Arielle Selya is a former academic turned consultant. She set out to prove that the “gateway hypothesis” – that vaping leads to smoking in young people – was real, but ended up disproving it. When many of her peers refused to listen to her results, she left for the private sector, and has been a critic of academic politics and groupthink ever since.
So set the scene for us: who are you and how did you end up where you are?
I got my PhD in neuroscience, and then I actually tried to leave academia because of the structural issues we’ll talk about today. But I ended up getting stuck for another ten years, and that’s where I got started in tobacco behavioral research.
When my postdoc started, I was researching adolescent smoking behavior, nicotine dependence, risk factors, time course of the addiction, all of that stuff. I had initially bought into the gateway hypothesis and that e-cigarettes are dangerous. But when I analysed the data, I realized that that’s not really what’s going on. So I tried to publish what I thought to be an objective reading of the data, and it got completely ignored.
Back when you were getting started, what was it that made you assume that the gateway hypothesis was real?
Honestly, academic groupthink. I had not heard any other opinion out there and it was the only narrative I was hearing from the published research the published research and the media. And, frankly, it seemed plausible to me.
But I was seeing assumptions being made in the gateway hypothesis that I didn’t see empirical support for: they were only speculation at that point.
In particular, there was this idea that if a young person is attracted to ecigarettes and starts using them, they develop nicotine dependence, and the nicotine dependence promotes smoking or causes them to go on to smoke. So there’s this implicit or assumed mechanism through nicotine dependence.
I had access to a great longitudinal study, and I thought, well, why don’t I look at that? So in my head, I thought, okay. I’m gonna get an interesting publication that proves that there’s this mechanism through nicotine dependence for the gateway hypothesis, and it’ll be a really useful paper. But I didn’t find any evidence for it!
I submitted this to Addiction (an academic journal) and I was lucky enough to get Clive Bates as a reviewer on that paper, and he started talking about the exit gateway or the off ramp idea in his review. That was really eye opening. So after that paper was published, I started thinking of more ways that I can look at the gateway question.
Then I realized population level trends are a great source. Smoking was already declining after 1999, so if we project that forward and see what would we have expected smoking rates to look like now had vaping not been invented, and compare it to what actually happened, smoking is lower than you’d have expected it to be based on those preexisting trends. Put simply: if vaping is a gateway to smoking, then where are all the new smokers?
That’s a really interesting way of putting it! Can you drill down into that statement a little more?
Well, if ecigarette use is increasing in the population, then you would expect more smokers than had they not been invented. But that’s where it gets complicated because that involves making a counterfactual comparison: we can’t create a parallel universe to test the theory!
You have to kind of get at that using time trend modelling, and even then, just the two time series being correlated doesn’t always mean there’s causation. In fact, it can lead to a lot of spurious associations.
It’s a tough case to prove mathematically. But we have an overall decline in the use of any tobacco or nicotine product, and this has gone on long enough that more of the mainstream research community seems to be starting to face the fact that there isn’t this huge surge in youth smoking. Some of them have shifted their focus to the young adult age group: the argument there is, “well, these are the kids that started during the peak of youth vaping, and let’s see what they’re doing as young adults”. And young adults do have a higher ecigarette use prevalence, but they also have the largest drops in cigarette smoking prevalence and there’s evidence that many of them are using ecigarettes to quit smoking.
So the most likely hypothesis right now is that there was a fad among young people a few years ago. The fad is declining, but we’re not seeing smoking rates rise back to similar levels where they were by cannibalizing what was the youth vaping market? Is other nicotine use making up, or are we seeing an overall decline in nicotine use?
That that’s a good question. I’ve also analyzed snus data in Norway going back a decade or two, and it’s the same kind of dynamics there. As snus became more popular, cigarette smoking declined. And that it’s the same sort of product substitution story where I think more or less there’s an approximate percentage of the population that finds a benefit in using nicotine, and these shifts are when they shift from one product to another.
So as I interpret the date, in Norway twenty years ago, the transition was away from combustible tobacco to snus. In the US, it’s to ecigarettes. And now there’s some evidence of oral nicotine pouches substituting for e-cigarette use.
You seem pretty clear that, empirically, this phenomenon just doesn’t exist, and yet we continue to see people proclaim with certainty that it does. What’s driving them?
I think a lot of it has to do with the culture in academia. Academics are sometimes discouraged from talking to people outside of academia, and there was no real dissent within it. So it’s easy to live in a bubble and not question received wisdom.
The other part is NIH funding (note: NIH is the US National Institute for Health, which funds most academic research in the US related to tobacco and nicotine). NIH funding is highly valued because of their high rate of indirects, meaning funding over and above the grant amount designed to cover an institution’s overhead. Because of that, I was told by an academic administrator that other types of grants don’t matter for your career progression. And the result that this has on faculty members is it forces them to conform with the NIH’s funding priorities.
So what NIH says goes?
Yes – here is an extent to which successfully obtaining NIH funding is a function of taking the scientific viewpoint of NIH and turning it into data.
When I was a naive junior faculty member and I stumbled across this gateway versus diversion issue, I did propose a project looking at what we might expect to see if the gateway hypothesis was real. I got a comment back criticizing me for suggesting that we shouldn’t recommend ecigarettes to kids just because they’re safer.
And, of course, I wasn’t recommending that kids vape, I was only proposing to undertake an observational kind of epidemiological study. But they feared that my results might lead to that.
But what about other sources of funding for this kind of work?
Well, I was applying to other major funders too. Charitable foundations such as Robert Wood Johnson. And I was encouraged to go after American Lung Association, Bloomberg Foundation and American Cancer Society grants. But they all have the same hostile view as the NIH.
I’m not sure where that opposition comes from. I don’t interact enough with those entities to say why they all came down on the side of anti-THR, but it’s unfortunate because there’s no big funders who challenge the accepted orthodoxy.
Global Action used to do it, but they’re tainted by the association with Phillip Morris. Other than that, there’s no-one outside of industry.
So in the end, you thought the only way to get the work done was to move to industry? Did your former colleagues have a view on that?
So things could have gone very differently. I almost gave up with the vaping story because I clearly wasn’t getting anywhere with grants. I was interested in it from a research perspective and a publishing perspective, but I was at the point where if I tried to go the traditional academic route, I would have been a failure. So I was trying to take a different turn and go somewhere completely different.
There’s a lot of hostility towards any ties with industry, and I knew that that would immediately and irrevocably undermine my credibility. But in the end, I decided I can probably do more in the real world working for industry than I could in academia.
I haven’t gotten a lot of open hostility, it’s more a problem of being ignored. So even before there was an easy reason to dismiss my work, like, in other words, before I was taking industry money, I was ignored by my colleagues. I was publishing on this gateway diversion issue, and I thought it was really pertinent and relevant. And I would reach out to other academics trying to collaborate, and that never went anywhere. I wasn’t really getting productive engagement. So actually, in some ways, very little has changed.
We can’t have an American on without talking about politics. We have seen huge NIH cuts come down from the White House. Do you think that will have an impact on the NIH’s ability to set the academic narrative as you’ve described? And if it does, who’s gonna step in to set the narrative?
Without going into politics too much, it’s almost annoying now that it’s become a politicized and polarized issue. It would have been better if the NIH had never had indirect rates so high. That that would have made a more level playing field in terms of incentives to go after grants. Given that they’re now high, I support lowering them again in principle, but I think it has to be done in the right way. I don’t think a sudden cut is really helpful for anyone.
That leads into the question of, well, who’s gonna step in to fill the gap? And the other academic funders: Bloomberg, American Lung Association, American Cancer Society, all have the same viewpoint of e cigarettes. So unless another big funder steps in, it’s not gonna solve the state of the science.
Arielle is an employee of Pinney Associates, Inc, which consults Juul Labs on nicotine vapour products to advance tobacco harm reduction. She also individually provides consulting services on behavioral science to the Center of Excellence for the Acceleration of Harm Reduction (CoEHAR), which received funding from the Global Action to End Smoking (GA). These funders had no role in this interview.
