Will Confirmatory Bias Keep You from Getting COVID-19 Vaccine?

In this episode of Lehigh University’s College of Business IlLUminate podcast, we are speaking with Shin-Yi Chou and Ernest Lai about their research examining the role that confirmatory bias plays in the decline in vaccination rates for measles, mumps, and rubella – or MMR – and what that might mean as a vaccine for coronavirus is developed.

The morning we recorded this podcast, it was announced that President Trump and his wife Melania tested positive for coronavirus. The COVID-19 pandemic has thrust the whole issue of vaccines front and center in the news over recent months, as more than 200,000 people in the United States, and more than a million worldwide, have died as a result of the coronavirus.

Dr. Chou holds the Arthur F. Searing professorship, and is the chair of the Department of Economics and Lehigh's College of Business. Her research focuses on health economics and applied econometrics. Dr. Chou also is a research associate at the National Bureau of Economic Research.

Dr. Lai is an associate professor who holds the Class of '61 professorship in economics. His research focuses on using the tools of game theory and laboratory experiments to study strategic communication. Dr. Lai teaches game theory at the undergraduate and graduate levels, and is the director of the Ph.D. program of the College of Business.

They spoke with Jack Croft, host of the ilLUminate podcast. Listen to the podcast here and subscribe and download Lehigh Business on Apple Podcasts or wherever you get your podcasts.

Below is an edited excerpt from that conversation. Read the complete podcast transcript. Their research was published in the March 2020 issue of the Journal of Health Economics.

Jack Croft: Confirmatory bias … sounds pretty obvious, and I'm guessing that it means that when we're looking at information, we're looking for things especially that confirm the bias we start out with. Is that correct?

Ernest Lai: Yes. I think the idea traced back to, actually, Francis Bacon, right? He had a quote that the human understanding, when it has once adopted an opinion, will draw all things to support and agree with it. It was his observation. But then psychologists have documented this in the laboratory. Once we form a position, form a belief, then we are becoming deaf to new information — in particular, information that contrasts with our initial beliefs, right? We will be selectively selecting information, we will be selectively interpreting information that fits our prior narratives. So I think that is the main idea of confirmatory bias. In our study, we document that in the context of health decisions, using what we call the real-world data.

Croft: And what kind of data were you looking at?

Shin-Yi Chou: The primary data is the National Immunization Survey, and we look at the periods from 1988 to 2011. So we know the MMR non-uptake rate among the children surveyed. And in addition to this main data, we also supplement our data with many other measures. So we look at the role of three factors to facilitate our findings. We look at the prevalence rates of relevant diseases. We look at the counts of relevant coverage in local newspapers. And we are also interested in the intensities of relevant online searches. All those are measured at the state level. And our underlying pre-assumption is that vaccination decisions of mothers with more frequent onsets of the diseases or more relevant newspaper coverage or more intense online searches of related topics are more strongly affected. At the end, we find that online searches have stronger impact on the biases of college-educated mothers than does newspaper coverage and disease outbreak. And we also find that both positive and negative exposures of online searches have a significant effect, although the effect of negative exposures remains dominant.

Croft: And what do you mean by positive or negative exposures?

Chou: For the online search, positive means the information that asserts that the vaccine is actually safe and effective. We went through the information, and then we grouped the information to positive and negative. Negative information means that the information asserts that the vaccine is not safe, is not effective.

Croft: It seems kind of surprising at first blush that college-educated parents in particular would be the ones most susceptible to negative information, false information, about vaccines. And I was wondering if you could talk about why that would be.

Chou: I think more educated parents have ways to obtain information faster and form their own beliefs. But when the information turns out to be erroneous, they will stick to the misinformation they believe in. So it shows us that the first perception when they formed the beliefs, it's quite important, the information that was floating at that time. We find that exposures to negative information strengthen the biases of the college educated mothers more than the exposures to positive information attenuated them. So it suggests that the effects of information linger when people amplify incoming information that confirms their erroneous beliefs. Once they formed their beliefs, they just follow through and stick to the information.

Lai: It's not so much about the educated groups or parents are more vulnerable to false information. It just so happened the first set of information they acquired about vaccines [based on a 1998 study published in the medical journal The Lancet that was later fully retracted] was not accurate, was false. And then, under the confirmatory bias, that sticks. Then it translates into the outcome that they are more vulnerable to the false information. So there are two separate concepts or issues here, right? They are better at absorbing information, and it so happened the information that they absorb initially turned out to be wrong. We have to see this by combining the absorption of information by the more educated parents with confirmatory bias, and that explains the findings.

Croft: I recently read a report from CNN talking to officials at the CDC and looking at [social media], and people at the CDC were admitting that they have not done a good enough job getting information out there about the coming coronavirus vaccine, in particular. And that in the absence of them having a public education campaign with what the facts are, that that void has been filled online by all kinds of false and even just pretty crazy information, ranging from the vaccines are part of a CIA plot to take over the world, that it's going to leave an invisible digital trackable tattoo on you, that it will hideously disfigure your face, and maybe my personal favorite, that Dr. Anthony Fauci, the CDC's leading expert on infectious diseases and a leading vaccine advocate, is actually Satan. How do you even begin to counter that kind of attack on scientific information and health information?

Chou: In the case of a COVID-19 vaccine, I think the scientific community needs to step up to scrutinize, for example, the clinical trials conducted to test the effectiveness of vaccines. So I think the studies need to be transparent. They need to publish the data so that the entire scientific community can evaluate whether the study is rigorous or not, and whether the public can trust their findings or not.

[With] the COVID-19 vaccine, time is of the essence here, so we all want to have it as soon as possible. But we really need to make sure that the public trusts the vaccine. So the first information disseminated to the public is crucial. According to our study, the first perception, once planted, is hard to eradicate. And people are going to stick to the misinformation and erroneous belief about the vaccines. So I think here, the health authorities, at every level, need to take a significant role in disseminating transparent information. I think that's the key.

Croft: Another recent survey by the Kaiser Family Foundation found that 62% of Americans are worried that a coronavirus vaccine is going to be rushed to approval without making sure it's safe and effective. So in terms of that first impression that people have, 62% of the American people have now formed that opinion of, "Oh, when this thing comes out, it's going to be rushed, and it won't have been properly vetted, so it may not be safe and effective." You're starting from a deficit in terms of trying to rebuild trust among the public right from the beginning now, it sounds like.

Lai: Yep, and that's devastating, I think, given our findings, right? The first impression really matters. That was the first impression that the public got, then we can predict that they are fighting an uphill battle to really promote the safety of the vaccines. Personally, I don't have a solution for that, but I'm just pointing out that in the future, or whenever there's any occasions, public officials or scientists have to make sure that they convey the correct information, the accurate information. So we have to be careful about what is the information that was first being released to the public, to ensure that they are proper and accurate.