Unhealthful News 140 - Cigarette addiction gets a lot of play this week, but still not clear what it is

Probably none of this will actually make the news.  That is too bad, because unlike the boring ranting and uninformative "science" about tobacco that does make the news, this actually matters for health and welfare.  I will leave that as my day's critique of the health press, and detour into an essay on the substance of the concept.

The story kicks off with anti-tobacco activist John Banzhaf, who built an empire based on suing tobacco companies, declaring:

although there is evidence that for many people smoking involves addiction, that addiction is to the drug nicotine, not to the act of smoking itself, which is a behavior. Because those who desire to quit smoking (e.g., for a medical procedure) can ingest nicotine from nicotine gum, nicotine patches, nicotine spray, nicotine inhalers, and e-cigarettes, their decision to ingest it by smoking rather than by using nicotine replacement products is a choice. Since it is a choice rather than an addiction, disease, or health status, it seems more legally justified to restrict access to medical care to smokers than to the obese.

As pointed out by Michael Siegel, this is obviously a cynical legalistic maneuver, to justify Banzhaf's current activism to encourage practitioners and payers to deny medical care to smokers.  Siegel is very disturbed by this because it provides a defense in product liability cases against cigarette manufacturers, Banzhaf's previous approach and one of Siegel's preferred anti-tobacco strategies.  He condemns Banzhaf's cynical inconsistency, making scientific claims based entirely on whatever seems to suit his activist goals du jour, though this is kind of like be shocked by two-year-old saying "no!" – it is just what they do.

Siegel contrasts Banzhaf's denial with the statement at the annual meeting of Altria (parent company of Philip Morris USA and U.S. Smokeless Tobacco Company) by CEO Michael Szymanczyk:

...tobacco use is addictive and it can be very difficult to quit... 

This statement is not surprising because major tobacco companies have for years been conceding this and other bits of the anti-tobacco conventional wisdom (some true, some false, and some meaningless).  While this might just be a throw-away line, it was presumably vetted by a team of lawyers and public relations people, so it is interesting to look at attentively.  The use of a conjunction in the statement – "and" rather than "so" – implies that being difficult to quit is not a necessary characteristic of addiction, though most people using the term seem to mean exactly that: addiction is the state of finding it difficult to quit something (though it obviously must really be something more or else it would be synonymous with "enjoyment").

A third assessment of cigarettes and addiction this week, was one that involved some actual analysis.  Hanan Frenk and Reuven Dar just published, in Harm Reduction Journal, the latest in their series of challenges to the simplistic conventional wisdom about nicotine and tobacco.  They are among the few authors with the nerve to take on the anti-tobacco industry and its mantras, and with the patience to sort through and identify some of bad science that is less obvious.  While I do not always agree with their analysis or conclusions, I find they generally offer prima facie effective challenges to the claims they are addressing, and I notice the consistent failure of those who support the original claims to present a substantive response (which tells us quite a lot about whose case is stronger, per Unhealthful News rule number one). 

Their latest contribution is a critique of the claims about addiction in the 2010 U.S. Surgeon General's report about smoking, which is an update of the 1988 SG Report, the document which is pretty much the only go-to citation for those who perceive some value in typing a little superscript number beside a statement that smoking is addictive.  They follow their preferred analytic approach of identifying ostensible empirical support for a claim that really does not support it.  For some studies (e.g., a study of rats pushing a lever to get nicotine as evidence of nicotine being reinforcing) they point out how the politically preferred interpretation is one of many possible explanations (the rats were already trained to push the lever before it started delivering nicotine).  With this, they employ scientific skepticism rather than the usual approach that is employed in public health and psychology research, which is to find a study result that is vaguely consistent with a preferred claim and declare it to be evidence the claim is true.  For other examples they observe that the study results actually contradict the purported conclusion.  In some cases they point out holes in the underlying logic (e.g., acquiring a tolerance for negative effects of nicotine allows greater consumption, but it cannot motivate greater consumption).  These observations do not demonstrate that a conclusion is incorrect, of course, but they undermine the stated basis for claiming that it is correct.

In the present case, the conclusion in question is that the nicotine delivery from cigarettes is addictive.  They implicitly propose alternative stories, that smoking itself is compelling and that believing you are addicted is self-fulfilling.  Frenk and Dar's conclusion, and Siegel's similar passing assertion, in contrast with Banzhaf's new "belief", tends to argue that harm reduction approaches to promoting smoking cessation – which is premised on the belief that low-risk alternative nicotine sources will fulfill a large part of the motivation to smoke – are less promising than they might be.  Personally, I find their hypothesis far from compelling, particular when they involve such claims as "we are not aware of any compelling evidence that nicotine has pleasurable effects in smokers", which seems to take skepticism rather too far.  Indeed, I tend to agree with Banzhaf's view that smoking is most usefully thought of as a choice, though this is conditional on people having knowledge about and access to alternative nicotine products, something that he and his friends have successfully prevented for most people.  (And obviously I am not endorsing his ethically-challenged assertion that we should punish people for that choice with the loss of access to medical care - no remotely humane person would argue that.) 

But though I disagree with where they take their arguments it would be foolish to not consider Frenk and Dar's critiques (though we can be sure that tobacco control types will not do so; you can finish the syllogism yourself).  As they point out, it does smokers no favors to thoughtlessly apply a simple model and label that might not be accurate.

Of course, for the label to be accurate, it has to mean something in the first place, and it is not clear that it does.  No doubt Frenk and Dar, along with most everyone else who uses the term "addiction", have something in mind.  At least this is true of those who are interested in the concept for scientific truth-seeking or trying to help people, as opposed to those who just use the term politically or as an inappropriate synonym for "use".  But vague definitions are not appropriate for ostensibly scientific claims, and likely make for bad policy too.  Remarkably absent from all of these statements about addiction is any statement of what addiction is.

Presumably everyone omits a definition for the same reason they do not define "nicotine" or "cigarette" – they believe that the word has a clear meaning.  But there is no excuse for making the error.  It should be obvious to everyone who is trying to seriously talk about addiction that there is no accepted definition.  Attempts to write a top-down definition often manage to create something so encompassing that it includes oxygen and breakfast in the list of addictive substances.  Even when they avoid such obvious problems, no top-down definition seems to exist (other than ones that are so gerrymandered as to be a joke) that does not include consumption of meth, cigarettes, and coffee, as well as most hobbies and sleeping in a comfortable bed.  (I was observing today that as a kid I would backpack for a week in the heat and humidity of August in West Virginia wearing cheap cotton clothing and wool socks, while today I squirm at the thought of just taking a four mile walk without changing into high-tech non-absorbent modern clothing.  I was fine for years not buying the expensive new tech and did not yearn for it, even finding it a bit uncomfortable at first.  But after I tried it, its benefits were reinforcing and I acquired a tolerance for its downsides, and now I find it extremely uncomfortable to withdraw to by previous behavior.  This story hits all the favorite characteristics of "addiction" definitions.  Does this mean I have a pathological condition, or just came around to a better way to do things?)

The non-absurd semi-definitions of addiction are not top-down, but by analogy.  This approach follows a pattern like, "the destructive behavior pattern that characterizes some users of hard drugs is what we call addiction, and so anything else that is sufficiently similar is also addiction."   Frenk and Dar argue the case that most of what we call addiction to opiods does not characterize the experience of using nicotine.  In so doing, they implicitly offer something of a definition and work with it.  They focus on technical responses, but we can expand to more obvious points:  Neither the rapid destruction of one's life nor the intense, encompassing, captivating desire that the archetypes drugs of addiction produce seems to characterize many smokers.

So if smoking or nicotine is addictive, addiction must mean something other than "very much like intense heroin use".  What?  Without an answer, what business do Surgeon General Reports, lawyers, pundits, tobacco companies, those who hate tobacco companies, medics, and activists have throwing the term around as if it telling us something useful?  And how much damage are they causing to public health by doing so? 

Banzhaf declared without scientific support that, in effect, tobacco harm reduction is the perfect solution to the problem of smoking and it is all about the nicotine.  Siegel, who tends to support THR (so long as it does not involve smokeless tobacco), attacked this because this was (a) strategically inconvenient and (b) wrong (though Frenk and Dar argue that his stated basis for this claim is misguided).  Szymanczyk, whose company does just fine with either continued smoking or THR dutifully repeated the Conventional Wisdom.  Frenk and Dar analyzed and argued the opposite conclusion to Banzhaf; they got some key points wrong, in my opinion, but they analyzed, made specific substantive arguments, and tried to define their terms, an approach seems like the only one that gets us closer to being able to improve anyone's life.  If you work long enough around tobacco politics, you almost forget that science can do that.