Framing vaccines, revisited: The “empathy” gambit

NOTE: This post, which is related to a discussion of Dr. Paul Offit’s Book Autism’s False Prophets, originally appeared over at The ScienceBlogs Book Club. However, now that the book club for this particular book has concluded, I am free to repost it here for those who may not have seen it and to archive it as one of my own posts. Besides, I know the antivaxers are more likely to see it here…

On Friday, while discussing what is perhaps the aspect of Autism’s False Prophets that is at the same time the most important set of observations (namely, how the media and government miscommunicate science and how the public seems hardwired to misunderstand science) and its most glaring omission (namely, suggestions how to overcome this problem), I talked about “framing” or how we could potentially represent the current science on vaccines in a compelling way that will be persuasive to the bulk of concerned parents. We know that hard core antivaccinationist parents will not be persuaded by virtually anything we say, but they are relatively small in number. It may not seem that way, given how noisy they are and how effective they’ve become at propaganda and media manipulation, but they are. Far more numerous are parents who hear the alarming rhetoric of antivaccinationists and wonder if maybe there’s anything to it all. They are the target audience.

In response, I got an e-mail from someone who wishes to remain anonymous. I will paraphrase what he said to some extent, because I don’t want to risk “outing” him, but one some carefully selected excerpts should be safe:

We are in trouble on this issue because we see it as a scientific issue, and if we just line up the science, people will be reasonable and decide that we are right and Jenny McCarthy is wrong.

I see his point there. It’s really no different than, for example, the evolution/creationism culture war. Evolutionary biologists know that the evidence is so overwhelmingly on the side of the theory of evolution and against creationism or its bastard offspring “intellgent design” creationism that they do tend to live under the delusion that, if the public were just informed of the science and the evidence, they would come to the reasonable conclusion: namely, that evolution represents the best theory of the origins of life that science has to offer, and creationism is without foundation in science. However, this attitude neglects the influence of fundamentalist religious beliefs, which are in direct opposition to the science of evolution and lead fundamentalists to reject evolution in spite of the overwhelming evidence in favor of it. When it comes to the whole issue of vaccines and autism, the evidence is nearly as overwhelming against the claim that vaccines or mercury in vaccines causes autism. So, a reasonable scientist thinks, just showing the overwhelming evidence that fails to support the hypothesis that vaccines cause autism should bring reasonable parents around. However, it can be argued, that viewpoint neglects the overwhelming emotional attachment parents have to their children, the malign effect of antivaccinationist propaganda which tells them that their child was somehow damaged and somebody must be made to pay, and the normal human cognitive quirks that make us seek patterns and be fast to see correlation as causation. I can see that.

My commenter then described a seminar about risk communication that he had attended and one presentation in particular. In this presentation it was described what’s important to people under “low concern” and “high concern” conditions. The presentation

showed two pie charts, on what determines credibility under “low concern” conditions (expertise and credentials are what matters) and under “high concern” conditions – and under the latter, it’s not expertise/credentials, but empathy and caring, and the big collective “we” are not good at that. I have great respect for Paul Offit (who, by the way, writes more books than I read – well, almost) but he is wrong on what we need to do to deal with this.

Instead of being clearer that vaccines don’t cause autism, we need to not overstate what we know (lots of studies have looked at MMR and thimerosal, and no evidence that either of them are a cause of autism at the population level).

I’m not entirely sure I see where scientists are “overstating” what they know, but more on that later. He then cited two posts by Peter Sandman:

Thimerosal, autism, and misleading toward the truth
Does taking the thimerosal out of vaccines reassure people or scare them?

From these he boiled down Peter Sandman’s message thusly:

…we need to stop gloating when we say that. We should express regret – because, if thimerosal had caused autism, well, we’d be through it by now. We would have had a tough couple of years, but we’d have vaccines without thimerosal by now and families who felt like their families were being torn apart by the stresses of dealing with a severely affected child would have have been spared that.

While there are some interesting and potentially useful suggestions in the links above, I have to strenuously disagree with this last sentiment. Strenuously. I will agree that gloating and rubbing antivaccinationists’ face in the exoneration of thimerosal as a cause of autism is not a good idea as far as communication of science and risk go (and, remember, I’m not a risk communicator; so I frequently lapse in this area, especially when it comes to Generation Rescue and Jenny McCarthy), but Sandman is taking it too far to the other extreme. There is nothing to be sorry for, and I find it strange that he would castigate scientists for supposedly “exaggerating” the science exonerating vaccines (something I also disagree that scientists do) and then tell us to exaggerate “sorrow” that a nonexistent risk believed by the public was found to be without a foundation in science. What’s there to be sorry about, other than at the money and effort being wasted chasing a failed hypothesis long after it’s was known with a high degree of confidence not to be valid? I fail to see how that would help or mitigate the damage already done. In fact, if I were a parent concerned about vaccines causing autism, I would find such an statement even more condescending and insulting to my intelligence than anything Paul Offit’s been accused of. But that’s just me, I suppose. Or maybe not. In any case, I would frame it more like, “This is great news. Now we can move on to other, more promising areas of research.”

Which is what scientists have been trying to do, but antivaccinationists won’t let them.

The other thing that I really detest in Sandman’s posts is his flagrant use of false equivalences:

The article you refer to is “Mistrust rises with autism rate,” by Anita Manning (USA Today, July 7, 2005). The article reviews the unending controversy between vaccination opponents, who charge that there has been a cover-up of evidence suggesting a link between thimerosal and autism, and vaccination proponents, who claim that the evidence of any such link is extremely weak and almost certainly false. (Thimerosal is a form of mercury that has been widely used to keep vaccines sterile.)

The latest wrinkle in this controversy came with an article by Robert F. Kennedy Jr. focusing on a June 2000 meeting of scientists and public health officials at the Simpsonwood Conference Center near Atlanta. Kennedy wrote that transcripts of the Simpsonwood meeting show the cover-up in action; vaccination supporters retort that the evidence discussed at the meeting was nowhere near persuasive enough to justify serious concern.

My key point is that both sides are right.

Jody Lanard and I first used the phrase “misleading toward the truth” to describe USDA risk communications about mad cow disease. People (including scientists) who believe they are right on the merits tend to withhold information that they fear might lead others to an erroneous conclusion. On vaccination safety, maybe proponents are 85% right. That’s not good enough for them, and too often they are less than candid about the other 15%. (Some of this is the sort of “conspiracy” the recent Kennedy/Simpsonwood kerfuffle is all about; some of it is more basic, like not bothering to mention that mercury is a poison in a disquisition the main thrust of which is the safety of thimerosal.) Leave aside whether or not this is dishonorable; it is demonstrably unwise. In a porous democracy like the U.S., the other 15% inevitably comes out – and the reluctance of proponents to acknowledge it makes it look much more compelling than it deserves.

Suppose the vaccination case isn’t 85% right but 99% right. All the more reason to be candid about that other one percent. The stronger your case actually is, the more foolish you are to try to make it look even stronger than it is.


Public health professionals claim that the weight of the scientific evidence shows that thimerosal does not cause autism. Critics claim that the public health profession is covering up the portion of the evidence that shows otherwise. Although I’m not qualified to assess either claim definitively, my strong impression is that both claims are true. (See for example my discussion of the June 2000 Simpsonwood Conference in the Guestbook entry linked above.)

That is, I think that the experts have solid grounds for concluding that thimerosal in pediatric vaccines is very unlikely to be responsible for the surge in autism diagnoses. And I think that once they reached that conclusion the experts have too often sought to reassure the public by overstating their degree of certainty, and have tried to ignore or discredit the evidence (a lot of anecdotal evidence plus a few studies) that suggested there might be something to the relationship after all. That’s what I mean by “misleading toward the truth.”

Overstating a mostly valid conclusion and hiding the small amount of contrary evidence is an incredibly common (and tempting) mistake. It is most common (and tempting) when people are upset, when you want to calm them down, and when all you have to work with is a pile of studies that didn’t find the effect they were looking for, plus a handful that might have found something. The evidence is maybe 85% on your side, but you’re afraid that acknowledging the other 15% might prolong the debate you’re trying to quell. So you suppress the 15%.

Sandman may know public relations, but he’s clueless about the science. Both sides are not, nor were they, “right.” The antivaccinationist side relied on conspiracymongering and quote-mining to create the legend of Simpsonwood as a place where the CDC plotted to hide the evidence that mercury in vaccines was the main cause of the “autism epidemic.” Indeed, it’s not 85% correct or even 99% correct to state that science has failed to find a link between thimerosal and autism. It’s more like 99.9999% right. Moreover, Sandman fails to present a single bit of evidence that, regarding thimerosal at least, the FDA, CDC, or vaccine proponents have been less than honest. He just asserts it as though it’s true. It’s also a total distortion that scientists don’t mention that mercury is a poison. They do, and I do. I have. So does Dr. Offit; indeed he says explicitly in the book that it is and even devotes a section of the chapter Mercury Falling to explaining why the dose makes the poison and why, at the doses used in vaccines, mercury is not toxic and that, contrary to the claims of antivaccine activists, even if mercury were toxic at the levels used the symptoms of mercury poisoning do not resemble the symptoms of autism. Dr. Offit hammers that point home over and over, and discusses why it is impossible to avoid mercury:

Because everyone drinks water, everyone has small amounts of methylmercury in their blood, urine, and hair. A typical breast-fed child will ingest almost 400 mcg of methylmercury during the first six months of life. That’s more than twice the amount of mercury than was ever contained in all the vaccines combined. And because the type of mercury in breast milk (methylmercury) is excreted from the body much more slowlly than that contained in vaccines (ethylmercury), breast milk mercury is more likely to accumulate. This doesn’t mean that breast milk is dangerous, or that infant formula is dangerous, or that water is dangerous. Not at all. It means only that anyone who lives on the planet will consume small amounts of mercury all the time. During legislative hearings to ban mercury-containing vaccines, some politicians have stood up and said: “I have zero tolerance for mercury.” This kind of statement makes for a great sound bite. But because mercury is an inescapable part of our environment, politicians with zero tolerance for it are going to have to move to another planet.

(Autism’s False Prophets, p. 114-115.)

I get the feeling that Sandman is attacking a gigantic straw man argument here, leading him to conclude:

The core problem for vaccination proponents, in short, isn’t that your critics exaggerate and distort. That’s true, but it’s not your core problem. Your core problem is that you also exaggerate and distort – and feel justified in doing so because you are (mostly) in the right, and don’t notice that it keeps backfiring on you. Or to put it a bit differently, your core problem isn’t that the public doesn’t trust you. That’s increasingly true too, but it’s not your core problem. Your problem is that you don’t trust the public.

And, worst of all:

My best guess is that vaccination proponents are “lying” (exaggerating and distorting) on behalf of the truth. That’s why I wish they’d stop.

That’s right. Sandman is accusing scientists like Paul Offit of lying in the name of defending vaccines. Of course, he can’t point to a single lie. Not one documented lie told by the CDC or other vaccine proponents to parents. Certainly he can’t point to one when it comes to thimerosal. In fact, the worst he can point to is a bit of a nonsequitur, namely a problem with the oral polio vaccine in Africa where, given the superstition and Islamic fundamentalists ready to jump at such problems, being reluctant to tell everything right away about a vaccine problem is somewhat understandable.

Sandman’s complaint is nothing more than a tu quoque fallacy writ large, and it’s a huge exaggeration. Antivaccinationists peddle outrageously false information and pseudoscience in a constant barrage of misinformation, exaggeration, cherry-picked data, and quote-mining on blogs like Age of Autism, while othe occasional vaccine proponent who may not be circumspect enough in not expressing too high a degree of certainty for Sandman’s judgment is to him equivalent to the tsunami of misinformation that comes from the antivaccine side. (Apparently, though, it’s OK for Sandman to “exaggerate and distort” while castigating scientists for “exaggerating and distorting” because he knows he’s in the right in his criticism.) In fact, it’s so wrong-headed that I have a hard time reading it without getting angry. Again, Sandman gives no specifics in the thimerosal controversy where vaccine proponents exaggerate and distort. He repeats the same essential message again in his later post, again bringing up the same straw men again and again not being able to give a single example, other than polio vaccine use in Nigeria, for which he castigated health officials for not acknowledging a very small risk of polio due to attenuated live virus polio vaccine. The only point he makes that I can sort of agree with is that if your position is 99% correct, at least acknowledging the other 1% is wise. However, I would also counter that there must come a certain level of certainty where acknowledging the “other side” gives pseudoscience an unjustified and undue appearance of validity. If you’re 99.999999% correct, are you obligated to acknowledge the other 0.0000001%, just to be “conciliatory”? Are evolutionists obligated to acknowledge creationist pseudoscience just to be “conciliatory”?

Then we come back to the exactly the same problem that Dr. Offit and I have described and why what Sandman says is a straw man: Scientists do couch their conclusions with “error bars,” so to speak. That’s a major problem in communcation, in fact. They do do exactly what Sandman decries them for supposedly not doing: Stating that, to the best of our knowledge, there is no association between thimerosal and autism and that multiple studies have failed to find a link but then qualifying it to point out that we can never completely prove a negative or that in a small number of susceptible individuals there might still be a possibility. That’s the very nature of how science works, especially in epidemiology: It can never completely prove a negative. Unfortunately, when scientists do acknowledge even the tiniest degree of uncertaint, antivaccine activists take that crack in the door and jump right on through: “See, see! There is a controversy! There might be a chance that vaccines do cause autism!”

I’m not saying that everything Sandman says is on that level of dubiousness. Certainly, I actually agree with him on this advice:

  • Attribute the change to the power of opposition groups. By far the easiest way to establish that a new precaution isn’t hypocritical is to concede that it’s a response to pressure. “If you’re so sure thimerosal is safe, why are you removing it?” “Because our critics won that fight!”
  • Explain the practicality of the decision. Whether it’s a genuine risk or not, a vaccine that significant numbers of people fear to take (or to let their children take) isn’t an effective vaccine. You’re not just deferring to your critics; you’re deferring to reality.

Exactly. Simply say: The antivaccinationists won that battle, and given how much fear they were provoking among parents banning thimerosal was the most practical way to blunt that fear.

Sandman is certainly correct in saying that science communicators would do well to be more empathetic, even giving some suggestions how to do it. However, his “sorry” gambit strikes me as far more condescending and paternalistic than telling it like it is and doing our best to make parents understand that we empathize with their difficulties. As we all realize, raising an autistic child is an incredibly difficult task, and I have grave doubts if I could ever manage it, for example. However, Sandman dresses up the simple message that given current science the chances that vaccines or mercury cause or contribute to autism but that we understand why at the individual level parents might believe they do in a whole load of false equivalences and advice that might make sense in the case of a scientific question whose anser is less clear-cut, but it’s advice that falls apart when it comes to defending science against pseudoscience.

ADDENDUM: I didn’t see this before, but it’s been pointed out to me that Peter Sandman has a rather infamous reputation among P.R. men, having done damage control for some of the largest corporations in the world. See:

  1. Sourcewatch: Peter Sandman
  2. Advice on Making Nice: Peter Sandman Plots to Make You a Winner
  3. Sandman’s Cagey Tactics
  4. Some Clients of Peter Sandman
  5. Mining PR Exec Lauds Peter Sandman
  6. Sorry is the hardest word for AWB