combating misinformation. Teaching people where to find credible and trustworthy information, how to validate claims by checking multiple sources, understanding the motives for different sources and platforms, and slowing down and thinking about the plausibility of claims are all important strategies for increasing people’s abilities to spot misinformation. 

 

Finally, creating infrastructure so people have access to low-cost, high quality health information is key for spreading true information and stopping false information during a pandemic. People need access to doctors, nurses, and community health providers that they trust. They need reliable news sources, including local and culturally-relevant news. And they need access to libraries and good schools where children and adults can learn about science, health, and the media. Healthcare, news, schools, and libraries are critical infrastructure for keeping people safe, healthy, and informed. 

 

Your question about communicating nuance is such an important one and we answer it below – keep reading! 

 

If you’d like to learn more about strategies for stopping misinformation, the Debunking Handbook has a great overview and it is available for free online: https://digitalcommons.unl.edu/scholcom/245

 

Question: How do changes in scientific understanding of a new disease contribute to misinformation? Can this be counteracted? (For example, in the early days of the COVID pandemic, scientists thought the virus couldn’t be spread without symptoms, but now we know that isn’t true.)

Answer: Despite many of us learning the scientific process in grade school, actual scientific deliberation has been occurring behind the closed doors of the academy for most of modern history. The only insights the public gets are those things which are published and reported on in the mainstream news. The way science is reported and written about in publications leads many to understand these things as some sort of ‘verified truth.’ Working scientists know that this is far from the case in just about any discipline, even though papers and pre-prints often use language that make it seem that it is. Nevertheless, by the time science gets to the public, there is usually some degree of consensus and validation of the results and that is what has built and maintained public trust in science. 

 

However, the early COVID pandemic was one of the first instances in which the general public was being updated on science in real-time. Especially in the era of pre-prints, this meant that instead of just seeing the end results, the public was seeing all the uncertainties along the way. Those uncertainties are normal and healthy in science, but only because scientists acknowledge and accept such uncertainty as being ever-present. At the beginning of the pandemic, however, there was a great deal of uncertainty which was not communicated to the public, and so the words of scientists were taken as truth. When those things turned out not to be true, many lost faith in “mainstream” science, leaving the door open for dis-/misinformation. 

 

At the workshop we recently held, panelists and attendees rallied around the idea that science should be more willing to say “we don’t know yet.” Instead of making shaky claims, saying “we don’t know” not only shows humility and prevents mistrust, it also closes the door on the possibility of misinformation by expressing that there is not enough data for anyone to make claims. 

 

Question: What are the most important ways you, as health professionals, and we, the public, can lay groundwork to improve health and vaccine science literacy, so that when the next pandemic rolls around, we have a populace with better foundational understanding to fight misinformation?

Answer: Answering this question as a communication scholar (and not necessarily a health expert!), I see two strategies that can work together to produce better science and health literacy: 

 

First, we need to prioritize health and science education in schools. Things like vaccines and epidemics are rarely covered in K-12 health and science classes, so it is no wonder that the general population has few tools to begin to understand things like epidemic spreading models and vaccine efficacy. These topics could easily be incorporated into classes, starting as young as kindergarten with basic concepts like how things spread from person to person, and how vaccines “teach” your body how to fight diseases. Kids experience these things every day - most kids have been sick and they have received vaccines. Building literacy from these common experiences is a great place to start. 

 

Second, we need to train science and health researchers and professionals to be better communicators. The truth is, for many of us in research jobs, there is not much incentive to communicate with the public. This is wrong, taxpayer dollars pay for most of our research and we need to do better! I think we should be offering training in public communication to all junior scientists, recognize public communication in hiring and promotion decisions, and give awards for outstanding health and science communication. 

 

I would also love to see scientists and regular people coming together around these topics in ways that are fun and engaging. Unless you’re really interested in science, it can be pretty boring to learn about epidemics and vaccines by reading science books and articles. But these things are amazing – vaccines literally save lives, they’re superheroes! There is no reason we could not have games and blockbuster movies about vaccines. Why do rockets and dinosaurs get to have all the fun in Hollywood?! 

 

Question: How can every day people identify medical misinformation when confronted with it in real life?

Answer: This is a great question! It is not an easy task! Fact-checking companies often hire experts or conduct lengthy reviews over individual facts, so for us everyday people who aren’t keeping up with every new publication coming out, it can be hard to spot. 

 

The best thing to do when confronted with any medical information is to search for the source material and supporting publications, but this isn’t really feasible for many of us that aren’t used to the jargon, don’t have access to the necessary journals, or simply have other things to be doing! In a pinch, doing a quick online search about the statement in question and considering the expertise of its author can be an effective strategy. There are reliable and free sources of medical information online such as https://www.mayoclinic.org/

 

All of this, however, is still a bit of work! And considering the sheer volume of information, even the simplest strategies for every day people don’t scale all too well. We can’t be doing a deep-dive into every post we see online. Thus, it seems worthwhile to think about the problem of misinformation. The harm comes not from simply seeing misinformation, it comes from believing, amplifying, or acting upon that misinformation. So, perhaps the best advice is simply to only believe, share, or act upon health information after you have tried some combination of the strategies above! 

 

Question: Has vaccine hesitancy worsened because of the COVID-19 pandemic? What can be done to combat it?

Answer: Unfortunately, it does look like vaccine hesitancy increased during and after the COVID-19 pandemic. There are a lot of reasons why this may be true – people went months or years without seeing a healthcare provider and many missed routine vaccines, vaccines became a politically polarized issue in some countries, trust in medical institutions declined in general, and anti-vaccine conspiracy theorists spread false and frightening information about vaccines during the pandemic. 

 

Getting people back to see their healthcare providers is of the utmost importance for reversing vaccine hesitancy. Trusted medical providers can help people understand the benefits of vaccines and discuss the merits of any particular concerns a patient might have. Making vaccines affordable and accessible is also a high priority, including funding community health clinics and outreach efforts in schools, churches, and other places where the public gathers. And working with media platforms to amplify accurate information and remove misinformation is key. Unlike in some areas of health research, vaccine efficacy and safety are settled science – we can identify claims about vaccines that are true and false, and most scientists and medical professionals will agree. This is one area where prebunking, fact checking, and removing false information may be especially effective.  

 

You can read more on this topic in this press release from UNICEF: https://www.unicef.org/rosa/press-releases/new-data-indicates-declining-confidence-childhood-vaccines-44-percentage-points-some

 

Question: Does most misinformation find its origins on the internet today? Can free speech be balanced with preventing the spread of misinformation?

Answer: Misinformation and disinformation are not new, people have been spreading lies and propaganda to manipulate, control, and gain power for as long as the media has existed (and probably before!) But, the Internet has made it easier and faster to spread misinformation, and recent advances in AI-generated content, including realistic fake photos and videos, will make it harder to determine what is real and what is fake. 

 

I think there are plenty of ways to balance the harm of misinformation against the right to free speech. Free speech rights already have limits in the United States. Some forms of speech are already not covered by free speech protections (e.g. libel, slander, threats, etc.). Other speech is limited in certain situations. For example, an adult can watch pornography in the privacy of their home, but they cannot broadcast it in a school. Similar blanket or situational restrictions could be placed on misinformation, limiting the most dangerous health misinformation during the most vulnerable times, such as a state of emergency during a pandemic.

 

But perhaps more to the point, we could compel platforms and content creators, by law or mutual agreement, to label or remove misinformation. My favorite everyday example of how this could work is the rating system on movies. It is not illegal to make movies with violent or sexual content, nor is it against the law to put that content in G movies. But, there is an industry standard based on social norms that movies are assigned ratings that more-or-less tell us whether we can expect to see violence, curse words, or sexual content. Movies designed for general audiences (G-rated) do not have that sort of content. We could similarly require platforms to make and enforce standards on misinformation. We could expect them to label or remove misinformation once it is identified, offer fact checking and community standards to identify misinformation before it spreads, and create an independent auditing board to monitor and report on the levels and types of misinformation that are circulating. We could also require safeguards and warnings for AI-generated content, including ethics reviews attached to research on new AI technologies. None of these restrictions infringe on freedom of speech. People may have the right to create whatever health mis- and disinformation they want, but we certainly do not need to guarantee them a broad and uninformed audience. 

 

Question: What happens when trusted authorities become sources of misinformation because of failure to keep up with the science? Can this be prevented? This seemed to happen during the COVID pandemic and now the trust is gone.

Answer: This is such a tricky problem. One major challenge with something like a pandemic is that information about public health and effective intervention strategies changes so quickly, it is hard to keep up. It is even harder for medical providers and public health officials, who are in charge of managing care and communicating pandemic responses at the same time. 

 

One strategy that has been shown to help is to simply acknowledge uncertainty. Saying “we don’t know yet, but we will know more in the future,” is a great way to be honest with the public and also avoid creating the impression that we will never know what to do. The more specific you can be about when more information will be available, and what is happening to produce that information, the better. 

 

Similarly, there is a tricky balance between getting people excited about an intervention, such as a vaccine, and being realistic about what it can do. The COVID vaccine is a marvel, in both its effectiveness and in how quickly it was developed, but it is not a silver bullet. People still catch COVID and it is still possible to spread COVID even if you are vaccinated. I think building vaccine literacy, bolstering the medical system so people have access to trusted providers who can provide realistic assessments, and using communication tools to paint a picture of the long-term vision of a world with COVID could all help people accept interventions while understanding their limits. 

 

Finally, I would also love to see the US and other countries invest in a “Health Communication Corps,” similar to the Army Medical Corps, that can be called up and deployed during a health crisis. There could be communication professionals, specifically trained in science and health communication, who can help track up-to-date health and science research during a crisis and work with public health officials, medical providers and journalists to communicate changing information to the general public.

bottom of page