draft Regulatory Science White Paper #3: “Addictiveness” of nicotine and e-cigarettes

by Carl V Phillips

See yesterday’s post for more about the white paper series and our requests for comments.

This one calls for a bit of explanation.  This topic is difficult to nail down and say much that is concrete about (for reasons that are explained) and would not have been high on our list for the series.  However, when we met with OIRA they were particularly interested in this point, so we decided to do what we could.  Keep the two implications of that in mind if you would like to comment: 1) we know it is difficult to get this right, but decided to do it anyway, and 2) we are particularly interested in comments if you have any idea how anyone could do any better than this.

Consumer Advocates for Smoke-free Alternatives

DRAFT Regulatory Science #3:  “Addictiveness” of nicotine and e-cigarettes

draft – not yet peer reviewed

Soundbite version

While it is widely claimed that nicotine in general, and e-cigarettes in particular are addictive, there is actually no evidence that nicotine is, by itself, addictive (setting aside that there is no accepted definition of that word making the claim vague, and just following the common usage of the word).  It turns out that claims about the addictiveness of nicotine are based on studies of smoking.  When nicotine is isolated, there is widespread insistence that it is not addictive (the common insistence about NRT) and substantial evidence that even people who were completely captivated by smoking have a relatively easy time giving it up.  Whatever “addiction” to smoking means, it is clear that users of nicotine have a very different experiencee.

Background

It is widely claimed that nicotine consumption causes the effects on behavior and preferences that are sometimes labeled “addiction.”  However, such claims are based on the effects of cigarette smoking, the dominant modern method of nicotine consumption.  This means that the data is perfectly confounded: it is impossible to distinguish the effects of nicotine from other effects of smoking.  While there is little doubt that nicotine is the strongest attraction of smoking, further examination suggests that there is little reason to believe that nicotine delivered in other ways has the same captivating effects as smoking.  Indeed, there is no affirmative evidence to support such a hypothesis, and quite a bit of evidence that suggests it is not true.

The widespread failure to differentiate between consumption of cigarette smoke and of nicotine appears to be largely a historical accident, stemming from failure to adequately test the nicotine hypothesis before accepting it.  The conflation has recently been embraced by anti-tobacco extremists as a convenient tactic for attacking tobacco harm reduction in general and e-cigarettes in particular.  It is more difficult to explain why the conflation became so embraced by tobacco control activists during the period when the only nicotine products (pharmaceutical nicotine products, a.k.a., nicotine replacement therapy, NRT) were actively touted by them, but that was how the politics played out.  Pointing out that the evidence did not support the popular claim largely fell to researchers who were not part of the tobacco control industry (e.g., Frenk and Dar), though recently some who conflated smoking and nicotine in the past have corrected that error (e.g., Fagerstrom).

Note that the word “addictiveness” is used in the title here because it is a common touchstone for the general concept.  That word, along with its other grammatical forms, has no defensible or even widely-accepted scientific or medical definition, and indeed are used in ways that make it nearly impossible to define and that create confusion.  When the word is used, it refers to some haphazard combination of dependency (having negative withdrawal effects), behavior patterns, appreciation of the product, other impacts on preferences, and biomarkers.  Sometimes it those using the word mean nothing more than someone uses, or even merely likes, a vilified product.  Thus, when speaking for ourselves in this document we avoid forms of that word and substitute other words (we have chosen “captivate”) that refer to the feelings or behaviors that the word evoke.  Those words are also not scientifically defined, of course, but their vagueness is clear and thus they are unlikely to be misinterpreted as if they are precise scientific claims.  When a form of the word “addiction” are used, it refers to specific or general claims made by others.

Some commentators have tried to salvage the concept of nicotine being addictive by suggesting that variations in user reactions to products are simple different degrees of addiction, with the nicotine still being the cause of addiction.  While this scale might be a valid approach for a more concrete concept like dependence, it clearly defies the general interpretation of the word “addiction,” which implies some kind of consuming and overwhelming motivation.  A scale of addictiveness that goes down to zero is nothing more than a scale of preference — how much someone likes a particular experience.  To call something addictive (or even “a little bit addictive”) because someone tried it and liked it rather than hating it, and thus might be inclined to try it again sometimes, is clearly disingenuous and intended to mislead people with a normal interpretation of the word.  So while there is no metric available and thus obviously no bright line, it is clear that “addiction” implies some substantial captivation and thus “a little bit addictive” should mean not addictive.  This is a necessary clarification for attempting to address the question at hand, since without it everything that people fits the definition of addictive and the question becomes nonsense.

Empirical evidence

Establishing a negative epistemic claim — that there is not evidence that nicotine delivered other than via smoking has the same captivating properties as smoking — is necessarily difficult.  The best support for the lcaim is the observation that when a references is given for the claim that nicotine is addictive, it is invariably to a study of smoking.  There simply appears to be no evidence that extracted nicotine has the behavioral or preference effects that are often called addiction.

NRT products are generally claimed to be non-addictive or close to it, even by those who aggressively condemn smoking for being addictive (example).  In the few studies of use of these products among non-tobacco-users (as a treatment for diseases where nicotine is beneficial), there were no reports of subjects having a difficult time quitting.  While the designers and manufacturers of NRT products attribute the lack of addictiveness to specific design details, a plausible alternative explanation is that nicotine alone never has the same behavioral effects as smoking.  There is no recorded history of NRT products that were abandoned because they were judged to be addictive, as we would expect if the design details really mattered much.  For some NRT products (nicotine gum, in particular), use defies the ostensible intentions of the designers because they can be used ad lib to generate nicotine pharmacokinetics that are quite similar to smokeless tobacco or e-cigarettes, products that are often declared to be addictive.  Yet there is little concern expressed about the gum being addictive.  While the oral NRT products have unpleasant taste and feel which makes them unappealing, the user can deliver as much nicotine as they want, as with e-cigarettes.  So if these NRT are non-addictive, it is difficult to see how nicotine or e-cigarettes could be addictive.

E-cigarettes have offered the first widely-studied large-scale experience of nicotine use without the other exposures from smoking or smokeless tobacco use.  (There has long been fairly widespread long-term consumption of NRT products, using them similarly to other tobacco products and not as temporary cessation aids.  But because this is not acknowledged by manufacturers and advocates of the products, there has been a glaring failure to study this population.)  Evidence from the e-cigarette experience strongly suggests that nicotine delivered via e-cigarettes this form is not so captivating as smoking, even among those who found that e-cigarette use was the only way they could manage to quit smoking.

Among the thousands of individual testimonials about successful smoking cessation using e-cigarettes, there are many reports of users easily or accidentally quitting the e-cigarettes after establishing their switch to them, or dropping the nicotine concentration they are using down to zero.  Many other users report that they feel less captivated by their e-cigarette use compared to smoking.  Many significantly reduce their nicotine content but do not eliminate it, which might suggest a dependency but not necessarily the other characteristics that are usually implied by “addiction.”  Those who keep vaping zero-nicotine may be captivated by the physical act of smoking/vaping, but clearly were not captivated by the nicotine itself.

Several surveys of e-cigarette users further support these observations, including those by CASAA (results not yet published), Farsilinos et al., Dawkins et al., and Heavner et al.  Quitting e-cigarettes and reducing and eliminating the nicotine occur even though experienced users have access to inexpensive products and realize that e-cigarettes are very low risk and thus they have little incentive to quit (let alone to use zero-nicotine e-cigarettes).  This suggests that not only are many former smokers not captivated by the nicotine, but become relatively indifferent about it in the absence of smoking.

None of this data allows an estimate of the prevalence of these tendencies.  This is obviously true of testimonials, but also all surveys to date have used convenience samples — mostly of experienced e-cigarette enthusiasts — with unknown sampling properties.  Still, it does demonstrate that for some established tobacco users, e-cigarette use is less captivating than smoking.  There is no evidence that suggests anyone has the opposite reaction.

There have been no reports claiming someone who was not previously a smoker becoming addicted to nicotine via e-cigarettes.  The newness of the products means that perhaps there will be such cases though none have yet been reported.  However, given that there are a substantial number of well-resourced anti-tobacco activists who insist that addiction to nicotine occurs almost instantaneously and are intent on showing that e-cigarettes are evil, their failure to find and report even a single such case suggests that it is not happening.  We can be sure that if even a single personal testimonial supported the claim, the story headlined “I thought e-cigarettes were harmless, but then I got addicted” would be irresistible to the media.  The story’s absence strongly supports the epistemic negative that no such testimonial exists.

The evidence about smokeless tobacco is more difficult to interpret.  While there can be no quantitative metric for a concept that is not defined in the first place, attempts to compare the “addictiveness” of smokeless tobacco to smoking have suggested that they are similar.  This occurs even in the USA, where the population was largely tricked into believing that the risks from smokeless tobacco were similar to those from smoking, and thus believed they had a compelling reason to quit.  Most research on this topic is suspect because it comes from sources with an enormous conflict of interest — they oppose the use of smokeless tobacco and thus are motivated to overstate its harms — but cannot be dismissed entirely.  Better evidence comparing the captivation effects of smokeless tobacco compared to smoking or e-cigarettes could distinguish among the hypotheses presented below

Use of traditional cigars (or “large cigars” as referred to in marketing and regulation) does not generally produce the same behaviors and feelings as cigarette smoking.  Most cigar users use them only occasionally.  A similar pattern is observed for hookah smokers (at least in Western cultures), where it is an occasional indulgence and not a captivating habit.  However, research on these products is limited, so it is difficult to draw any firm conclusions.

Scientific analysis

There are several plausible explanations for the observations.  All of these hypothesized explanations suggest that nicotine per se is not particularly captivating.  No single explanation is going to be universal, given population heterogeneity:  Some former smokers who switch to e-cigarettes are inclined to then quit e-cigarettes or nicotine, while other become dedicated vapers.  Some smokers find nicotine alone to be an appealing substitute, while others find that smokeless tobacco is a good substitute but nicotine alone is not (and, of course, many find neither to be satisfying).

The only plausible hypothesis that is both consistent with the data and maintains the centrality of nicotine is that differences in the pharmacokinetics explain the different reactions.  Cigarettes deliver a much faster Tmax (time until the peak level is reached) for nicotine than any other tobacco product and unless the user of a smoke-free product uses a lot for an extended period, cigarettes also produce a higher Cmax (how high that peak is).  That this would result in greater captivation is consistent with what we know from research on other drugs as well as human experience.  Under this hypothesis we would expect that smokeless tobacco, e-cigarettes, and NRT would have similar levels of “addictiveness,” and those would be less than those of all variations on smoking.  This does not tend to conform to the current interpretations of the evidence (though, as noted, that evidence is sketchy).

Tobacco includes other psychoactive chemicals that might act independently or synergistically with nicotine to make it more captivating than nicotine alone.  There are people who find that nicotine alone is not a particularly pleasant drug, but that whole tobacco is.  The systematic affirmative evidence that these constituents are the important difference is limited to rat studies (example) which have little relevance to human preferences and behavior, but do suggest biological plausibility.  It theoretically also may also be that the smoke itself has constituents that make it more captivating.

Inhalation versus other delivery methods is another frequently proposed contrast, since delivery is more rapid and might reach different receptors.  However this hypothesis is quite inconsistent with the data (e-cigarettes use is unlike smoking; smokeless tobacco use may be somewhere in between; NRT inhalers are claimed to not be addictive).

Alternatively, the contrast may be caused by nothing more than established habits.  Those who quit the nicotine-only products — either e-cigarettes or NRT — had been using them for far shorter periods than the average smoker or smokeless tobacco user who tries to quit.  A lack of established routine or change of routine makes behavior change easier.  This hypothesis is compatible with nicotine being the major contributor to the appeal of the products, but suggests that the captivation has other causes.  A related hypothesis can be found in discussions about how to market alternatives to cigarettes, with an emphasis on consumer “moments.”  If the marketers are right, and much of the appeal of smoking is that it creates focused carved-out short periods, then the context of nicotine delivery, and not just the chemical, is critical.

Whatever the explanation, it is clear that the claim that nicotine is highly addictive is naive and unsubstantiated.  Overwhelming evidence suggests that e-cigarettes do not have the same captivating properties as smoking, and a similar claim is made about NRT.  There is no apparent basis for claiming that nicotine alone, absent a history of smoking, produces “addiction.”

Our position

The concept of addictiveness is a worthwhile area of study and concern, but in the politics of tobacco products, it is used merely as rhetoric.  Many of the same people who tout NRT condemn e-cigarettes because of the supposed addictiveness of the nicotine they deliver, even though there is very little difference in nicotine delivery between the two categories.  If asked what they even mean by the word, almost no one condemning e-cigarettes because the are addictive could provide a remotely coherent answer (and even the few reasonably cogent answers are still scientifically imprecise and either do not apply or do not reflect the dire implications usually assigned to the word).  These observations alone are sufficient to show that the supposed worry is disingenuous.  The lack of evidence that nicotine alone is particularly captivating, and the affirmative evidence that suggests it is not, means that there is no scientific basis for such worry.  Unless some other evidence emerges, or products are developed that have different capitvation properties, the trope that nicotine is addictive should be seen as mere political rhetoric or blind repetition by those who do not know it is political rhetoric and believe there is such evidence.

It can be argued that even cigarettes do not meet a reasonable definition for “addiction,” a word that evokes the much more intense and all-consuming reactions that some people have to the use of drugs like cocaine, amphetamines, or strong narcotics.  Still it does appear that some tobacco products are more captivating than others and that cigarettes are the “most addictive” among them (granting that since “addictive” is ill-defined, the phrases “most addictive” is perhaps unforgivably ill-defined).

This creates some considerations that matter little to the tobacco control extremists, but which are important to those interested in health and consumer welfare.  Improving the quality of a THR product will make it a satisfying substitute for more would-be smokers and will increase the welfare of those using it.  To anti-tobacco extremists, this very appeal is a downside and they tend to call it “abuse potential,” but to those whose motives are not dominated by a quasi-religious objection to tobacco use, making people happier, ceteris paribus, is considered a plus, not a minus.  However, to the extent that those same quality features make it unpleasant to quit (as opposed to pleasant to use — a subtle but important distinction), should someone decide that is what they would prefer, there is a downside to quality.  The perfect low-risk product would be highly appealing and a very satisfying substitute for smoking, but the only barrier to quitting would be that it is appealing and thus there are benefits that the user does not want to give up.

The current holy grail of e-cigarette developers seems to be to make them as much like cigarettes as possible.  This has obvious benefits for harm reduction, but has the downside that it might be making them more captivating (a downside for consumers, that is — for the producers this may be considered an additional benefit).  If there were some legitimate research into the concept of “addictiveness” that did not merely conflate it with attractiveness, it might be possible to figure out how to create more of the latter and less of the former, which would be beneficial for consumers.  Unfortunately, this is a line of research that is unlikely to be pursued by the tobacco control special interests or the industry, and they control almost all of the available research funding.