New CDC and FDA Study Presents More Lies and Misinformation on Youth E-Cig Use

By Lindsay Fox Posted October 22, 2014


CDC and FDA Study Presents More Lies and Misinformation on Youth E-Cig Use


Apparently not satisfied with their first attempt at misleading the public about the extent of e-cigarette use among young people (nor with stoking the fires of the absurd e-liquid poisonings fears), the CDC has published a new study
(working alongside researchers from the FDA’s Center for Tobacco Products) furthering it’s abundantly obvious anti-e-cigarette agenda. This time, they looked into the use of e-cigarettes and the possible association with the “intention” to smoke, and over the course of the paper they manage to twist and distort reality sufficiently for lead author Rebecca Bunnell to appear justified in saying the following in the press release.


“The increasing number of young people who use e-cigarettes should be a concern for parents and the public health community, especially since youth e-cigarette users were nearly twice as likely to have intentions to smoke conventional cigarettes compared with youth who had never tried e-cigarettes.”


If you’re aware of the CDC’s previous attempt to further the “won’t somebody please think of the children!” cause, you’ll be well aware of the sort of shining examples of willful idiocy that need to be contained in the full paper in order to “justify” this statement. They apparently don’t want you to know, though, so they hid it the paper behind a paywall. Thankfully, Carl V. Phillips was annoyed enough by this decision (since tax-payer funded research is not usually allowed to be paywalled) to put the full paper up on Anti-THR Lies.


I can understand why they don’t want people to read it, though, because they CDC-FDA team seems to be convinced that “probably not” means “yes,” demonstrating their level of expertise well before you even consider the abundant methodological errors and statistical flaws in the paper. Worst of all, this pure junk science somehow survived the peer review process and made it into the journal Nicotine & Tobacco Research, giving a false air of credibility to a study that any layman reader could identify as transparent deception.




  • Using data from the National Youth Tobacco Surveys from 2011 to 2013, the researchers aimed to determine the relationship between e-cigarette or other tobacco product use and intention to smoke.


  • Respondents had to answer “definitely not” to both of two questions related to future tobacco use to class as having no intention to smoke. If they said “definitely not” to one and “probably not” to the other, they were classed as intending to smoke.


  • Out of the sample of 43,873 never-smoking middle and high school students, just 0.9 percent had ever tried an e-cigarette, and only 0.3 percent had done so at least once in the last month.


  • 21.9 percent of never-smokers who’d never tried an e-cigarette were classed as “intending” to smoke, compared to 43.9 percent of never-smokers who’d ever used an e-cigarette.


  • Because of the classification system used for “intentions” to smoke, this finding really means that never-smokers who weren’t absolutely certain they wouldn’t smoke in the next year were more likely to have tried an e-cigarette at least once. Or: students interested in smoking are more likely to try vaping.


  • According to the study, exposure to more sources of cigarette advertising was also related to “intentions” to smoke, with more exposure making it less likely students would respond that they definitely wouldn’t smoke.


  • Multi-variable analysis suggested that never-smokers who’d used an e-cigarette were 1.7 times more likely to have intentions to smoke, but an inappropriate and misleading statistic was used, not to mention an obvious confounding variable being ignored.


  • The discussion section of the paper suggests harsher restrictions on e-cigarette advertising and enhanced efforts to reduce youth use, despite the study providing no evidence e-cigarette use among middle and high schoolers is any cause for concern.


What They Did – Investigating Vaping and “Intentions” to Smoke


The researchers used data from the National Youth Tobacco Surveys conducted in 2011, 2012 and 2013 on students in grades 6 to 12. This is designed to be a nationally-representative sample, and as you would expect, asks many questions related to tobacco use to a large amount of eligible students. They used this data to determine whether never-smokers who’d tried e-cigarettes were more likely to have intentions to smoke than those who had never vaped.


The core issue with the previous CDC youth e-cig use study was that the definition of “current” user was absurd. According to the CDC, if you’ve vaped on one day in the past 30 days, you’re a “current” user, even if that’s the only time you’ve ever done so and you only had one or two puffs. This mistake persists in this research, and the same misleading criteria are also applied to other combustible (such as pipes, hookah and cigars) and non-combustible (like chewing tobacco, snus and dissolvables) tobacco products. Never smokers were defined as those who had never had even one or two puffs on a cigarette.


In addition to e-cigarettes, the authors also investigated the impact of other tobacco product use on the middle and high school students’ intentions to smoke, as well as their level of exposure to “pro-tobacco” advertisements and other characteristics (including sex, race, and whether they lived with someone who uses tobacco).


The questions designed to measure intention to smoke are where the problems really get started for this paper. The “intention” to smoke was assessed using a pair of questions: “Do you think you will smoke a cigarette in the next year?” and “If one of your best friends were to offer you a cigarette, would you smoke it?” There were four possible answers to these questions: definitely not, probably not, probably yes and definitely yes. The correct course of action is simple, either “yes” answer would count as an intention and either “not” answer would count as a lack of intention.


The researchers apparently saw this a little differently (but saw no cause to explain why at any point in the paper), arguing that the only way you could be classified as not having intentions to smoke was if you answered definitely not to both questions! In other words, if you said you definitely wouldn’t smoke a cigarette in the next year, but when asked what you’d do if a close friend asked you to take a few puffs softened a little to “probably not,” the CDC and FDA class this as you having “intentions” to smoke.


As Carl Phillips comments, saying you definitely won’t smoke is quite an extreme statement, and any realistic teenager (who doesn’t just repeat “just say no” propaganda) would at most respond with “probably not.” Even if you don’t think you will, there’s always a possibility of something happening in the future. Their idiotic stance to classification is typified by the fact that those respondents who only answered one of the two questions (just 0.45 percent of the sample, but still) were counted as having “intentions” to smoke, simply because there wasn’t a “definitely not” answer to both. If the only piece of information they have is you saying you definitely won’t smoke a cigarette if offered to you by one of your best friends, they still class you as intending to smoke.


The bulk of the analysis was centered on kids who had never smoked (43,873 in total), and as well as the more straightforward assessments, they conducted multivariate analyses on the findings, to attempt to account for things like gender, race, tobacco advertisement exposure and the presence of another smoker in the household. In order to increase the size of the sample of never-smoking vapers, they pooled the three years’ worth of surveys together for the analysis.


What They Found – Non-Smokers Who Try Vaping are More Likely to “Intend” to Smoke




If you’re concerned about the overall issue of non-smoking youths using e-cigarettes, and the feasibility of concerns about a “gateway” to tobacco, all you need to do is look at the most basic results of this study (or any other done into this issue). Of all of the young people in the survey, only 6.1 percent had ever used an e-cigarette, with rates of 20.2 percent among ever (at least one puff at some point) smokers and just 0.9 percent of never smokers. Using the closest measure to current use in this survey (vaping at least once in the last 30 days), 6.9 percent of ever smokers were classed as current vapers compared to just 0.3 percent of never cigarette smokers. That’s less than one in 300 never smokers, and even that is using a ridiculous definition of what constitutes a current user, which will almost certainly include recent experimenters rather than actual regular users. This simply serves as yet another piece of evidence that non-smokers barely ever use e-cigarettes.


The numbers of never-smoking participants who’d used other forms of combustible tobacco or other non-combustible products were higher, but still ultimately very low. 7.7 percent of never-cigarette-smokers had ever used another combustible form of tobacco, and 2.6 percent had smoked a non-cigarette product at least once in the last 30 days. For non-combustible tobaccos, 3 percent of never-smokers had ever used them and 1 percent had used one in the last 30 days. It’s interesting that e-cigarettes remain the focus of the fears about youth addiction, despite the fact that almost nine times more never-cigarette-smoking youths had smoked a pipe, cigar, hookah or other combustible product in the 30 days prior to the survey.


The headline finding comes from the results to the highly flawed “intention to smoke” results. According to their highly questionable definitions, 21.5 percent of never smokers who’d never vaped had an intention to smoke, in comparison to 43.9 percent of the never smoker’s who’d tried vaping. This is the justification for the claim that never smokers who use e-cigarettes were twice as likely to intend to smoke cigarettes in future as those who had never vaped.


I think we can respond to this together with an emphatic “duh!” Oh really, CDC and FDA, teens who try something that looks like smoking are more interested in smoking than those who don’t? Stop the presses – the people must know about this!


Of course, this study doesn’t provide enough information to say that trying e-cigarettes causes a doubling in never-smokers’ desire to smoke, in fact, it’s much more plausible that having an existing greater interest in smoking than other never-smokers would make somebody more likely to try an e-cigarette. But since they didn’t even bother with an accurate measure of intentions to smoke, it’s more like: never-smoking middle and high school students who weren’t completely and unilaterally opposed to trying smoking in the next year were more likely to have tried a puff on an e-cigarette at least once. Clearly, the correct response is “so what?”


Carl Phillips further points out that those who’d have access to an e-cigarette would logically be offered to try it by a friend, and since almost all vapers are smokers or ex-smokers, people with access to e-cigarettes would almost definitely have smoker friends. Somebody who makes friends with smokers is obviously less likely to respond that they definitely wouldn’t smoke than somebody who isn’t friends with smokers.


Since this all comes down to the classification of responses to the “intention to smoke” questions, researcher Brad Rodu has compiled a more realistic version of these findings based on the 2012 National Youth Tobacco Survey (the 2013 data hasn’t been released yet). If you’re wacky enough to think “definitely” or “probably not” should mean “no” and “definitely” or “probably yes” should mean “yes,” and that different answers to the two questions (a “yes” to one and a “no” to the other) constitutes mixed intentions, then the findings look very different.


It seems that from the whole eligible sample from the 2012 survey (150 never-smokers who’d ever vaped and 17,672 who hadn’t), only 9 never smokers who’d ever used an e-cigarette (just 8 percent of all never-smokers who’d vaped) intended to smoke a cigarette, with 13 (11 percent) having mixed intentions. Using the CDC and FDAs classification, the very same data set would suggest that 59 percent of these never-smokers who’d vaped at least once “intend” to smoke. How you classify intentions is clearly very crucial to what you find.


The other findings as reported by the paper seem to support the idea that some never-smokers are less likely to say they will definitely not smoke than others, and these people are more likely to have tried an alternative nicotine-based product. Although the figures for the intention to smoke are downright misleading for the reasons outlined above, it’s interesting that the corresponding figures for never-smokers who’d ever used other combustible or non-combustible products are very similar to those found for the e-cig experimenters. For combustibles, 38.3 percent show “intention” and for non-combustibles it’s 41.4 percent, compared to the 43.9 percent for e-cigarettes.


Based on this (and the obvious fact that every smoker was once a never-smoker), it seems like there are just some non-smokers who are slightly interested in smoking, and these individuals are more likely to have tried any non-cigarette tobacco product, especially ones with greatly reduced health risks. Whatever the potential cause of this is, it represents a confounding variable; a crucial factor that’s completely unaccounted for (and unacknowledged) in the study. E-cigarettes strike the perfect balance between pleasantness (unlike pipes or cigars), comparative safety and similarity to smoking (unlike snus or chewing tobacco), so with this hypothesis in mind it makes sense that never-smokers with an interest in smoking would be more likely to give them a try than anything else. Viewed like this, it’s a safer outlet for ordinary curiosity, rather than a menace luring them into a greater desire to smoke.


Finally, they looked at the impact of advertising on smoking intentions. Exposure to “pro-tobacco” advertising did appear to correlate with the idiotic interpretation of the responses on the “intention to smoke” measure, with those exposed to content from more sources (online, in magazines, on TV and so on) being more likely to have “intentions” to smoke. Of those with no exposure, 13 percent were classed as intending to smoke, compared to 20.4 percent of those exposed to one or two sources and 25.6 percent of those exposed to three or four sources. Although the real figures (with “intentions” better defined) would undoubtedly be smaller, it seems that exposure to advertising does have some limited impact on whether or not middle and high schoolers say they won’t smoke in the next year.


The Multivariate Analysis and Statistical Issues


Carl Phillips also criticizes the type of figures reported for the analyses that took several variables into account. In short, odds ratios aren’t appropriate when you’re working with proportions, which he illustrates with an example. If 50 percent of group A intends to smoke and 25 percent of the reference group does, it’s clear that twice as many of group A intend to smoke. Using odds ratios, the individual 1:1 and 1:3 ratios for the groups lead to a final odds ratio for group A of 3.0, which makes it appear like three times the chance rather than just twice the chance, as we know it to be.


He adds that, “It is so reassuring that the people we depend on to protect us from Ebola cannot figure out first-year statistics.”


So while the study reports from the multivariate analysis that never-smokers who’d used e-cigarettes were 1.7 times more likely to have intentions to smoke, Phillips roughly calculates that the real figure would be under 1.5. Even this is a huge over-estimation because the multiple variables accounted for didn’t include the confounding one the study completely ignores: there’s some other reason that a certain subset of middle school and high school students are more likely to want to try e-cigarettes or alternative tobacco products and less likely to say they definitely won’t smoke in the next year.


The Real Intended Message: E-Cigarettes Are Bad!


With the pesky facts swept under the rug or twisted to suit their end, the CDC and FDA researchers go on to editorialize and suggest restrictions on e-cigarette marketing and efforts to prevent all forms of tobacco use (including e-cigarettes) among youth. In particular, they say the evidence from this study on advertising and smoking intentions is especially relevant because of the rise in e-cigarette advertisements, and also point out that three times as many never-smoking students had ever used e-cigarettes in 2013 than in 2011. By 2013, more than a quarter of a million non-smoking middle or high schoolers had ever tried an e-cigarette. This sounds much more impressive than 0.9 percent (or for use in the last 30 days, 0.3 percent), and that’s why this raw number was in the headline to the press release – big numbers sound like a lot even if they are really a tiny fraction of the total. They also say these people “used” e-cigarettes, which most people would associate with regular use rather than having had at least one puff, ever.


The discussion section of a study is supposed to look at the limitations of the findings. This would have been a good place to address the whole “probably not” = “yes” issue, but apparently they see this as needing no justification. They acknowledge the potential issues due to the use of two questions in combination, but somehow completely ignore this very obvious problem.


The whole point of the paper was the criticize e-cigs, with the “science” taking a backseat to their ideologies, political or otherwise. This is obvious because the science is flawed beyond all belief and is clearly designed to continue their fabricated story of how e-cigarettes are luring children into cigarette smoking. This is illustrated by their acknowledgement that the study can’t establish causality or whether the intention to smoke came before vaping or vice-versa, which is then followed up by the claim that either way, e-cigarettes are bad.


They argue that if intention to smoke comes before vaping, then it’s bad because they’re exposed to nicotine (although the study didn’t actually ask if the e-cigs the never-smokers tried contained nicotine), which increases the odds of combustible product use (which is a wholly unsupported assertion). If it’s the other way around, then e-cigs magically cause the intention to smoke and are even worse. In other words, they take a valid critique and then twist it to mean that e-cigs are either bad or they’re even worse. They don’t suggest the alternative: that e-cigs provide a safer outlet for interest in smoking and regular vaping is highly unlikely in never-smokers, and even less likely to lead to smoking, even though they acknowledge that intention to smoke has been on the decrease while e-cig use has been increasing.


Despite the actual evidence, the take-away message they’re trying to implant is that e-cigs are going to lead to the deaths of millions of children, lured into lifelong smoking thanks to the fabricated “gateway” effect.


Conclusion – The CDC and FDA are Trying to Mislead Us All


This study is devoid of competent scientific analysis, and is quite obviously intended to smear e-cigarettes despite the abundant evidence of their harm reduction potential and continuous findings that non-smokers, adult or kids, aren’t using them much at all. It’s important to remember that the CDC and FDA are supposed to be concerned with the health of the nation, and health decisions must be based on sound science. It’s disgusting enough that this study made it through peer review, but the really horrendous thing is that these people charged with such an important task are either scientifically illiterate, or more likely, purposefully misleading everybody. They want to destroy something with the potential to save millions of lives, and there’s only one reason they would do that: they put financial or ideological interests ahead of improving public health.