Daily Vaper

How Low Will They Go? Academics And Journals Spread Disinformation

Photo via Shutterstock

Carl V. Phillips Contributor
Font Size:

Pushing disinformation is standard fare for the propagandists in the tobacco control movement. More disturbingly, that disinformation is also produced by people who are supposed to be university scholars and journals that are supposed to be scientific. This once again raises the question: How low will they go?

The latest of many examples of the disturbing behavior of academics betraying the truth is a peer-reviewed article in the journal “Preventing Chronic Disease” by Sarah D. Kowitt and colleagues at the University of North Carolina. Their asserted results are not very interesting. They looked at survey data and, unsurprisingly, found that if high school students (specifically, non-vapers who expressed no interest in smoking) believed that vaping was more harmful, either to the vaper or to bystanders, they were less “susceptible” to taking up vaping. (A proper reporting of what they really measured appears below, though it is absent from the paper’s abstract.)

This is not exactly shocking. Fewer people who are overly worried about health risks from vaping are likely to want to try it. Moreover, anyone actually considering vaping is likely to (accurately) convince herself it is low-risk, either because she researches the topic or just because of natural psychological tendencies. Thus there is undoubtedly some causation in both directions, and there is undoubtedly confounding that makes the association even stronger.

What is shocking — in the sense of being disturbing, rather than surprising — is this assertion in the authors’ conclusion: “To prevent initiation of e-cigarette use…educational campaigns could target harm perceptions associated with e-cigarettes” (emphasis added).  Since their results show that greater perceived harm is associated with less interest in vaping, what they are really saying is “convince kids that vaping is more harmful than they think.” This is not education. This is lying. The only way to send a message that vaping is more harmful than people already believe would be to claim that it is much more harmful than it really is.

When asked about the perceived risk of vaping, the “susceptible” students’ average response was a bit lower than “somewhat harmful” on a four-point scale of harmless/somewhat/very/extremely harmful. The “not susceptible” group averaged between “somewhat” and “very.” It is difficult to make much sense of this because of the terrible categories used: Someone who understands the truth faces the choice of asserting that vaping is harmless (which is, strictly speaking, not true for any exposure) or not putting vaping in the lowest category of risk on a four point scale (if risks are divided into four categories in any natural way, vaping clearly belongs in the lowest one). In addition, the researchers made an error that biased the average downward, by coding anyone who said “not sure” into the “harmless” category.

But interpreting the responses as the authors did, apparently unaware of the huge flaws in their methods, the study suggests that making someone substantially less “susceptible” means moving their perception of risk up to “very” or “extremely.” Thus, these university researchers, endorsed by the editors and reviewers of a scientific journal, are clearly advocating that “education” be used to move average beliefs from vaping being “somewhat harmful” — which is not a precise statement but feels like an overstatement of the risk — toward clearly inaccurate perceptions.

Imagine the outcry if credentialed history scholars published journal articles calling for teaching disinformation about slavery, perhaps because the truth makes high school students unhappy and discourages blind patriotism. There are actors who call for exactly that, and indeed implement it to varying degrees. But scholars, journals and respectable historians are not complicit. Tobacco control has damaged academia and scholarly publishing to the point that we see such unthinkable behavior.

Or consider the outcry from this same “public health” political faction whenever someone advocates disinformation about condoms and HPV vaccines. Teenagers are at least somewhat more likely to have sex if it they know it is safer. There is a political faction that does not like that, and so advocates misleading messages about the risks. Experts push back, whatever their personal views about teenage sexual activity, simply because the claims are false. There would be the equivalent of riots in the virtual streets if academic journal articles advocated teaching disinformation about safer sex.

A journal article that advocates for scientific disinformation would still be appalling if the research suggested this would be effective. In this case the weak research design meant that it did not actually do so.

The sample was restricted to subjects who were “not susceptible” to smoking. What this really means is students who asserted they would “definitely not” smoke in the next year, no matter what. This is an absurd assertion on its face. Asserting this, rather than “probably not” (the next strongest answer), is just thoughtlessly expressing political correctness. A high school student should be sufficiently aware of life’s complexity to realize that it is impossible to say “definitely” about almost anything. By looking at only those making this assertion, the researchers were removing the more insightful students from the sample.

The definition of “not susceptible” to vaping was also defined based on an absurd assertion of “definitely not.” The vast majority of those in the sample, those who “definitely” would not smoke, unsurprisingly made the same unrealistic assertion about vaping. The authors do not report the results, but presumably almost all of the others said “probably not.” It would be odd for a nonsmoker to be interested enough in vaping to actively expect to start, but to not have already done so. Thus all the results hinge on whether someone said “definitely not” or “probably not,” which are no different for all practical purposes.

Moreover, if we ignore all these flaws and interpret the data as the researchers did, the associations are remarkably weak. With causation in both directions and confounding, our prior expectations would be for substantial association between perceived harm and inclination to vape. Yet the results suggest that a huge change in perception — from vaping being “very harmful” rather than “somewhat harmful” — reduces “susceptibility” by only about 25 percent. Thus, if the results were to be believed, they would suggest risk perceptions matter less than we previously thought.

Kowitt and her coauthors are not merely calling for promoting scientific illiteracy and disinformation. Their contempt for honesty is so great that they call for this despite their research suggesting that badly misinforming someone merely modestly increases her chance of saying she will “definitely not” vape, as opposed to “probably not.”

Follow Dr. Phillips on Twitter