Nuance versus certainty: The disadvantage scientists and physicians have in communicating risk


This post, although it is about an interview with a CDC scientist named William W. Thompson that resulted from the whole “CDC whistleblower” manufactroversy that’s been flogged relentlessly for the last two weeks, since antivaccine “heros” Andrew Wakefield and Brian Hooker released a despicable race-bating video flogging Hooker’s utterly incompetent reanalysis of a ten year old study that had failed to find an association between autism diagnosis and age of first vaccination, is about a more general issue as well, an issue that can apply to discussions of just about any trumped up risk, not just vaccines. It also demonstrates why we in the scientific community will always be at a disadvantage when it comes to countering fear mongering, exaggeration of risk, and even making up risks that do not exist, you know, the sort of things that antivaccinationists, the Food Babe, and anti-GMO activists (among others) routinely do.

I’m referring, of course, to the recent interview between the first author of the study that Brian Hooker “reanalyzed,” Frank DeStefano, and antivaccine former CBS reporter turned “independent” investigative journalist. Remember Sharyl Attkisson? On this blog, we first met her seven years ago, when she posted on the CBS News website an article so full of antivaccine crankery that it gave RFK, Jr. a run for his money. Since then, Attkisson has been reliably antivaccine, flogging a horribly incompetent review article by Helen Ratajczak blaming autism on vaccines, being a conduit for intelligence about happenings at CBS for the antivaccine crank blog Age of Autism, sucking up to Andrew Wakefield, abusing the Hannah Poling case, and, just last year, hideously misreporting the murder of an autistic boy while promoting Andrew Wakefield’s Autism Media Channel without telling anyone that the AMC belongs to Andrew Wakefield as well as Polly Tommey, in essence lying by omission. Most recently, she’s resurrected a tragic case of a boy who died to try to link it to vaccines.

In other words, if you’re the CDC press office, she’s the last reporter you’d want to give access to CDC scientists to. After all, she’s not even affiliated with a major news organization anymore; so you don’t even risk pissing off CBS by not granting one of its reporters an interview. Now, given the feeding frenzy of utter crankitude that’s been going on since Wakefield compared DeStefano’s study to the Tuskegee syphilis experiment and the crimes of Hitler, Stalin, and Pol Pot all rolled into one, who would be the last journalist to whom you’d want to grant access? Yep, Sharyl Attkisson. Yet, grant Sharyl Attkisson access to Frank DeStefano they did.

In a certain way, you almost have to admire the ballsiness, if it weren’t for the concern that this was done more out of incompetence than anything else. What you don’t have to admire is how DeStefano appears to be ill-suited to be interviewed by a hostile journalist. Or maybe he didn’t know that this was a hostile journalist, although he should have been prepped beforehand. In any case, even though DeStefano didn’t sound too bad on the audio, Attkisson, in her transcript of the interview, includes every single “um,” “er,” “uh,” or other pause is included. In most transcripts, these sorts of normal pauses are left out, because they are unnecessary. It was clear that Attkisson wanted to make DeStefano look as bad as possible. In any case, DeStefano repeats the usual defense of his study, similar to the CDC response. There’s nothing new here. What inspired this entire post as a sterling example of why scientists are always at a disadvantage in the PR war with cranks comes from a second part of the exchange, which was trumpeted across antivaccine blogs and Twitter feeds right after Attkisson posted it yesterday under the headline CDC: “Possibility” that vaccines rarely trigger autism. Here’s the key part of Attkisson’s spreading of fear, uncertainty, and doubt:

A coauthor of the questioned study is Dr. Frank DeStefano, Director of the CDC Immunization Safety Office. In a telephone interview last week, DeStefano defended the study and reiterated the commonly accepted position that there’s no “causal” link between vaccines and autism.

But he acknowledged the prospect that vaccines might rarely trigger autism.

It is a significant admission from a leading health official at an agency that has worked for nearly 15 years to dispel the public of any notion of a tie between vaccines and autism. Vaccines are among the most heralded medical inventions of our time. Billions of people have been vaccinated worldwide, countless lives have been saved and debilitating injuries prevented. The possibility that vaccines may also partly be responsible for autism, in individual cases, is not something public health officials are typically eager to address.

And that, my friends, is why scientists dealing with cranks are at an extreme disadvantage. Scientists and the reality-based community recognize in DeStefano’s statement nothing more than scientific open-mindedness and caution, the recognition that there might in very rare cases be an association that suggests that vaccines might trigger autism. He’s being cautious, but it’s clear from the context that he doesn’t really believe vaccines cause autism. In fact, take a look at the interview in context:

Attkisson: What about secondary?

DeStefano: Sec—I don’t understand what do you mean “secondary”?

Attkisson: What about not “causal,” but “as a result of” vaccines, as in the Poling case? The medical expert found, you know, as a result of the damages she had from the vaccines, she ended up with autism. And the distinction was made in the medical expert, ‘well, that’s not ‘causal’, it’s sort of a ‘but for’ but it’s not a ‘causal.’

DeStefano: Yeah, I mean, I mean in that case, you know, she had a, I mean, you know, she had an underlying uh biological illness that uh either vaccination, or it could’ve been an infection that that would trigger some physiological stress in her, uh, seems to have, you know, could’ve, could’ve caused uh, um, manifestations that, characteristics of autism which, you, you know, appears to be what happened in her case.

Attkisson: But I mean doesn’t that, is—isn’t that a “link”? It’s not a “causal” link, but isn’t that a potential link between vaccination and autism if certain children with a “underyling biological illness” can have a “trigger” through vaccination?

DeStefano: [Unintell] as you call it, a secondary link if you wanna call it that way, w– in certain children, I mean ri—I mean, I, maybe that, but, you know, then I guess, that, that is a possibility.

Attkisson: Do you think that’s an important area of study so we could figure out which kids might have that predisposition?

DeStefano: uh, [phone noise] Yeah, I mean, I think um…You know, I think it’s something that, uh, well I mean, you know, in terms of uh… I mean, It’s hard, it’s hard to say, you know, I mean it’s like, um…I mean how how important that is. I mean, it’s a theoretical possibility, I guess the, the Poling case maybe suggested it could happen. Uh, but [unintell] cause it’s hard to predict who those children might be, but certainly, um individual cases, uh, can be studied to try to, uh, to look at those, uh, those possibilities.

Yes, predictably, Attkisson is flogging the Hannah Poling case again, which, as has been pointed out many times before, was a “rebranding” of autism, in which children with mitochondrial disorders can develop encephalopathy if they have significant fevers. In fact, the recommendation is that children with such disorders be vaccinated. The whole claim that children with mitochondrial disorders develop autism after vaccination is yet another highly dubious claim with no compelling evidence to support it. It’s how the story morphs, just as the story about DeStefano’s paper is morphing now.

Unfortunately, DeStefano blundered right into this issue, either insufficiently familiar with antivaccine talking points on the issue or unable to articulate a message about just how uncommon mitochondrial disorders are and how the risk of not vaccinating such children far outstrips any risk of vaccinating. But the case of Hannah Poling is a side issue to a more general problem when science is communicated to the public by scientists. The real issue is this. When scientists say, conservatively and scientifically correctly from their point of view, that something, say vaccines, “might” cause a problem, no matter how unlikely causation might be (e.g., it’s “theoretically possible,” as DeStefano said, that under some circumstances in some children vaccination might “trigger” autism), what cranks hear is that something does cause a problem, and what the general public hears is that a scientist is saying that concerns about that something causing a problem are plausible—even not unlikely. The same phenomenon is common for any risk factor in the environment about which there is a lot of unjustified fear mongering, such as genetically modified organisms (GMOs) or ingredients in food. Indeed, such statements by scientists are part of the reason why even the risibly ignorant “Food Babe” can get traction when she decries the “yoga mat chemical” in bread or—gasp!—calcium phosphate and, reminiscent of antivaccine lies, “antifreeze” in beer.

It all comes down to this. Scientists see nuance everywhere. It’s our job, particularly in biomedical sciences, medicine, and epidemiology where it’s rare for anything ever to be 100% “proven.” Moreover, it’s rare, leaving out pseudoscience like homeopathy and reiki, for anything ever to be completely “disproven.” Indeed, those of us in the biomedical sciences tend to dislike even using the words “prove” or “proven” in the context of a hypothesis. That, however, is not how non-scientists think, and it’s even less how cranks, such as anti-GMO, antivaccine, and Food Babe-type activists think. Non-scientists frequently confuse correlation with causation and vastly overestimate certain risks. Not uncommonly, they seem to think that there should be no risk whatsoever to something like vaccines, when such a goal is completely unattainable for any medical intervention, regardless of how benign.

As I’ve pointed out time and time again, people happily hop in their cars every day, even though the risk of serious injury or death in a car crash is orders of magnitude greater than any risk of injury from vaccines. Several years ago, I pointed out that about four children per year in the U.S. die as a result of an injury from playing baseball or softball, an average that, as of the mid-1990′s, had not changed appreciably since the 1970′s. But deaths are only the worst injuries. Among children ages 5-14, around well over 100,000 children per year are taken to the emergency room for injuries suffered playing baseball. These include sprains, contusions, fractures, dental injuries, head injuries and concussions, and internal injuries. In fact, baseball has the highest child fatality rate of any sport. And those are just the injuries severe enough to cause the coach and parents to take the child to the emergency room. In fact, if you look at all sports injuries in children, there are approximately three million sports injuries per year! Meanwhile, An estimated 5,000 children ages 14 and under are hospitalized due to unintentional drowning-related incidents each year; 15 percent die in the hospital and as many as 20 percent suffer severe, permanent neurological disability. Yet, antivaccinationists focus on the incredibly tiny risk of vaccines compared to the benefits and happily let their children play baseball, swim, be driven around in cars, and all sorts of far more risky behaviors than receiving a few shots.

Scientists, being scientists, must acknowledge even the “theoretical possibility” of infinitesimal risk and then weigh that against other risks, such as the risk of disease in the case of vaccination. The balance of risks versus benefits clearly comes out on the side of vaccinating, but because the benefits of vaccinating are not so clearly seen any more (vaccine-preventable diseases having been reduced to very low levels by mass vaccination programs), parents tend to focus far more on even tiny risks of vaccinating, even risks that aren’t scientifically supported, such as the “risk” of autism.

So what’s the way around this? We as scientists and physicians can’t be like pundits on the news, dealing in absolute statements. (“Only a Sith deals in absolutes.”) That wouldn’t be intellectually honest. We can, however, be very aware of how the general public will perceive our statements (and how cranks will try to cherry pick from them) and be very careful to make them (1) bullet-proof as far as quote mining goes and (2) reflect the correct level of risk. DeStefano failed in both of these areas. However, his failure might remind those of us who try to communicate science to the public, be it GMOs, various environmental risks, whether real or perceived, or food safety, or whatever, to avoid these pitfalls. We are already at a disadvantage when it comes to communicating science and countering crankery and quackery. We shouldn’t dig ourselves into an even greater hole.

Oh, and we shouldn’t agree to be interviewed by quacks and cranks, like Sharyl Attkisson.