Updated: 15 February 2019 to add section 17
Update 19 July 2019: Note that Scott Gottlieb, who features in this post, left the FDA on 5 April 2019 and joined the board of Pfizer on 27 June 2019: Ex-FDA boss joins Pfizer 83 days after leaving US drug regulator, Financial Times.
I have been following FDA Commissioner Scott Gottlieb on Twitter and was alarmed at the threat he issued to vapers and the vaping industry in response to recent trends in US adolescent vaping.
If this is his analysis, @SGottliebFDA hasn’t tried hard enough to understand what is – and is not – a material risk in these youth vaping numbers. Yet he appears willing to put millions of adult lives at risk by imposing regulatory overkill on a life-saving product for smokers. https://t.co/HB64pgtjRJ— Clive Bates (@Clive_Bates) January 13, 2019
I disagree with the FDA’s analysis of what is happening with adolescent vaping and also what should be done about it. So I need to put some flesh on the tweet above and examine some of FDA’s claims in more depth. Please dip to these talking points… it’s a long blog but I hope at least some of it will be illuminating.
- Adolescent vaping is not an ‘epidemic’
- Vaping is not the same as nicotine addiction
- There is more to tobacco addiction than nicotine dependence
- Most adolescent vapers are not ‘hooked on vaping’ whatever anecdotes you might hear
- Most adolescent vapers are not regular or daily users (updated 19 July 2019)
- Frequent vapers are likely to also be smokers and this may be positive for public health (updated 19 July 2019)
- Not all vaping is with nicotine
- Marijuana use may be inflating the vaping numbers
- Nicotine is not “uniquely harmful” to the young developing brain
- Vaping is a low priority risk behaviour among adolescents
- E-liquid flavours may be beneficial to public health and adolescents
- The flawed and disturbing ethics of trading-off youth vaping and adult smoking
- It is not possible or useful to separate the welfare of adults and adolescents
- We do need a mature debate about nicotine, but FDA won’t have one
- FDA is the enemy of innovation
- “It’s the smoke, stupid” (and we are winning)
- A significant ‘gateway effect’ from vaping to smoking is not established and not likely (new)
Introduction – the youth vaping epidemic
New data suggest there was a steep rise in youth vaping between 2017 and 2018. FDA/CDC rushed out a partial disclosure of the 2018 National Youth Tobacco Survey (NYTS).
During 2017–2018, current [high school] e-cigarette use increased by 78% (from 11.7% to 20.8%, p<0.001). The proportion of current e-cigarette users who reported use on ≥20 of the past 30 days increased from 20.0% in 2017 to 27.7% in 2018Notes from the Field: Use of Electronic Cigarettes and Any Tobacco Product Among Middle and High School Students — United States, 2011–2018 [link]
I took a hard look at the partial disclosure of the NYTS survey data here: More heat than light – new US statistics on youth vaping provide no basis for FDA policy (November 2018). The basic issue is that you have to know more about: (1) who is a regular vaper; (2) what the regular vapers would be doing in the absence of vaping; (3) who is using nicotine. That data is not available at this time but we know enough to be sceptical.
However, these data have prompted a very strong response from FDA Commissioner, Scott Gottlieb. He has defined the rise in youth vaping as an ‘epidemic’ and says that this poses an ‘existential threat’ to the category.
I’ll tell you this. If the youth use continues to rise, and we see significant increases in use in 2019, on top of the dramatic rise in 2018, the entire category will face an existential threat.Scott Gottlieb, Remarks at Public Hearing on Eliminating Youth Use of Electronic Cigarette and Other Tobacco Product Use: The Role for Drug Therapies, 18 January 2019 [link]
The threat, it should be clear, is made by Dr Gottlieb in his capacity as a regulator and is based on his analysis of what the headline trends mean, and the options FDA has to respond. I will use this post to show why I do not believe his analysis is sufficient to justify ANY of the actions he is threatening.
In September 2018, FDA was jolted into action by the (at the time undisclosed) news of a rise in teen vaping. Since April, FDA had been captivated by an activist campaign and dozens of critical news reports focussed on the highly successful Juul vaping product. But the data was being treated uncritically. See my blog: The US media is losing its mind over vaping and Juul – the questions a credible journalist should ask. Alas, there has not been a more careful examination of the data or how it should be interpreted. If anything, the hype and panic has intensified.
We didn’t predict what I now believe is an epidemic of e-cigarette use among teenagers. Today we can see that this epidemic of addiction was emerging when we first announced our plan last summer. Hindsight, and the data now available to us, reveal these trends. And the impact is clearly apparent to the FDA.
Unfortunately, I now have good reason to believe that it’s reached nothing short of an epidemic proportion of growth.
I use the word epidemic with great care. E-cigs have become an almost ubiquitous ‒ and dangerous ‒ trend among teens.Scott Gottlieb, Commissioner FDA, 12 September 2018
Three months later, the US Surgeon General is no more measured:
U.S. Surgeon General Vice Adm. Jerome M. Adams issued an ‘advisory’ today stressing the importance of protecting children from a lifetime of nicotine addiction and associated health risks by immediately addressing the epidemic of youth e-cigarette use. E-cigarette use among youth has skyrocketed in the past year at a rate of epidemic proportions.US Surgeon General, advisory on E-cigarette epidemic among youth, 18 December 2018
How should we understand these claims from the top public health officials in the United States? Here are fifteen talking points…
1. Adolescent vaping is not an ‘epidemic’
The formal definition of an epidemic:
Epidemic refers to an increase, often sudden, in the number of cases of a disease above what is normally expected in that population in that area.Centers for Disease Control and Prevention (CDC). Lesson 1 Introduction to epidemiology [link]
Let’s be clear, the rise in vaping is not an epidemic. Vaping is not a disease and not even a cause of any disease. Vaping is a behaviour with relatively minor health consequences, if any
Perhaps they mean it as a metaphor – but technocratic entities such as FDA, the Office of the Surgeon General and CDC should not be trading in emotionally charged hyperbole while claiming to take great care with the choice of language. The appropriate response from FDA to the rise in teenage vaping would be to say:
There has been a steep rise in teenage vaping, and this is a real concern. We must understand the nature of the rise, its causes, and its interactions with other tobacco and substance use before we determine options to intervene. This is an area of regulation where a simplistic analysis and crude interventions can have unintended consequences and cause more harm than good to public health.
There are many other problems with the use of ‘epidemic’ to apply to the recent rise in teenage vaping and Carl V Phillips has written more than you need to know here: Calling vaping/tobacco use an “epidemic”: it’s even stupider than you might think
Dr Gottlieb reaches for “epidemic of addiction” as the basis of his epidemic claim, and says he uses it ‘with great care’. But it is a gigantic and inappropriate conflation to equate a rise in teenage vaping with a rise in ‘nicotine addiction’, as we shall see.
2. Vaping is not the same as nicotine addiction
Neither Dr Gottlieb or Dr Adams defines the term ‘nicotine addiction’. In real life, addiction terminology is highly contested. For example, some see addiction as a brain disease, while others reject the brain disease model as far too reductionist. But let’s look at the conventional definitions: Sussman and Sussman surveyed the literature and identified five characteristics of addiction:
(a) engagement in the behavior to achieve appetitive effects, (b) preoccupation with the behavior, (c) temporary satiation, (d) loss of control, and (e) suffering negative consequencesSussman S, Sussman AN, Considering the definition of addiction, Int J Environ Res Public Health. 2011 Oct; 8(10): 4025–4038. [link]
How many adolescent vapers have lost control and are suffering negative consequences? Or we could draw on definitions that the American health establishment uses. Here is the American Psychiatric Association:
Addiction is a complex condition, a brain disease that is manifested by compulsive substance use despite harmful consequence. People with addiction (severe substance use disorder) have an intense focus on using a certain substance(s), such as alcohol or drugs, to the point that it takes over their life.American Psychiatric Association, What is Addiction?
Wait, takes over their life? How many adolescent lives have been taken over by vaping? Or perhaps the definition from DSM-5 definition of “Tobacco Use Disorder”
A problematic pattern of tobacco use leading to clinically significant impairment or distressDiagnostic and Statistical Manual of Mental Disorders(DSM-5) Tobacco Use Disorder [link]
Note that Dr Gottlieb and Dr Adams use the more emotively powerful term ‘addiction’, rather than dependence. But the distinction between the two is important:
When people talk about addiction, they are usually referring to the harmful behavior associated with substance abuse. Dependence refers to the physical symptoms of withdrawal and tolerance.Addiction Center [link]
But even if we allow sloppy interchangeability of addiction and dependence, how many teen vapers meet the criteria for nicotine dependence?
Many experts stress the ‘time to first cigarette’ as an indicator of nicotine dependence.
Thus the time-to-first-cigarette item appears to tap a pattern of heavy, uninterrupted, and automatic smoking and may be a good single-item measure of nicotine dependence.
Baker T, Piper M, McCarthy D, Bolt D, Smith S, Kim SY, et al. Time to first cigarette in the morning as an index of ability to quit smoking: Implications for nicotine dependence. Nicotine Tob Res. NIH Public Access; 2007 Nov;9 (SUPPL. 4):555–70. [link]
The famous Fagerström Test for Nicotine Dependence includes time-to-first-cigarette in its index. However, this widely-used index scores zero for those having their first cigarette later than one hour from waking. This isn’t even daily use, but a subset of daily users who feel a strong compulsion to smoke soon after ending their overnight abstinence.
How many teenage vapers are having their first vape within an hour of waking?
3. There is more to tobacco addiction than nicotine dependence
Nicotine is the primary psychoactive substance in tobacco smoke or e-cigarette aerosol. But the dependence-forming characteristic of nicotine is strengthened by interactions with other substances present in cigarette smoke, for example, monoamine oxidase inhibitors (MAOIs).
Nicotine is the major neuroactive compound of tobacco, which has, by itself, weak reinforcing properties. It is known that levels of the enzymes monoamine oxidase A (MAO-A) and MAO-B are reduced in the platelets and brains of smokers and that substances, other than nicotine, present in tobacco smoke have MAO-inhibitory activities. Here, we report that inhibition of MAO dramatically and specifically increases the motivation to self-administer nicotine in rats.Guillem K, Vouillac C, Azar MR, Parsons LH, Koob GF, Cador M, et al. Monoamine Oxidase Inhibition Dramatically Increases the Motivation to Self-Administer Nicotine in Rats. J Neurosci. 2005 Sep 21;25(38):8593–600. [link] emphasis added.
There is no evidence that vape aerosol contains MAOIs, such as β-carboline alkaloids found in cigarette smoke.
These results suggest that β-carboline alkaloids from cigarette smoke acting as potent reversible inhibitors of MAO enzymes may contribute to the MAO-reduced activity produced by tobacco smoke in smokers.Herraiz T, Chaparro C. Human monoamine oxidase is inhibited by tobacco smoke: β-carboline alkaloids act as potent and reversible inhibitors. Biochem Biophys Res Commun. Academic Press; 2005 Jan 14;326(2):378–86. [link]
What then is the basis for treating all nicotine use as if it is equally reinforcing, dependence-forming and addictive?
4. Most adolescent vapers are not ‘hooked on vaping’ whatever the anecdotes say
Scott Gottlieb draws on anecdotal evidence to support his concerns about nicotine addiction, though he does not document any of it.
But, already, I’ve heard too many painful stories from parents of teenagers, pediatricians, and young people themselves. The stories make clear that, for many young e-cigarette users, addiction has already taken hold. Perhaps the kids didn’t realize that what they were inhaling contained nicotine, a highly addictive chemical. For these addicted young people, the reasons that they tried e-cigarettes may not matter anymore. These young people are hooked on vaping, and their worried parents, physicians, and the public health community are searching for tools to help them quit.Scott Gottlieb, Remarks at Public Hearing on Eliminating Youth Use of Electronic Cigarette and Other Tobacco Product Use: The Role for Drug Therapies, 18 January 2019 [link]
Obviously, it is wholly unscientific to approach the impact of vaping in this way. The use of second- or third-hand accounts with no knowledge of the cases or on what basis they are classified as addicted, or what proportion of users are in the ‘hooked’ category does not contribute meaningful information to the debate. But it does suggest that emotively charged terms are being used uncritically, and perhaps politically. Regrettably, the Surgeon General has also reported third-hand anecdotes as if they are useful contributions to the debate – much to my dismay.
There are plenty of reasons why senior public health officials would hear only the most extreme cases, and that those involved in relaying such cases would exaggerate or distort the cases – many of their interlocutors are engaged in a political campaign to use FDA regulation to apply a da facto ban to vaping. Selection bias and confirmation bias are everpresent dangers here.
As the discussion above suggests, ‘addiction’ is a slippery concept to define and senior officials are using it extensively without defining it. For it to be a meaningful term, it must mean something different to liking something a lot or feeling an urge to drink water when suffering water withdrawal (i.e. being thirsty). In the way the term is generally used to apply to tobacco use, it might have these characteristics:
- Loss of control
- Some sort of harm, impairment or distress to the user
- First-use each day within 60 minutes of waking
We don’t know how much adolescent vaping meets these criteria – I suspect only a tiny fraction if any. But neither does Scott Gottlieb nor the Surgeon General, Jerome Adams. Yet they are the ones making the claims and shaping public policy on the back of them. The available data come nowhere close to providing an adequate basis for these claims – and could even support the case that adolescent vaping is beneficial to public health.
But if we look at what we do know, the picture is far from nightmare portrayed in the press and by public health officials. Let’s try to inject the facts we do have into the discussion.
5. Most adolescent vapers are not regular or daily users
Updated 19 July 2019
E-cigarette frequency of use matters. There are at least two different vaping behaviours in adolescents and the difference really matters:
- Experimentation and frivolous use (most users)
- Dependence-forming substance use (some users)
Those using e-cigs only a few times a month are doing something much less concerning – just messing about, experimenting etc. We don’t know exactly how many because CDC/FDA has only published one data point for the 2018 frequency of use distribution: of those who vaped at least once in the last 30 days, only 27.7% did this on more than 20 days. This suggests the large majority (>72%) are not ‘substance users’ but have a more experimental or frivolous pattern of use – blowing clouds etc.
Given that 20.8% of high school students had vaped in the past 30 days, that means only 5.8% of high school students are regular or daily users (≥ 20 days in 30). This figure is confirmed (with rounding error) in the analysis below, which shows only 5.7% of high school-age adolescents are regular users and on 3.8% are daily users.
Of course, we don’t have any data on how many adolescents are vaping within one hour of waking, but it will likely be a small subset of the daily users.
Potentially problematic use is a small subset of the headline numbers. But as we shall see, the vast majority of the ‘potentially problematic’ use may actually be beneficial because it is concentrated in former or current smokers.
6. Frequent vapers are likely to also be smokers and this may be positive for public health
Update 19 July 2019
The original version of this post said we didn’t have the full breakdown of vaping frequency by tobacco use history. This is crucial to understanding the ‘anatomy’ of the so-called youth vaping epidemic. Now the data is available it confirms the picture we expected.
But the analysis does not stop there. What would this group of regular or daily users be doing instead of vaping if e-cigarettes did not exist? The data strongly suggest that many of these more regular users would be smoking.
If these frequent e-cigarette users are mostly already smokers or likely to become smokers, then this may be the start of them switching away from smoking – and therefore it could be beneficial for public health (and we have seen rapid declines in adolescent smoking as vaping has risen – see section 16). From previous studies, we know that frequent vaping is almost entirely concentrated in young people who smoke or use other tobacco products. Less than one in one thousand young people had become frequent vapers having never used tobacco before. And even in these cases, we do not know what this tiny minority would have done in the absence of e-cigarettes – possibly gone on to smoke.
Few never tobacco users had used e-cigarettes on 10 or more days in the past month (absolute percent < 0.1%).Villanti AC, Pearson JL, Glasser AM, Johnson AL, Collins LK, Niaura RS, et al. Frequency of youth e-cigarette and tobacco use patterns in the U.S.: Measurement precision is critical to inform public health. Nicotine Tob Res. December 2016 [link]
Scientists with access to the underlying datasets for the 2018 National Youth Tobacco Survey on which the ‘youth vaping epidemic’ claim is based have provided the breakdown of the frequency of vaping (number of days vaping in the past 30) by tobacco use history. Absolutely unsurprisingly, the most frequent vapers are also previously or currently smokers – the more they have smoked, the more they are likely to be more frequent vapers. This is the critical point: vaping is ‘intruding’ on the smoking behaviour of US adolescents.
Luckily, you don’t need to squint at the table above – you can access this data in this Google spreadsheet.
From this, we can derive some key results:
|High school students – Nation Youth Tobacco Survey, 2018|
|20.8%||Proportion of students who used an e-cigarette in the past 30 days|
|27.6%||Proportion of vapers who are regular vapers (≥20 days per month)|
|72.4%||Proportion of vapers who are not regular vapers (<20 days per month)|
|58.6%||Proportion of vapers who are occasional users (≤9 days per month)|
|80.3%||Proportion of vapers who had not previously used tobacco who are occasional users (≤9 days)|
|89.5%||Proportion of regular vapers who previously used tobacco|
|87.9%||Proportion of regular vapers who previously smoked|
|10.5%||Proportion of regular vapers who never used tobacco|
|12.1%||Proportion of regular vapers who never smoked|
|5.7%||Proportion of students who are regular vapers (≥20 days per month)|
|5.3%||Proportion of students who vaped and had never used other tobacco products|
|0.6%||Proportion of students who were regular vapers and never used tobacco products|
And some talking points:
- Much of the youth vaping epidemic is occasional use: almost half (48.7%) is on 5 days or less – approximately weekly – rising to 58.7% vaping on 9 days or less in the last 30. Among the 26% of vapers who had not previously used other tobacco products, occasional use rises to 71.2% vaping on 5 days or less and to 80.3% on 9 days or less. It’s hard to see how all this occasional or irregular use can amount to an epidemic of nicotine addiction.
- As I hope it is clear: the area where most concern should arise is those tobacco-naive users taking up vaping and becoming regular users – only 0.6% of high school students fall into this category. Even with this group, we cannot know whether some of these users would have gone on to smoke in the absence of vaping as an option.
- The other, more positive story is that regular vaping (or at least 20 days per month) is dominated by people who had previously smoked: 8 out of 9 (88%) of regular vapers had previously smoked. For these users, there is a positive story: vaping is in the mix of their tobacco use and could provide longer-term protection against smoking, which is the actual risky behaviour.
I am grateful to Professor Martin Jarvis OBE of UCL London for providing underlying data from NYTS datasets – see his table here. Any mistakes beyond that are mine!
I am not alone in examining this question. Professor Brad Rodu of the University of Louisville has produced two excellent posts on the subject.
Rodu draws similar conclusions to those set out above… (original emphasis)
In summary, the oft-cited teen vaping epidemic involves not three million youths, but rather 95,000 underage teens who vaped frequently but never used other tobacco products – or 0.6% of the nation’s 14.8 million high school students.
This post contains a very striking chart on dramatic changes in young adult smoking.
We need to keep in mind that smoking is the problem behaviour – at least twenty times as dangerous as vaping. The sharp decline in the red and pink on this chart are the big public health win. Instead, the anti-vaping activists are ignoring this – the actual public health win – and concentrating on the green. But as a youth or young adult risk behaviour, vaping is really of little concern.
7. Not all vaping is with nicotine
Most of the youth e-cigarette use is probably not nicotine-based… again the main National Youth Tobacco Use Survey doesn’t actually even ask if the vaping is nicotine-based, yet still counts it as “tobacco use”. But if we look at the University of Michigan Monitoring the Future survey we can get some idea – at least what the adolescents actually report. This letter in the NEJM with data from the University of Michigan Monitoring the Future survey gave the youth vapers reporting using nicotine in the past 30 days as 84% of the total (Table 1).
But this does not tell us the mix of vaping flavours with and without and nicotine through the past 30 days. The MTF survey previously asked about the last use of nicotine. This would be a better indicator of the mix of use within the 30 days. For 2015, the MTF reported that about two-thirds had not used nicotine the last time they vaped:
A common assumption among researchers and policymakers is that adolescents are vaping nicotine, although this assumption has yet to be examined closely.
Among students who had ever used a vaporiser, 65–66% last used ‘just flavouring’ in 12th, in 10th and in 8th grade, more than all other responses combined.
Miech R, Patrick ME, O’Malley PM, Johnston LD. What are kids vaping? Results from a national survey of US adolescents. Tob Control. BMJ Publishing Group Ltd; 2017 Jul 1;26(4):386–91. [link]
Correction: this section was updated 22 January. There has been a substantial increase in either those using nicotine or those realising they were using nicotine. But there is still vaping without nicotine incorrectly counted as tobacco or nicotine use in ‘epidemic’ claims.
8. Marijuana use may be inflating the vaping numbers
Despite the moral panic about vaping in the United States, the current use of marijuana was almost one in five 19.8% in 2017 among high school students. In 2018 e-cigarette use was 20.8%. It is quite possible that some of the increase in vaping arises from adolescents vaping cannabinoids instead of smoking. Though one would not wish to condone this, it is better to vape than to smoke – whatever the substance. However, such an effect would exaggerate the rise in vaping. It could simply represent a change in the way young people take marijuana.
CDC/FDA approach vaping as if it is a tobacco use behaviour, but that is an invalid assumption: some vaping involves no nicotine, some involves other substances.
9. Nicotine is not “uniquely harmful” to the developing brain
The Surgeon General has made a strong claim about nicotine and the developing brain.
I absolutely want people to understand that for young people, this presents a very unique danger. … Nicotine is uniquely harmful to the young, developing brain. We know that the brain continues to develop up to the age of 25. For the developing brain, [nicotine] can cause learning, attention and memory problems and it can prime the brain for future addiction.U.S. Surgeon General speaks at MD Anderson about the e-cigarette epidemic among American youth, 19 December 2019 [link]
A unique danger? Uniquely harmful? Really? More harmful than cannabis, alcohol, crystal meth, fentanyl? More harmful than American football? Worse than a car crash? Obviously, it is inappropriate to make wild claims like that nicotine is uniquely harmful.
But what of the more modest claim that young people’s brains are harmed by nicotine? Where would you look for evidence? How about all those young people who have been exposed to nicotine as smokers? There are millions of them happily pursuing adult life: but are they suffering any form of damage? In 1980, 30.5% of US 12th graders were past-30-day smokers, 21.3% daily smokers [Monitoring the Future]. This group would be in their 50s by now, and most would have quit smoking years ago. Where is the evidence of impairment in this population arising from youth nicotine use?
So an important rider on this claim is that: there are no human data that show nicotine causes material harms to the adolescent brain. The evidence that there exists is based on rat and mouse studies. Even the Surgeon General’s 2016 report recognises just how indirect and speculative the claims of harm are:
Limited direct human experimental data exist on the effects of nicotine exposure from e‑cigarettes on the developing adolescent brain, but experimental laboratory data have been found to be relevant in animal models to contextualize effects in humans.US Surgeon General. E-Cigarette Use Among Youth and Young Adults: A Report of the Surgeon General. Chapter 3: Health Effects of E-Cigarette Use Among U.S. Youth and Young Adults, December 2016 pp.104 [link]
If there was good epidemiology suggesting some sort of mental impairment arising from youth nicotine use in humans, then animal studies would be useful in confirming biologic mechanisms and broadening the evidence base. But the supposed problem is not visible in human populations of nicotine users and the animal studies are not convincing in isolation. David Abrams and colleagues reinforce this point:
The possible harms to youth of nicotine cited by Flannery are based primarily on animal studies. While concerning, direct causal evidence is sparse from human studies that can quantify nicotine’s harms or benefits. Our framework is agnostic to benefits or harms if they can be quantified beyond speculation and weighed against the well-known harms of nicotine when delivered in smoked tobacco with carbon monoxide and its toxins.Abrams DB, Glasser AM, Villanti AC, Pearson JL, Rose S, Niaura RS. Re: Disregarding the impact of nicotine on the developing brain when evaluating costs and benefits of noncombustible nicotine products. Prev Med (Baltim). 2019 Mar 1;120:158–9. [link][PDF]
The appropriate comparison is not between the adult and the adolescent brain, but the effects of different things that young people do have on the developing brain and whether these effects are harmful in some way.
I am unaware of evidence suggesting that nicotine “can cause learning, attention and memory problems” in young people. Please advise.
In fact, Abrams et al. point to research suggesting cognitive benefits for adults from nicotine use:
Our framework allows for additional harms informed by science and recognizes potential benefits of nicotine use in adults (e.g., increased alertness, concentration, memory, and modulating mood) (Heishman et al., 2010; Newhouse, 2018; Talati et al., 2016).Source: as above [PDF]
10. Vaping should be a low-priority risk behaviour among adolescents
The Youth Risk Behavior Surveillance system provides some insight into the risks that young people take.
So 29.8% of high school students are using alcohol, and 13.5% “binge drinking” (according to CDC’s definition) but 20.8% vape. There are about as many marijuana users as vapers. What is the response to that?
If there are any material risks with vaping, they would arise from decades of use, and only if the vaping becomes a smoking habit that would not otherwise have happened. Alcohol use, in contrast, can end a life instantly, pulverised in a road traffic accident, or put young people into violent, abusive or vulnerable situations that can be acutely harmful.
Also, what do we know of the effects of alcohol on the adolescent brain? Here’s one review:
These differences in structure and function are related to poorer neurocognitive performance on tests of attention, working memory, spatial functioning, verbal and visual memory, and executive functioning. Adolescents who engage in heavy episodic drinking (binge drinking), as well as those who have greater withdrawal and hangover symptoms, tend to show the greatest deviations in functioning. Family history of alcohol is related to cognitive disadvantages that may leave youth more vulnerable to the neurotoxic effects of alcohol use.Squeglia LM, Jacobus J, Tapert SF. The effect of alcohol use on human adolescent brain structures and systems. Handb Clin Neurol. NIH Public Access; 2014;125:501–10. [link]
But where are the calls to limit adult access to alcohol? In fact, many adults in responsible positions seem quite sanguine about alcohol
The right way to address young people’s problems is to look at the individual in the round – substance use or other risk behaviours are often an outcome of stresses in their personal, family, social and community circumstances. The reason FDA is focussed on nicotine use is threefold: (1) jurisdiction: it is limited by its legal authorities to work on tobacco – it is a hammer and nicotine is its nail.; (2) money, from 2019 it will be receiving $712 million per year from tobacco companies via user fees demanded under the Tobacco Control Act S.919; (3) politics, it faces pressure and scrutiny. It has strong incentives to adopt eye-catching, politically salient campaigns that show it is putting money to good use. It has to define the problem as a characteristic of the product (flavours, packaging, marketing) rather than a byproduct of the individual’s situation
11. E-liquid flavours may be beneficial to public health and adolescents
Dr. Gottlieb is quick to attribute the youth vaping epidemic to ‘flavors’. And quick to determine that taking them off the market would roll back the epidemic. Far too quick.
I find myself debating with tobacco makers and retailers the merits of selling fruity flavors in ways that remain easily accessible to kids. But if the epidemic continues to mount, I’m sure that the debate will change to one of whether these products should continue to be marketed at all without authorized pre-market tobacco applications.Scott Gottlieb, Remarks at Public Hearing on Eliminating Youth Use of Electronic Cigarette and Other Tobacco Product Use: The Role for Drug Therapies, 18 January 2019 [link]
In doing so he misses much subtlety about how flavours and product appeal actually work in public health terms – for both adults and adolescents. Many of us have tried to explain it, but apparently without success. If it can be shown that certain flavours change behaviour, it is quite possible that change will be beneficial – an attraction away from smoking.
To summarize, the chain of reasoning required to justify rule-making to prohibit particular flavors, flavor categories or flavor descriptors in non-combustible products is extremely challenging, with the real possibility that FDA intervention could cause harm both to adults and young people if it makes misjudgments about:
(1) the effects of vaping on health, and (2) the effect of flavors on vaping. FDA would need to show that vaping itself is a source of net harm (this is unlikely) and show that particular flavors or descriptors were increasing uptake and contributing to harm (this is difficult). Finally, it would need to show its proposed intervention would be proportionate and effective, and not prone to excessive unintended consequences (for this there is no credible evidence). The FDA does not have a reliable case at any point in this chain of reasoning.Attorney General Tom Miller, Iowa. Response to Flavors ANPRM [link]
The absurdity of this approach reached a peak in November 2018, when the FDA decided to remove flavoured e-cigarettes from convenience stores and allow them only to be sold in age-restricted locations.
The changes I seek would protect kids by having all flavored ENDS products (other than tobacco, mint and menthol flavors or non-flavored products) sold in age-restricted, in-person locations and, if sold online, under heightened practices for age verification.Statement from FDA Commissioner Scott Gottlieb, M.D., on proposed new steps to protect youth by preventing access to flavored tobacco products and banning menthol in cigarettes, 15 November 2018 [link]
So cigarettes were unaffected, and the only vapes allowed were those most like cigarettes. FDA has completely lost track of what it is doing and why.
AG Miller also highlighted the obvious absurdity of leaving only those products which most closely resemble cigarettes on the market.
Generations of young people have taken up cigarette smoking usingAttorney General Tom Miller, Iowa. To Scott Gottlieb. Youth tobacco and nicotine use – proportionate and responsible reaction, 14 November 2018. [link]
almost exclusively tobacco and menthol flavors. There is no obvious logic to restricting e-liquid flavors to only those that mimic the most dangerous tobacco products and have been the main basis for initiation in the past – and no justification of such a move has been provided.
12. The flawed and disturbing ethics of trading-off youth vaping and adult smoking
Scott Gottlieb has asserted that it may be necessary to make it harder for adults to quit smoking in order to make it harder for adolescents to start vaping. To do this he draws of the ‘population health standard’ that underpins FDA’s role
The legal standard for FDA premarket review of a new tobacco product includes consideration of whether the product would be appropriate for the protection of the public health. We must factor in whether these products get kids addicted to nicotine. It’s that simple. And we are taking aggressive actions today to address this.
This may create some obstacles for some adults who also enjoy e-cigs. These are the hard tradeoffs that we’re grappling with. But the youth risk is paramount.
It’s now clear to me, that in closing the on-ramp to kids, we’re going to have to narrow the off-ramp for adults who want to migrate off combustible tobacco and onto e-cigs.Statement from FDA Commissioner Scott Gottlieb, M.D., on new steps to address epidemic of youth e-cigarette use, 12 September 2018 [link]
The problem with this reasoning is that it is not mediated by the respective risks and benefits – smoking is very harmful, nicotine use without smoking is not. The law that Dr Gottlieb refers to does require consideration of the impact on the whole population. The following formulation is used at multiple points in the Tobacco Control Act
…appropriate for the protection of the public health shall be determined with respect to the risks and benefits to the population as a whole, including users and nonusers of the tobacco productFor example, in Section 910 of the Federal Food, Drug, and Cosmetic Act – Application for Review of Certain Tobacco Products, [link]
But nowhere in the Act is a false equivalence created between, for example, dying in agony from cancer or the debilitating misery of COPD and adolescent vaping, which is, objectively, a low-risk behaviour with minimal health consequences. Quitting smoking by switching to vaping is a huge benefit to a smoker. Starting to vape causes negligible harm to an adolescent – and even those vaping regularly may be doing it as an alternative to smoking, in which case it is a benefit.
On top of this, Dr Gottlieb is bizarrely indifferent to the effect that vaping might have on youth smoking. He is not prepared to countenance the idea that e-cigarettes may reduce youth smoking, even if this was proven. He rules this out on principle.
No child should use any tobacco product. We’ve seen cigarette use decline among kids, while e-cig use has grown sharply. This is happening even as overall rates of tobacco use among kids has declined, according to recent data.
This is still not acceptable, even if the trends are moving in a more positive direction of reduced overall use of tobacco products. Even if kids are using ENDS instead of cigarettes — and that migration in part accounts for the decline in youth cigarette use — that’s still not an acceptable trade.
Parents who see their children using e-cigs and say, “well at least my child isn’t smoking,” should take no comfort.
No child should be using any tobacco product.Scott Gottlieb, FDA’s Nicotine and Tobacco Regulation and the Key Role of Regulatory Science, 18 June 2018 [link]
Actually, parents should take a great deal of comfort – smoking is actually the harmful behaviour. They should equally be concerned if Dr Gottlieb’s misguided ethics stop them vaping instead of smoking and so causes them harm. Public health does not work by declaring that “No child should do …X” and then expecting that to happen. It is about intervening in real-world behaviours, where appropriate, to improve public health.
13. It is not possible or useful to separate the welfare of adults and adolescents
With his “off-ramp, on-ramp analogies, Dr Gottlieb tries to separate the welfare interests of adolescents and adults. This is short-sighted and counterproductive. In a letter to Dr Gottlieb, Attorney General Miller and others argued that, in practice, it is not possible to divide these populations and their respective interests so neatly for tobacco policy purposes – giving three reasons:
Youth tobacco and nicotine use – proportionate and responsible reaction, 14 November 2018 [link]
- Parental smoking and adult role-modelling are important risk-factors and predictors for youth smoking initiation. By modelling a different, far less harmful behavior, adult vaping and associated smoking cessation is likely to have a beneficial effect on youth smoking initiation and prevalence.
- The loss of a parent or close relative to smoking-related disease is a significant detriment to most young people. Likewise, adult ill-health imposes costs on the family in terms of lost economic activity and increased caring responsibilities. Harm reduction for adults has collateral benefits for a whole family, including its younger members.
- Adolescents grow into adults, and today’s youth have an interest, not necessarily acknowledged, in having better options for their future.The serious harms of tobacco or nicotine use are not instantaneous and mainly develop over many decades of use. They are an outcome of the pattern of tobacco use over the life-course – almost all of the premature mortality risk of smoking is avoided by stopping smoking by age 35. So, opportunities to stop smoking in the first two decades of adult life are especially valuable and continue through the whole life-course.
14. We do need a mature debate about nicotine, but FDA won’t have one
FDA officials like Mitch Zeller have been calling for a debate about nicotine. For example, Zeller posed six questions.
How comfortable are we with long-term, or possibly permanent, use of less harmful nicotine delivery mechanisms by adults, if they help keep currently addicted smokers from relapsing to combustible tobacco products?
How much weight should be placed on diminished interest in quitting nicotine altogether?
Given the potential health impacts of dual use of tobacco, how acceptable is a short period of dual use while transitioning to less harmful nicotine-containing products?
What if many current smokers engage in dual use on a long-term or permanent basis?
Can we revise labeling and indications for medicinal nicotine to increase quitting?
How might youth initiation be affected by the availability of different nicotine-containing products and how should we account for youth uptake of these products?Zeller M.The Future of Nicotine Regulation: Key Questions and Challenges. Nicotine & Tobacco Research. 2018 Oct 12. [link]
Good questions – and I provided my answers: Rethinking nicotine: FDA asks six questions about the future of nicotine regulation. So did Brad Rodu, but with some scepticism that anyone cared to hear the answers: The Answers to FDA’s Tobacco Questions Are Obvious, But Perhaps Unwanted.
My view in brief: we are a drug-using society, and nicotine – delivered without smoke – is a relatively benign drug. Efforts to stop people using nicotine run counter to efforts to reduce their exposure to smoke. The government should not be intervening to obstruct much safer alternatives to smoking while allowing cigarettes to be ubiquitously available.
So when is this debate about rethinking nicotine actually going to happen? Or is it just a diversionary tactic to look serious without being serious? At the moment we have FDA and the Surgeon General taking to social media with slogans that appear to have been copied from the playbook of activist organisations. That is not the debate we need now.
15. FDA is the enemy of innovation
FDA, CDC and the Surgeon General have lost sight of the most important fact in tobacco policy. To paraphrase Bill Clinton: “it’s the smoke,stupid“. Yet all have become obsessed with the ‘threat’ from much safer alternatives to smoking, such as vaping. FDA is hardly doing anything to address the 2.3 million teens still using combustible tobacco products (CDC, 2017).
- Vaping products will be forced to go through a hugely burdensome pre-market authorisation process before 2022 [PMTA draft guidance] – and FDA is now threatening to bring that date forward (here). Most products and companies will exit the market as a result. In contrast, most cigarettes on the market before 22 March 2011 were ‘grandfathered’ and thousands of cigarette brands and variants have access to the US recreational nicotine market without the need for any pre-market review. [link]
- FDA has a “Youth Tobacco Prevention Plan” but it is entirely focussed on e-cigarettes [statement] and not on the actually harmful tobacco products, cigarettes.
- FDA is insisting that e-cigarette products are sold under more restrictive conditions than cigarettes: all flavoured ENDS products (other than tobacco, mint and menthol flavours sold in age-restricted, in-person locations [announcement, October 2018]. In fairness, FDA is proposing to ban menthol in cigarettes – but flavours are integral to the e-cigarette value proposition.
- FDA made extremely demanding disclosure requests of vaping businesses. For example, to Juul and others. No equivalent curiosity was shown about the cigarette businesses and preventing youth smoking.
- FDA issued letters to five manufacturers asking each company to “submit within 60 days plans describing how they will address the widespread youth access and use of their products”. No equivalent demands have been made of cigarette businesses and youth smoking.
- FDA has focussed its Real Cost campaign on vaping. For example, at the time of writing, its Facebook page and YouTube channel are dominated by anti-scientific fear-based advertising about vaping. Smoking, the actual public health issue, is barely mentioned.
- The Office of the Surgeon General has released a 2016 report and a new advisory raising alarm about e-cigarettes. But the SG has done nothing to put these risks into context with smoking-related risks.
- FDA proposed a rule that would take most smokeless tobacco products off the market by setting a nitrosamine standard that is impossible to meet for most products. The standard was not applied to the tobacco in cigarettes.
- FDA, CDC and the Surgeon General have done almost nothing to correct wildly inaccurate public perceptions of the relative risk of combustible and non-combustible tobacco products. Only 2.3% of Americans think (correctly) that e-cigarettes are much less harmful than cigarettes [link]; Only 13.4% believe smokeless tobacco is less harmful than cigarettes (it is much less harmful)[link]; over 50% believe nicotine causes most of the cancers caused by smoking and only 8% correctly strongly disagree that nicotine is the main cause [link]. In fact, the focus on vaping and demonising nicotine has been pushing the vaping and nicotine perceptions in the wrong direction. Yes, they have been making a bad situation worse.
- FDA has yet to issue a single Modified Risk Order, even though it is beyond dispute that smokeless tobacco and vaping products are much less harmful than cigarettes. The effect is to protect cigarettes from competition on the basis of health differential – a key USP of non-combustible products. See MRTP application stats. The cost of even applying to state the obvious runs to millions of dollars and given FDA’s intransigence, the cost is a clear deterrent to even trying.
This the FDA’s response to the most promising innovation in several decades to address the ‘epidemic’ of smoking-related disease is to define it as a threat, overstate or misinterpret its risks and ignore its potential – smothering it with bureaucracy, opaque procedures and hostile propaganda. They are truly the enemies of innovation.
16. “It’s the smoke, stupid” – and we are winning
If the public health task is to change behaviours that cause the greatest harm, that would be smoking, then we really are winning.
There is more to this than the appearance on the graph. Analysts examined the trends in youth and young adults smoking and concluded:
There was a substantial increase in youth vaping prevalence beginning in about 2014. Time trend analyses showed that the decline in past 30-day smoking prevalence accelerated by two to four times after 2014. Indicators of more established smoking rates, including the proportion of daily smokers among past 30-day smokers, also decreased more rapidly as vaping became more prevalent.Levy DT, Warner KE, Cummings KM, Hammond D, Kuo C, Fong GT, et al. Examining the relationship of vaping to smoking initiation among US youth and young adults: a reality check. Tob Control. BMJ Publishing Group Ltd; 2018 Nov 20. [link]
Note that this is not the trade-off acknowledged by Dr Gottlieb between the ‘on-ramp’ to vaping for youth and ‘off-ramp’ from smoking for adults. This is vaping displacing smoking among youth. The obvious corollary is that measures to limit youth vaping uptake could have adverse unintended consequences, and cause a rise in youth smoking. At the very minimum, FDA should recognise this possibility and incorporate it into any assessment of the public health impact.
The adult trend (above) is extremely encouraging. Adult smoking cessation provides the most rapid health gains and healthcare system benefits. In the age of vaping, the US has got back on track to meet its Healthy People target to reduce smoking to 12% by 2020. There may be many causes of the improving trend, but it is hard to argue that vaping is creating a public health disaster.
It is clear that vaping products are having a profound impact on the market for recreational nicotine, currently dominated by smoking. The tobacco majors have seen their stock value savaged as investors project forward the impacts of profound shifts in consumer preferences for nicotine – the migration from harmful combustion products to much less harmful vapour or smokeless products. On 24 January, Altria hit a new 52-week low and is down 37% in the past year. BAT is now down by over 53% in that time. Even PMI is down by about a third. Something big is happening!
The question is: does the investor in the cigarette business have a friend or foe in the FDA, Surgeon General and activist groups like Campaign for Tobacco-Free Kids? From where I sit, the champions of public health look like they are striving hard to hold back the transformational innovations that will alter the tobacco and nicotine market permanently and for common good. Big Tobacco’s Little Helpers, as my friend David Sweanor memorably put it.
17. A significant ‘gateway effect’ from vaping to smoking is not established and not likely
The gateway effect, whereby a less risky behaviour causes the initiation of a more risky behaviour, has a disreputable history with origins in the ‘reefer madness’ moral panic about cannabis. Its primary function is as an activist talking point that contrives to load relatively benign behaviours or drugs with the risks and harms of more harmful behaviours or drugs: thus cannabis becomes as harmful as heroin and, currently, vaping becomes as harmful as smoking.
I mention this because Dr. Gottlieb has been suggesting that vaping may cause smoking.
Though much is made of this study, the authors themselves insert the all-important disclaimer:
… the PATH Study data are observational; we cannot establish causal relations or rule out the possibility of residual confounding by underlying risk-taking propensities.Berry KM, Fetterman JL, Benjamin EJ, Bhatnagar A, Barrington-Trimis JL, Leventhal AM, et al. Association of Electronic Cigarette Use With Subsequent Initiation of Tobacco Cigarettes in US Youths. JAMA Netw Open. 2019 Feb 1;2(2):e187794. [link]
This is at the heart of the intractable problem with gateway claims. It is impossible to eliminate “underlying risk-taking propensities” as the reason for an observed association between vaping and subsequent smoking. This means people might have an underlying tendency to indulge in risky behaviours and both the low-risk and the high-risk behaviour might be attributable to this. That is quite likely given that vaping is a ‘smoking-like’ behaviour – though at a dramatically lower risk. This is known as ‘confounding’. But what it means is that without the existence of vaping these young people may have gone on to smoke anyway, and therefore that vaping is not the cause of their smoking.
Although it is possible to identify associations between prior vaping and subsequent smoking, they may not be causal. But a further issue is just how few adolescents these relationships apply to. Professor Mike Siegel of Boston University School of Public Health recently published an interesting reanalysis of ‘gateway effect’ data from a major study:
The investigators were unable to report a single youth out of the 12,000 in the sample who was a cigarette naive, regular vaper at baseline who progressed to become a smoker at follow-up. Why? Because the number of these youth was so small that it was impossible to accurately quantify this number.Siegel, M. Landmark Study Fails to Find Any Evidence that Vaping is a Gateway to Smoking among Youth, The rest of the story (blog), 18 February 2019 [link]
Based on my own calculations from the data presented, out of the 12,000 youth, there were only 21 who were even past 30-day e-cigarette users at baseline (cigarette naive) who progressed even to ever smoking at follow-up. These youth reported using e-cigarettes only 1-5 days in the past 30 days at baseline, so it is not even clear that they were regular vapers. They literally could have tried an e-cigarette once 30 days earlier.
It is possible that vaping was not a gateway to smoking for any of these 21 youth, but even if it was, they represent just 0.2% of the youth population.
By the way, he says that even for the 21 that started vaping and then moved on to cigarettes, vaping may not have been a gateway. This is because in a hypothetical world (‘the counterfactual’) they may have become smokers anyway.
Didn’t the National Academies say there was a gateway effect?
The National Academies (NASEM) report, The Public Health Consequences of E-cigarettes, appeared to give gateway-effect enthusiasts a lift with its conclusion (16.1) that:
There is substantial evidence that e-cigarette use increases risk of ever using combustible tobacco cigarettes among youth and young adults.National Academies of Sciences, Engineering, and Medicine. 2018. Public health consequences of e-cigarettes. Washington, DC: The National Academies Press. [link].
There are many problems with this:
- “Ever-use” is not in itself a problem behaviour – it more an indication of a tendency to experimentation. The problem behaviour is sustained regular use. While ever-use is a necessary step to sustained regular use, it is far from sufficient – most experimenters will not become ‘addicts’. Use of ever-use draws in a huge population ‘froth’ of teens who are not at risk and are just messing about.
- The NASEM committee simply asserts that it has sufficiently eliminated confounding to draw a causal conclusion. But that is far from the case. However, the covariates used in each study evaluated by NASEM are limited and crude approximations of the actual drivers of underlying risk-taking propensities. The actual drivers may be related to subtle phenomena like mental health status, self-esteem, family and social circumstances etc – yet it is impossible to approximate these fully with the proxy measures that are captured in surveys and used for deconfounding. For example, even if a study adjusts for ‘depression’, it can hardly do justice to the complex landscape of depression with a yes/no variable. I defer to Carl V Phillips for his excellent science lesson on confounding on why it is impossible to eliminate confounding in the way claimed by some researchers. For full academic treatment, see Phillips C V. Gateway Effects: Why the Cited Evidence Does Not Support Their Existence for Low-Risk Tobacco Products (and What Evidence Would). Int J Environ Res Public Health; 2015 May 21;12(5):5439–64. [link]
- The challenge of deconfounding is further exacerbated by the focus on ‘ever use’ This because ever-use is a different behaviour to sustained and has different underlying drivers to regular use. It would require different deconfounding variables to eliminate common risk factors – variables that reflect the propensity to experiment or goof about, rather than variable that reflect the propensity to use tobacco more intensively – a form of substance use.
- NASEM recognises gateways in both directions – a catalytic effect in which vaping causes smoking and a diversion effect in which vaping prevents smoking. But its conclusion rests only on looking at transitions from vaping to smoking: “Removal of baseline ever smokers from the analyses is critical for eliminating the possibility of reverse causation“. By this, NASEM means that some e-cigarette use might be caused by smoking (for example, people trying to quit smoking) – it doesn’t eliminate reverse causation because that is a real-world effect. It simply ignores it in the analysis. Any assessment of the impact of e-cigarettes on young people needs to make a net assessment of all the possible transitions (harmful and beneficial), not just look at one pathway in isolation (note this is nothing to do with adults – it is about whether young people may be benefiting).
- Claiming a gateway effect is meaningless without quantification. It is a near-certainty that at least one e-cigarette user will have become a smoker because of vaping. But how many would it take before it was of public health relevance?
- For further views: National Academies report – just what the FDA ordered and Does vaping lead to smoking? by Carl V. Phillips.
In its 2015 Evidence Review, Public Health England found no compelling evidence of a catalytic gateway effect and that the concept was too poorly defined and difficult to assess to be useful:
No clear provenance exists and its origin appears to derive from lay, academic and political models. It is apparent that discussions about the natural progression of drug use observed in longitudinal studies of young people appear to have morphed into implicit conclusions on causality without any evidential backing.
The gateway theory is ill-defined and we suggest its use be abandoned until it is clear how it can be tested in this fieldMcNeill A, Brose LS, Calder R, Hitchman SC, Hajek P, McRobbie H. E-cigarettes: An Evidence Update. A Report Commissioned by Public Health England. London; 2015. [link]
Please use comments to suggest improvements or further talking points and I will add them to this list.