Phillips testimony for FDA hearing on recognizing tobacco harm reduction

by Carl V Phillips

The title overstates a bit, but next week’s FDA hearings (official government announcement here) on whether to allow pharma nicotine products to be able to officially (“on label”) be used for long-term THR rather than just for the magical rapid weaning off of nicotine that they are supposed to accomplish.  Since they are approximately useless for the latter and are mostly used for the former anyway, this is not a huge stretch.  But recognizing the reality of THR would be a big step for the FDA, similar to what the UK’s MHRA has already done.

FDA is not really admitting that this opens the door to properly recognizing THR, but it does, and so several of us who are interested in THR are testifying on various points.  Annoyingly, the government wanted us to submit not just our slides, but an actual text of our statements a week in advance.  There are various conspiracy theories about the reason for that, but mostly I just find it a bother since I never write out the text of my talks, but usually put a lot of detail in the slides.  But since I had to do it, I might as well make use of it here, so….

Testimony for: FDA Actions Related to Nicotine Replacement Therapies and Smoking Cessation Products; Report to Congress on Innovative Products and Treatments for Tobacco Dependence; Public Hearing

by Carl V Phillips PhD

17 December 2012; Silver Spring, Maryland

I am testifying as a private citizen and stakeholder, speaking for myself based on extensive professional experience and expertise.  I am a long-time advocate and researcher on tobacco harm reduction.  I have worked in this field for more than a decade as a researcher and educator, and have published roughly a million words on the topic.  I am also an informed, consciously-motivated, occasional user of low-risk tobacco products, which I find to be quite beneficial, and clearly worth the unmeasurably small risk that they pose.

In response to Question 4.4, which asks about identifying harm in this context and identifying which harms are of particular importance, I have the following observations:

[Note:  Full text is: 4.4. With regard to innovative products and treatments for “reduction in the harm associated with continued tobacco use”: a. How should the “harm” be identified and measured? b. Is there a range of harms that might be addressed, and if so, which are the most important to address?]

Harm should refer to actual harmful effects.  This seems rather obviously, but there is a remarkable failure to follow this guideline in the discourse on this issue.  Life-threatening disease risk is an important harm.  Immediate problems of functioning are also an important harm.  “Addiction” is not a harm in itself (even setting aside that it is not even scientifically defined).  It might be a reason that people do something that actually causes harm, but it is only that actual harm that should be counted.

“Harm” must be considered in net term, as is done for treatment drugs or anything else the government properly regulates.  Nothing is harm-free, but harm may be low (or trivial) compared to benefits.  The common approach of naïve extremist activists approach, saying “this has a harm and therefore is bad” makes terrible public policy, but that is what you get if you do not consider full net costs and benefits.

To consider net effects, first requires considering the health benefits of using a tobacco harm reduction (THR) product as compared to abstinence.  There are psychological benefits and aids to basic functioning from nicotine use for a non-trivial minority of the population, and this is a highly important consideration.  Second, the net must consider health benefits that may not be considered “health”.  Acting as if nicotine is purely a treatment for disease, rather than recognizing it as a consumer good that serves other preferences, is simply out of touch with the obvious reality.  Many people use nicotine because they like the effects, and this is another highly important consideration.  If the “D” side of FDA are not accustomed to dealing with such goods, perhaps those on the “F” side could help (or anyone else in government who regulates any non-drug consumer goods).

Third, net health benefits need to based on the probability-weighted “but for” choices.  For example, if there is a 20% chance someone would be abstinent, but for the promotion of the THR option, and 80% they would continue smoke, then you need to compare any harms from the THR to .8 times the risk of smoking (a comparison that overwhelmingly favors the latter).  It is simply out-and-out wrong thinking to compare the harm from one option to some unrealistic world that some people consider a best-case.  This is not an odd concept: treatment drugs are evaluated based on their net effects, considering the harm from not getting the treatment.  Moreover, the calculation of the net harm needs to recognize that some of the methods for pursuing abstinence (e.g., Chantix) are quite hazardous in themselves.  And, most important, if someone tries to achieve abstinence but relapsing for a while, and as a result smokes for just a couple of more months, the resulting risk is (for the average smoker) greater than from than using a low-risk alternative for a lifetime, as I argued in my 2009 Harm Reduction Journal paper.  This means that pursuit of abstinence is often a higher risk alternative to THR.

In response to Question 4.6, which asks about how FDA and other HHS agencies can help promote public health in this context,…

[Full text: 4.6 In regulating the innovative products and treatments referenced in section 918(b), how can FDA and other HHS Agencies act to ensure that the three effects mentioned in section 918(b)–total abstinence from tobacco use, reductions in consumption of tobacco, and reductions in the harm associated with continued tobacco use–are achieved as broader outcomes, in a manner that best protects and promotes the public health?]

…far and away the most important thing to do is:

Stop lying!

Units of HHS are the world’s biggest anti-THR liars, and have been since I started documenting anti-THR disinformation campaigns about a decade ago.  The pervasive message to smokers about low risk tobacco and nicotine alternatives coming out of HHS has been and still is “you might as well smoke” (a phrase that was the title of my 2005 BMC Public Health paper that reports some of my research on this disinformation).  FDA has now joined other agencies in publishing disinformation designed to discourage smokers from switching to low-risk alternatives by grossly exaggerating the risks of those alternatives using false and intentionally misleading claims.

These anti-THR messages encourage people to smoke when they would otherwise use a smoke-free alternative that poses risks that are two orders of magnitude lower.  No conceivable amount of low-risk product usage could cause as much health harm as the smoking this causes.  Following the phrases from Question 4.6, respectively “total abstinence from tobacco use” and “reductions in the harm associated with continued tobacco use”, the current practice of promoting the former at the expense of encouraging the latter is guaranteed to be bad for public health.  Obvious and simple arithmetic makes that clear.

Regarding Question 4.7, about what to consider in FDA’s premarket evaluation of new product candidates,…

[Full text: 4.7. How can these broader outcomes be taken into account in FDA’s premarket evaluation of new product candidates?]

…I reiterate that focusing on the net on smoking is most important.  Indeed, from a population health perspective, it is the only thing that matters.  Everything else is a rounding error.

In addition, premarket evaluation depends on having a correct basic model for what motivates consumers to choose particular products.  People act based on incentives which include health risks, of course, but also include many other factors.  Consumers making choices are not mere patients – sick people seeking treatment for a disease and nothing more.  If they are treated as such when predicting what they will do, the predictions will inevitably be wrong.  Even worse, if they are just treated as black boxes who are assumed to just make choices at particular rates because that is sort of what other consumers did in the past under difference circumstances, then you will have garbage in yielding garbage out.

I have a lot more to say about this, but only seven minutes total, so that is where I will end.