“Doctor’s death after Covid vaccine is being investigated,” read a headline in the New York Times on January 12, 2021.
Does that frighten you? Let’s look at the first paragraph of the story:
Health authorities are investigating the case of a Florida doctor who died from an unusually severe blood disorder 16 days after receiving the Pfizer coronavirus vaccine.
The story so far unequivocally links the Pfizer/BioNTech COVID-19 vaccine to a death. People firmly in the anti-vaccination world had a field day with this news that a physician in Florida died after being vaccinated. Most people are not ardent “anti-vaxxers,” but many—especially back in January—were vaccine hesitant, worried, that is, that the new vaccines were developed so rapidly that shortcuts had been taken establishing their safety. A headline and first paragraph like this one are just what the vaccine hesitant were afraid of—a seemingly clear link between the vaccine and death.
The story continues in this vein until the fifth paragraph when we finally get some context:
About nine million people in the United States have received at least one shot of either the Pfizer or Moderna coronavirus vaccine, the two [at that time] authorized in the United States. So far, serious problems reported were 29 cases of anaphylaxis, a severe allergic reaction. None were reported as fatal. Many people have had other side effects like sore arms, fatigue, headache or fever, which are usually transient.
So let’s say just for a moment that the doctor’s death was in fact a result of the COVID-19 vaccine. This paragraph tells us that it is one death out of about 9 million people who had received an authorized COVID-19 vaccine by the beginning of last January. How do we convey to people that that risk is essentially non-existent once we consider the risks we accept in our daily lives? The risk of dying in a car crash is about 10.7 deaths out of 100,000 people. The odds of getting hit by lightning in a given year are about one per 1.2 million people. Perhaps more important is the fact that the odds of dying for a 55 year old man if he gets infected with COVID-19 are about one in 10. If you drive a car or walk outside during a thunderstorm you should not worry about the risk of dying from a COVID-19 vaccine. You should be much more worried about getting COVID-19 than about having a COVID-19 vaccine to prevent that from happening.
None of this is laid out for the reader explicitly even in that fifth paragraph of the New York Times story when some data are finally offered. It would not be a big surprise if reading the story increased one’s fear of the vaccines and perhaps even induced the common “wait-and-see” attitude that many people have taken about them, that is, waiting and seeing how others fare from being vaccinated before agreeing to have one oneself.
We do not believe it was necessarily the intention of either the New York Times editor who wrote the headline or the reporter who wrote the story to stoke fears of vaccines or to drive people away from being vaccinated against COVID-19. We also do not question that the public has a right to know if there is a death associated with vaccines. Rather, we are concerned about the way in which this story and so many others like it are written and their placement. We question whether stories about science and health that entail a great deal of uncertainty belong on the front page. Perhaps they should be run in the science section where there may be more room to explain nuance.
Stories are Stickier Than Data
We know that most people do not read newspaper stories beyond the headlines. In order to get readers to go a little farther, editors, who write headlines, need to make them as dramatic as possible, sometimes bordering on what is called “click-bait.” When we read a novel, we expect to have to wait a while, sometimes until the end, to get to the really exciting part. Not so when we read news articles. Journalism students are taught they must put what is the most exciting part of a story in the first paragraph or two, acknowledging that that is about as far as most readers will get.
Our own research, done in collaboration with Fluent LLC, shows that stories about negative events are far more “sticky” in people’s minds than are explanations that involve data. So, for example, when the rare death occurs following a COVID-19 vaccine, we are treated to a lively account of the individual’s life story. This creates a lasting memory. Deaths from COVID-19, by contrast, are typically presented only with numbers; we are regularly told the number of people who have died—562,000 in the U.S. at the time of writing this article in early April. Although that is a tragically big number, numbers don’t stick with us as well as stories. Importantly, we are given very few dramatic stories of individuals who have succumbed to COVID-19 in the news. So, although there are many, many more deaths from COVID-19 than there are deaths following COVID-19 vaccinations—none of which have yet even been definitively attributed to vaccines—the emotional balance is exactly the opposite of this reality, in large part because of the way the stories are told.
Looking at the New York Times story, then, we see that from a journalism course point of view, it is nearly perfect. The headline is dramatic and makes you want to read more. The first paragraph is similarly dramatic—a death is being linked to the vaccine. The boring stuff about the actual risk of this happening doesn’t appear until paragraph five, before which point many readers will have quit reading.
A Better Example
On April 8, a story in the Atlanta Journal-Constitution began with the headline “Coroner: Man who died after vaccine died of natural cause” and goes on to explain that no link could be established between vaccine and the cause of the Florida doctor’s death. That means that almost two months elapsed between the time the New York Times reported the possible link and the coroner’s report stating no such link could be established appeared in the media. It is impossible to know how many people may have been dissuaded from being vaccinated by the first story and unlikely that the second one restored their faith in vaccine safety. As Miles Parks noted last March in an NPR report, “The odds of dying after getting a COVID-19 vaccine are virtually nonexistent.” That bold fact does not appear in the New York Times story.
Here’s an example of the way we believe these things should be reported. A March 31 headline in the Sacramento Bee reads “Sheriff linked vaccine to death despite experts’ cautions.” This story concerns the death of a California man who died after receiving a COVID-19 vaccine. The local sheriff immediately went on record linking the death to the vaccine even though the local coroner insisted that was premature pending an autopsy. Note that right in the headline there is the warning that the purported link may not exist.
In the very first paragraph of the story, this caveat is again made clear:
A California sheriff announced in January that a man died hours after receiving a COVID-19 vaccine even though his county’s health officials said the declaration was premature and detrimental, The Sacramento Bee reported, citing emails obtained under the state Public Records Act.
One thing we would have preferred is that the opening to the fourth paragraph in the Sacramento Bee story had come even earlier in the story: “Eventually, it would be determined that the vaccination was coincidental to the death…”
There are of course journalists from some fringe media who deliberately twist the facts about health issues like COVID-19 vaccines, taking things out of context and cherry-picking studies in order to make spurious points. Here, however, we are talking about responsible journalists from mainstream media who want to give the public correct information. These journalists do not generally intend to manipulate readers’ emotions to drive them toward incorrect ideas, such as that there is a link between COVID-19 vaccines and deaths. Inadvertently (and we must acknowledge sometime even here purposefully), however, this is exactly what they sometimes do.
We fully understand as well that journalists and editors need to have their stories read. Science when properly explained is not generally dramatic. It proceeds at a slow pace as scientists take their time to design and conduct experiments, review and analyze data, and then write their manuscripts. One study rarely is sufficient to change scientific theories or alter the course of treatment patients get for a particular disease. This disconnect between the slow acquisition of scientific knowledge and the need to write stories that are sufficiently dramatic to gain readership inevitably leads to headlines and stories that fail to place things in their proper context. It may sound boring to explain to readers that a one in 9 million chance of death is a virtually non-existent risk but failing to do so makes a very rare death seem much more frightening than it should.
Recently, there have been reports of people developing blood clots following receiving an AstraZeneca COVID-19 vaccine. The vaccine is not yet authorized in the U.S., but around the world about 25 million people have received it and there are about 18 reported deaths from blood clotting following vaccination. Headlines about this could be written in one of two ways:
AstraZeneca Covid Vaccine Linked to Rare Blood Clotting Disease
Rare Cases of Blood Clotting Disease Linked to AstraZeneca Vaccine
Both headlines would be correct, but it is easy to see how the former is more frightening than the latter. The first headline does not put the blood clotting disease in its proper context vis a vis the vaccine; the second one makes clear that serious blood clotting is a rare phenomenon following vaccination.
Guidelines on Reporting
There is precedent for offering journalists and editors guidelines on how to safely report certain kinds of events. After it was observed that media attention to suicides can increase the risk for further suicides, a phenomenon called suicide contagion, several agencies issued recommendations on how to safely report about suicides in the media. These recommendations include not going into details about the method of suicide and showing that help is available for people with suicidal thoughts.
Critica is now working on developing similar recommendations for reporting on topics involving health and science. In doing so, we understand there must be a balance between providing the public with information it needs to know in readable fashion with the duty to present accurate information that is put into its proper context. Newspaper stories cannot read like published scientific papers, but they also must not be so dramatized that readers will be misled and frightened. Our recommendations will largely concern the placement of context and use of narratives within a news story. From the headline on through the first few paragraphs a proper news story about a science or health topic must include something about the context and significance of the finding upon which the story is based.
Explaining risk is not easy and humans are not programmed to grasp risk rationally. We all tend to overestimate small risk and underestimate large risk. In the case of COVID-19 vaccines, it is easy to see how we might come to overestimate the risk of death after a vaccine and underestimate the risk of death from COVID-19 itself. Nevertheless, responsible journalism has to do a much better job explaining risks and telling stories that reflect the real risks. We will keep all of our readers posted as we come up with our own recommendations.