Science Lesson: How We Know Vaping Poses Little Or No Risk
Assessments about risks – from smoking, smokeless tobacco use, driving, obesity, or whatever – are usually based on epidemiology, observations about the health outcomes of people who have the exposure. But for vaping, we only have a few years of observations, and the data is difficult to interpret because of the many and varied effects of past smoking. Yet, it is still possible to conclude that vaping is almost certainly approximately harmless.
Our basis for that conclusion begins with smokeless tobacco. ST use allows us to observe the consumption of nicotine without the harms from inhaling smoke. Fairly extensive epidemiology shows that there are no measurable risks from using the varieties of ST that are popular in Europe and North America. This makes it safe to conclude that consuming nicotine causes no measurable risk. (It is theoretically possible that some other aspect of ST exposure counteracts some negative effect of nicotine alone, but this seems unlikely.) Despite their ceaseless rhetoric about the harms from ST and nicotine, FDA endorsed this conclusion as part of their approval pharmaceutical nicotine products (NRT) for long-term use.
Absent any other information, it would be a good start to assume that the risk from vaping is the same as the (negligible or zero) risk from ST use.
Of course, vaping is not quite the same as ST use. It involves inhaling various chemicals which creates other potential pathways for harm. But we already know quite a lot about exposures to those chemicals, mostly from occupational epidemiology which often provides information about exposures that are novel and small for consumers but present at high levels in some workplaces. Based on that and the quantities of the chemicals in vapor, Laugesen, Burstyn, and others have assessed the risks to be negligible.
Despite all the rhetoric about us not knowing what the effect of inhaling vapor, the only aspect of the exposure that is novel is inhaling the large quantity of the carrier chemicals, propylene glycol and glycerine. That quantity turns out to be a bit beyond what is well-understood from occupational exposures. However, since these chemicals appear to be benign (beyond minor irritation that some people experience and can immediately recognize), they too pose little worry.
Ongoing toxicology research provides a reality check on those predictions. It never hurts to test and confirm our predictions, no matter how confident we are. This research, mostly being conducted by the major tobacco companies, confirms that exposing cells to vapor does not produce the effects that we understand to be precursors to disease.
Notice two areas of research that have not been mentioned.
The first is harm from smoking. There is absolutely nothing to be learned about the health effects of vaping from smoking. The exposures are far too different. Those who suggest smoking is a useful starting point – because vaping looks like smoking or substitutes for it – are not only wrong, but demonstrate they simply do not understand health science. It is akin to suggesting that the risk from cave diving in shark-infested waters is a good starting point for understanding the risk from doing laps in a pool (“both are swimming!”). This would be true even if ST use did not exist, but given that it does exist, it is especially absurd to ignore the useful comparison in favor of a useless one.
It is theoretically possible to estimate the risk from vaping and then compare it to the risk from smoking, though this turns out to be a bad idea. But you cannot get there by starting with the risk from smoking and trying to adjust from it.
For different reasons, short-term epidemiology adds basically nothing to what we already know. A couple of recent studies of lung function were overhyped as showing that vaping does not harm the lungs. But long-term health effects are often preceded by a period of no noticeable changes. In addition, these studies tend to only look at a few measures of possible harm in a few people. Any harm from vaping must be subtle and unpredicted, something these studies are unlikely to detect. It is still a little bit useful to have a reality check about predictions being wrong, but we would have to be wildly wrong for these studies to provide any new information.
Contrary to the uneducated claims of naysayers, we can be quite confident that any risk from vaping is trivial based on what we already know from the epidemiology of smokeless tobacco and occupational exposures, along with the reality check from toxicology experiments. That is possibly the best we can ever expect. It is not clear that we will ever have epidemiology that offers a more precise estimate. Comparing vapers to other ex-smokers would be an interesting exercise to demonstrate their similar health outcomes, but the residual confounding – from slight differences in smoking history, pre-quitting health status, or other differences between the groups – would be larger than the possible risk from vaping. It would not allow us to refine our estimate of the risk beyond the “trivial or zero” estimate we already have.
Also contrary to the uneducated claims of naysayers, if smoking were invented today, we would not be in the dark about it causing substantial risk. The reasons why will be covered in next week’s science lesson article.
Follow Dr. Phillips on Twitter