FYI

As New Mexico health officials gear up to tackle vaccine hesitancy, a new LANL study examines how COVID-19 conspiracies spread on social media

by

Earlier this week, New Mexico’s health department began allowing all New Mexicans 16 years and older to self-schedule their COVID-19 vaccines. This move—forecast the week prior by Health Secretary Dr. Tracie Collins—signaled yet another inflection point in the pandemic’s trajectory as officials move from managing the vaccine-eager to addressing the vaccine-hesitant.

As I write this, close to 700,000 New Mexicans have been fully vaccinated, with the state receiving national recognition mid-month as it passed the 50% mark for residents with at least one dose, “surging,” the New York Times wrote, “past states with far more resources in the race to achieve herd immunity against the coronavirus.”

Nonetheless, New Mexico, like the rest of the country, is now preparing to address folks who, for whatever reason, don’t want a vaccine. In an April 21 news conference, Collins said that while interest in the vaccine had increased 40% over the last month, “…there are certain communities and pockets of communities where people are skeptical or hesitant” and the state is in the process of reviewing data about those areas in order to create a program in which “community champions…folks who represent these communities that might be skeptical” will act as surrogates for the health department and relay information intended to assuage concerns. The department also will continue holding town halls to “really get beyond misinformation,” Collins said, “this idea that vaccines are not safe, which is not true. We’ve really got to be available not only to understand, ‘do people not like vaccines?’ but what sort of misinformation have they received. And we have to be able to correct that.”

A comparable program, already running, uses the same philosophy. “Trusted Voices” has close to 50 people and counting from across the state discussing their experiences getting the vaccine and encouraging people to get their own (you can also nominate folks to be a “trusted voice” in the campaign at togethernm.org/trusted-voices-nomination).

Certainly, correctable misinformation may be driving some vaccine hesitancy. A US Census Bureau report at the end of March calculated the percentage of people in New Mexico hesitant about vaccines at close to 11%, with nearly 50% of those citing concerns about side effects; 39.1% simply not trusting them and just over 37% saying they’re waiting to see if they’re safe.

I asked DOH Deputy Cabinet Secretary Dr. Laura Chanchien Parajon if 11% sounded accurate, and she told me that in terms of the data the state is seeing, the percentage of people hesitant about receiving a COVID-19 ranges—anywhere from 11% to 20%. So far, she says, apprehension about cost and identification, particularly from immigrant populations, has emerged as a top concern, with the state responding with a campaign reinforcing that the vaccines are free and no government ID is required to receive one. Parajon says the state wants to be “data driven” in its vaccine hesitancy campaigns, and is using not only its own data collected from its outreach efforts, but from partners, such as Presbyterian Health Care.

But beyond data, she said, DOH wants to directly respond to concerns.

“What we want people to know is we want to partner with people. We want to partner with communities and listen to what people’s concerns are. We want to be able to answer them. That’s why we have the town halls: We want to answer your questions.”

I’ve listened in on some of the town halls and most of the questions people have seemed, to me, fairly reasonable. People want to know about how to arrange for a home-bound vaccine; they want to know the latest on the Johnson & Johnson vaccine (as of this writing, the state was set to start administering it again); they want to know when people younger than 16 will become eligible.

And then there are those questions that drift from the practical into the less moored territory, such as: “Does the COVID-19 vaccine involve a microchip of any kind?”

Dr. Jennifer Pyeatt, president and pharmacy director of Covenant Health in Hobbs, took that one on during a recent town hall.

No, Pyeatt said: “There is no microchip. I’ve pulled up thousands of those vaccines and it would be impossible to get a microchip through that tiny little syringe.”

I heard the question as I was reading through a new study from researchers at Los Alamos National Laboratory, published recently in the Journal of Medical Internet Research. Titled “Thought I’d Share First,” the study used publicly available, anonymized Twitter data to characterize four COVID-19 conspiracy theory themes and provide context for each through the first five months of the pandemic. Those themes were: 5G cell towers spread the virus; the Bill and Melinda Gates Foundation engineered or had otherwise malicious intent related to COVID-19; the virus was bioengineered or developed in a laboratory; and that the COVID-19 vaccines (still in development at the time of the study) would be dangerous.

Ashlynn Daughton, an information scientist in LANL’s Information Systems and Modeling Group and co-author of the study, told me in an interview she had been interested in health and misinformation for several years and had done some work on the topic during the emergence of the Zika virus in 2015 and 2016. But “all of the circumstances surrounding the emergence of COVID,” including its scope and how quickly the science was evolving in real time “made the importance of studying health information crystalize,” she says. “And it was pretty easy to identify some things that were interesting and seemed kind of relevant and seemed [to be] popping up repeatedly.”

The study’s title, she says, references “a handful of different tweets, all in the same vein [basically saying] ‘I haven’t vetted this, but here’s a link to an article I saw.’ It’s an homage to how easy it is to re-share things without critically examining them.”

Starting with a dataset of approximately 1.8 million tweets (Facebook’s terms of service prevent scraping its data), the researchers then used the data collected for each of the four theories and built artificial intelligence models that were able to categorize tweets as COVID-19 misinformation or not. But the study also required a great deal of manual labor, with the scientists reading the tweets and labeling them (I wanted to know if reading those tweets had been funny or depressing and Daughton said it had been both).

One of the most interesting observations from the study, Daughton said, was “there seems to be evidence that [conspiracy theories] kind of grow and intertwine as time goes on, and we saw a lot of overlap.” This was less true for the 5G theory, which was the smallest subset and the most contained. The Bill Gates conspiracy theories, on the other hand, “were very diverse.”

While the data from this study now dates to the earlier part of the pandemic, it’s become increasingly clear that understanding how such theories evolve could be helpful in the future. Daughton, who has a public health background, says one of the goals for future work is “to continue to explore how we could move this more into the public health communication realm and develop stronger collaborations with public health departments and figure out which of these components are more helpful to them.”

My guess is all of them.