Mad Stan clutching at straws (again)

Today Sara Kalkhoran and I published “Modeling the Health Effects of Expanding e-Cigarette Sales in the United States and United Kingdom” is the beginning of one of Frampton’s latest blogs. It is always amusing (if a little difficult to digest his grasp of the English language) to read these as most of the time (95% CI) it’s all based on mythical hypothesis and conjecture, none of which surprises me when it comes to the “leading tobacco control activist”.

It becomes more amusing when he says things like “excellent accompanying commentary” and “excellent studies” when in fact reading the ‘commentary’ and ‘studies’ they are anything but.

So what has driven Mad Stan to put his fingers on the keyboard (other than to drive us all insane with his language skills)? Well, two things unsurprisingly and both refer to papers published in the Journal of American Medical Association. The first of the two has been bounced around for a while already, but it seems Mad Stan has only just heard about it, either that or it’s taken him this long to figure out what the paper was all about and how he could work it to his advantage (which hasn’t really worked).

PowerfulEvidence

Of course, this is Mad Stan so anything that lends itself to being manipulated is “powerful” or “excellent”. But we already know the paper itself isn’t worth the time or the effort as it doesn’t really show us anything we didn’t already know. Kids try stuff, it really is that simple. Once again though Mad Stan completely misses the point:

Never smoking youth who used e-cigarettes at baseline were 3-4 times more likely to be smoking cigarettes or other combustible tobacco products a year later.

Hate to break it to you Stan, but there were an awful lot of assumptions, cofounders and exceptions to reach those figures which the authors themselves admit isn’t exactly hard evidence. I’ve read the study, and whilst yes it can be interpreted in the way that Stan has done it isn’t what it actually says. There wasn’t any effort made to identify if the e-cig users became regular users as no attempt had been made to measure actual use it only measured if the kid had tried an e-cig. Don’t forget that the “baseline” excluded all self-reported users of tobacco all, 768 of them who would likely have moved onto vapourisers and not gone back which would have defeated the whole point behind the study; which is of course to try to find ways of “proving the gateway theory”.

There’s two sentences which I’m just going to copy and paste in here which highlight Stans woeful understanding of the study and the wider population level implications of vapourisers:

This is a huge effect that is consistent with the earlier cross-sectional studies that suggested a “gateway effect” of e-cigarettes in promoting smoking of conventional cigarettes.

But this paper demonstrates quite unequivocally that never smoking kids who use e-cigarettes are a lot more likely to become smokers.

If you mean “trimming unnecessary participants out” then yes it is consistent but that isn’t the point. The methodology of the study doesn’t allow for sustained use, whether a baseline tobacco smoker switches to vaping, or in fact if a participant is actually followed up at the defined intervals. Yes, you read that right. If a participant only showed up for one follow-up (either the 6 or 12 month), they were included for analysis. So if a participant shows up for the 6-month follow-up, but not for the 12 the information gathered is still analysed as part of the 12-month follow-up.  True there are going to be statistical covariants were applied to basically “best guess” the responses, most likely answering “yes” to certain questions. Stan’s conclusion?

Reasonings

“It’s a big problem that must be acted on by the FDA and other regulatory bodies”; In other words, regulate them out of existence. But there’s more, and this refers to the second JAMA study which is all about trying to predict the future. No seriously. The authors are trying to figure out if there are any negative health effects of the “expanding e-cigarette sales in the US and UK”, odd thing about this particular study, and something that intrigued me was of course how it was funded. Checking the grant reference for Mad Stan as listed in the study (grant reference 1P50CA180890-01) it was awarded in 2013 to the tune of $3,977,218. Now of course two years have passed since that grant was awarded to Stan (via UCSF of course, not directly to Stan) for the project “Improved Models to Inform Tobacco Product Regulation”.

Just shy of FOUR MILLION US Dollars to put together wishy-washy models that have little or no basis in the real world. I’m totally in the wrong job. So what has Mad Stan come up with? Not much as you’d expect. The author, and no I’m not including Mad Stan as an author even though his name is on the paper, developed a model that “quantifies transitions from an initial state of no cigarette or e-cigarette use to 1 of 5 states:”

  • Never use of either
  • Cigarette use
  • e-cigarette use
  • dual use
  • quit

From the design, setting and participants section of the abstract alone it’s pretty clear that whilst there is some ‘real world’ data, the rest is pure speculation. Computer generated speculation for the separate sets of “model parameters” for both US and UK populations. Intriguing stuff, but as with any speculative model there are hundreds upon hundreds of variables, life affecting events that simply cannot be accounted for. Using this speculative model, the results are pretty much what you’d expect to see:

Maintain the status quo

Nothing changes, e-cigarette use increases exclusively among smokers interested in quitting, and no increase in the uptake of e-cigarette use among never-smokers showed population level health benefits, regardless of “health costs” of e-cigarettes in both the US & UK.

Uptake of vaping amongst smoking youth

Again, not much changes. More smokers quitting, net population-level health benefit.

Renormalisation of smoking

Here we go again, the old “renormalisation” argument (which by the way is still not the case, even with the latest CDC data) where “aggressive promotion” leads to an uptake of both tobacco and e-cigarettes in youth never-smokers shows a net-harm. Well of course it would wouldn’t it? If you have an uptake of a product that is harmful (like smoking) then there is always going to be an overall net-harm. I suspect that in this particular “scenario” the uptake of tobacco exceeds the uptake of vaping.

All other scenarios had “varied” net health effects.

The author concluded that “widespread promotion of e-cigarettes may have a wide range of population-level health effects”, which is absolutely true. The reason here is that e-cigarettes are one of many factors that can affect population level health, however the author(s) went on to state:

Absent the primary effect of e-cigarette promotion being only to divert current or future conventional cigarette smokers to e-cigarette use, the current uncertainty about the health risks of e-cigarettes, increasing e-cigarette use among youth, and the varying health effects at different e-cigarette health costs suggest a potential for harm.

So whilst the first line of the conclusion suggests that e-cigarettes as a whole have a varying degrees of impact on population level health, the “current uncertainty” surrounding ‘health risks’, and youth uptake, the authors conclude that there is potential for net-population harm. Oddly enough, this isn’t far from the truth but that isn’t what Mad Stan wildly penned on his blog.

The key finding of the paper is that under all the likely scenarios, if ecigs turn out to be 20-30% as dangerous as conventional cigarettes there will be net harm.  No one know what the actual risks will be, but I would be surprised if the risks are not at least that high.

Mad Stan has been very quiet since the release of the PHE Review, no doubt feverishly putting the finishing touches to his latest masterpiece of deception, but he is assuming that vaping is 20-30% as dangerous as conventional cigarettes. I wouldn’t mind too much, but as always Mad Stan is showing his incredible tenacity for wishful thinking:

The real situation is more complicated.

The entry of e-cigarettes is having several interacting effects:

  • Some people are switching to ecigs
  • Some people end up continuing to smoke because of dual use with ecigs
  • Some people (kids and young adults) initiate nicotine use with ecigs, some of whom go on to smoke cigs
  • The actual long-term risk of ecigs are not known, but there are reasons to think that they could be substantial (albeit lower than cigarettes)

So then. People switch to ecigs, this is a good thing overall. It means that tobacco prevalence declines, which the latest CDC data shows (again). Some people dual use, again a good thing. It is after all, harm reduction. It isn’t necessarily the level of harm reduction that folk would like to see, but it is harm reduction nonetheless. Initiation of nicotine use with ecigs then going on to smoke. The gateway argument. Again.

Of course, we know that according to the data from the UK this is not happening. The gateway to cigarettes is closing, we also know that the gateway is closing in the US thanks to the latest CDC data. Whilst it is still a valid concern that some may decide to move on to cigarettes, the data gathered so far is showing the exact opposite. It may of course change, but it is highly unlikely.

“The actual long-term risk of ecigs are not known, but there are reasons to think that they could be substantial“; based on the research I’ve looked at and making a layman’s “best guess” this statement is utterly superfluous. Whilst there may be some risk in the long-term, it is utterly ludicrous to even begin to suggest one way or the other as a factual statement. Out the 185 studies reviewed for the PHE report, it is reasonable to extrapolate a “best case” scenario where long-term risk is minimal with the occasional outlier. It is true of course that the long-term risks of cigarettes are substantial, especially directly to the user (I’m definitely not convinced about the whole “second-hand smoke” shenanigans). But anyway, I digress as always. Let’s have a closer look at the “scenarios” that these esteemed authors suggest might happen, for this I’m only going to look at the UK.

Potential "pathways"
Potential “pathways”

According to the baseline data available (the real stuff, not the computer generated speculation), 58% of UK adults have never smoked, 19% are current smokers with 23% being former smokers (using the ONS data from 2013). The key results, based upon the other research available and specifically the PHE review suggests that at the “95% safer” region, an increase in e-cigarette prevalence of 10% (from existing smokers) gives a net-health benefit, as does using vapourisers in a harm reduction capacity.

Of course, the authors decided to up the relative harms of vaping to around 50%, which we all know to be utter bunkem. The net-benefit decreases rapidly, as you would expect with a “status quo” balance of no gain or loss when the “health risks of e-cigarettes” is approximately 25%.  Based on all the research, we can expect positive gains if vaping is allowed to continue as it is, and even greater gains if there is a 10% increase in prevalence among smokers switching. Amusingly, the authors predict that the UK (the most liberal when it comes to vaping, at least for the moment) doesn’t fare so well even at the current estimated “95%” figure of ‘safety’ with a 10% increase in smokers switch having only a 25% net-gain instead of the near 60% in the US.

We all know that Mad Stan believes that vapourisers are harmful (what with his “ultra fine particles”), but his post title of “likely to have negative population health effects if more than 20-30% as bad as cigs” once again doesn’t accurately reflect what the study actually says. According to the mad prof:

Bottom line:  The effects of the entry of e-cigarettes into the market is a lot more complicated than just smokers switching to them.  E-cigarettes can also be an entry product for nicotine addiction and can lead to and discourage quitting cigarettes.  Unless e-cigarettes turn out to have very low risks, the net effect on the population will be negative.

There’s a lot of assumptions loosely based on research, such as a “five-fold increase in e-cigarette initiation” where they assume one-third of never smokers and two thirds of current smokers take to vaping. The assumption made is that there would be a 20% decrease in interest in quitting smoking among smokers due to “an increase in e-cigarette use in smokefree areas (bans anyone?), and a doubling of e-cigarette use for cessation”; because the mad prof doesn’t believe vaping helps folks quit.

Think we all know what to say to that don’t we?