Home » Columns, Stats4Good

Data for Good Can Drive COVID Vaccine Confidence

1 February 2021 908 views No Comment

David CorlissWith a PhD in statistical astrophysics, David Corliss leads a data science team at Fiat Chrysler. He serves on the steering committee for the Conference on Statistical Practice and is the founder of Peace-Work, a volunteer cooperative of statisticians and data scientists providing analytic support for charitable groups and applying statistical methods in issue-driven advocacy.

As the COVID-19 pandemic has unfolded, we have seen a series of challenges. One many statisticians can help address, even without any particular medical expertise, is the distrust of vaccines.

Many Americans are reluctant to receive an inoculation when a new vaccine arrives on the scene. Years of deliberate destruction of public trust in science and data, coupled with very real incidents of abuse and neglect, have reduced confidence in clearly life-saving vaccines to a dangerously low level. As a result, a substantial portion of people have adopted a “wait and see” attitude or even refuse to be inoculated. While few of us have been involved with vaccine trials, all of us as statisticians can work to address and reduce the instances of vaccine hesitancy.

Not all root causes of vaccine hesitancy are based on falsehoods. The infamous Tuskegee Syphilis Experiment from 1932–1972 is an important cause of vaccine distrust among often underserved and highly vulnerable populations such as people of color. As a result, this 20th-century crime against humanity is still killing people today by leading many to avoid COVID-19 vaccines.

As advocates and practitioners in Data for Good, it is important for us to be aware of problems both historical and present so we can respond to concerns and help drive vaccine adoption. One often-stated concern is that the vaccine has been rushed and the medical establishment is, in effect, experimenting on early adopters.

Another important factor is widely circulated reports of medical staff disregarding symptoms of patients who are often women and persons of color such as Dr. Susan Moore. Knowing the primary reasons for this deep-seeded distrust allows us to address the questions driving vaccine hesitancy.

Data for Good advocates need to be aware that statistics alone rarely convince nonstatisticians of anything. Telling human stories and conveying experiences to which people can relate before presenting supporting statistics is much more effective. To address vaccine hesitancy, we must first listen to and fully acknowledge people’s concerns, both historical and ongoing. Once a relationship of trust is established, we can offer answers to their questions.

The vaccines were not rushed. In the crisis of the pandemic, many teams collaborated as never before, working long hours to complete the standard approval process without omitting any steps. The vaccines aren’t just being tested on a few people in a few places. The trials are worldwide, with testing and approval in many countries. People who don’t trust US trials or approvals can be referred to statistical studies in other countries. For example, the BBC reported the Pfizer / BioNTech vaccine “has been tested on 43,500 people in six countries and no safety concerns have been raised.”


Science alone simply isn’t enough to fight this pandemic and, to this virus, irrational fears are just as deadly as rational ones.

We need to listen and patiently respond to “yes, but…” statements. Always keep in mind the barrier to acceptance is emotion, not information. Be sure to include examples of real people with good reason to have the same concerns who are choosing to take the vaccine.

One phrase I’ve heard statisticians, epidemiologists, and other experts use over and over is “public health moves at the speed of trust.” It’s a powerful message, underscoring the critical importance of developing relationships of trust in the public health space. But the concept was already familiar to me, a message I first heard while working at a large manufacturing company where Simon Bailey taught business analytics moves at the speed of trust. It’s a principle that can be applied across all areas of Data for Good: We can build the models, but using them to make a difference requires building relationships, as well.

I certainly don’t mean to suggest a false equivalence between myth, pseudoscience, and deliberate deception on one hand and scientific fact on the other. I am saying the science alone simply isn’t enough to fight this pandemic and, to this virus, irrational fears are just as deadly as rational ones. This is an all-hands-on-deck event in the D4G community. While we don’t all work in medicine, pharma, or public health, we have our own circle of connections in which we can build trust to help save lives. It’s a principle to remember, whatever the particular subject area and methods applied: Data for Good always moves at the speed of trust.

Getting Involved

Coming up February 17–19 is the ASA’s Conference on Statistical Practice (CSP). It’s a virtual conference this year, making it easier for many to attend. CSP includes a wealth of excellent D4G content. Analysis of the COVID-19 pandemic includes the keynote by R. David Parker, several papers, and the annual ethics panel.

Another opportunity I would like to mention is an interesting data set on animals in shelters. The Basic Animal Data Matrix incorporates selected shelter data from several leading organizations. For those interested in statistical analysis of the drivers of more positive outcomes, this can be a great resource.

1 Star2 Stars3 Stars4 Stars5 Stars (No Ratings Yet)
Loading...

Comments are closed.