Contrary to popular belief, smokeless tobacco is not a major health hazard. In fact, it has not been definitively linked to any deadly disease. More important, to the extent that it causes any risks, the scientific evidence clearly shows that the risks are very small (See Lee & Hamling 2009).
This might surprise you. We will admit that it surprised every one of us when we first learned it. Because there is so much misinformation about ST, much of it spread by otherwise legitimate health authorities, it can be difficult to find the truth. Below, we take up the question of how to sort out truth, lies, and genuine scientific uncertainty. Here, we present what the scientific evidence actually shows. (Reading list.)
Doesn't smokeless tobacco cause oral cancer (cancer of the mouth)?
-If so, very rarely.
There is overwhelming evidence that any risk for oral cancer (cancer of the mouth) from ST is very low.
This is an important question in the ST debate, so here is a little background first. Oral cancer (cancer of the mouth and surrounding areas) is a fairly rare disease in Western countries, and in most cases is caused by smoking. Smoking can increase your risk of oral cancer by eight or ten times, much more than even the worst-case-scenario estimate for ST. The American government attributes about 75% of all oral cancer deaths in that country to smoking combined with drinking. In other words, the best thing you can do to reduce the risk for oral cancer is not to smoke.
Even if ST did cause a measurable increase in oral cancer risk, the total risk would still be small. The "baseline" risk for oral cancer (the risk for people who do not smoke or drink a lot of alcohol) is quite low. So even if ST multiplied that risk by 2 or even 4, the total risk of dying from it would still be very small compared to the risks of the lung and heart diseases caused by smoking. (Consider if you had a choice between doubling your chances of dying by getting hit by lightning versus doubling your chances of dying in a car crash. Doubling your risk from lightning would hardly matter because it is so rare in the first place; two times a very small number is still a very small number. But doubling your risk from a car crash would be a lot more worrisome.)
In one of the most remarkable marketing successes ever, anti-ST activists have convinced people that, "if I switch from smoking to dipping, I will just trade lung cancer for mouth cancer." In reality, switching from cigarettes to ST will dramatically reduce your oral cancer risk (and your risk of lung disease, heart disease, and many other diseases), and even if that were not true, the tradeoff would still be a good one because the risk of oral cancer for non-smokers is so low in the first place.
Having said all that, what is the answer to the original question? What does the scientific evidence say about the risk for oral cancer from smokeless tobacco?
It turns out that the evidence clearly shows there is very little risk. When looking at scientific research, it is necessary to look at all of it, not just one or two particular studies. As with most things we study in health science, the results vary. A few studies find that people who use ST have a higher risk for oral cancer but most studies have found that the risk is very low. Some even show a negative association. This is the same to say, if someone just picked out those studies and ignored the rest, they might conclude that using ST protects you against getting oral cancer. Of course, picking just a few studies with extreme results is just bad science.
When we look at the all the scientific evidence, it averages out to there being either no risk or very little risk (it is impossible to tell the difference between those two conclusions because health science methods are always imperfect). By "very little", we mean that using smokeless tobacco might cause a 10% or 20% increase in the risk for oral cancer compared to not smoking or not drinking heavily. The evidence shows it is extremely unlikely that the increase in risk is as high as 50%(which is still much less than the risk from smoking). (Oral cancer reading list.)
If you do not want to delve into the technical details, the previous points and our calculation of thecomparative risk of smoking and smokeless tobacco are all you need to know.
But doesn't smokeless tobacco cause oral lesions which can become cancerous? - If so, very rarely.
Much of the following is taken from Rodu and Godshall's Tobacco harm reduction: an alternative cessation strategy for inveterate smokers. See the original for further reading or for details on the research behind the conclusions.
Oral leukoplakia is an ominous sounding term used frequently in discussions about oral lesions resulting from ST use. The term literally means "white plaque," and it is used to describe areas of the mouth lining that become thickened by ST use or smoking. The World Health Organization has determined that leukoplakias resulting from ST use are considerably different from those resulting from smoking. The distinctions are based on the frequency of occurrence, the location in the mouth, and how often these leukoplakias result in mouth cancer.
The condition is rare, occurring in less than 1% of the general population, primarily in long-time smokers 40 to 60 years old. Smoking-related leukoplakias most commonly involve the undersurface of the tongue and throat area, locations that account for 75% of oral cancer in the U.S..
Oral leukoplakias occur in up to 60% of ST users, within 6 months to 3 years of starting ST use. They primarily occur at the site of ST use and are largely a result of local irritation. The frequency of appearance depends on the type of ST that is used. Moist snuff, which is more alkaline than chewing tobacco, more often leads to leukoplakia. However, moist snuff in pre-portioned pouches (like snus) causes fewer cases of leukoplakia than does the loose form.
There are distinct differences in how often ST and smoking leukoplakias show pre-cancerous changes called dysplasia. Dysplasia is seen infrequently in ST leukoplakias (less than 3%). Furthermore, even when dysplasia is present in ST leukoplakia, it usually is found in earlier stages than in leukoplakias among smokers, where it is seen in about 20% of cases.
ST leukoplakias only rarely progress to cancer. For example, one study found no case of cancer in 1,550 ST users with leukoplakia who were followed for 10 years, and a second study reported no case of oral cancer among 500 regular ST users followed for six years. A retrospective study of 200,000 male snuff users in Sweden found only one case of oral cancer per year, an extremely low frequency. In comparison, a follow-up study reported that 17% of smoking leukoplakias transformed into cancer within seven years.
In conclusion, oral leukoplakia occurs commonly in ST users, but it primarily represents irritation and only very rarely progresses to oral cancer.
Does smokeless tobacco cause other cancers?
- If so, very rarely.
Again, the answer has to be that there is no conclusive evidence that it does, but it is certainly possible that it does at some very low level. There is no clear evidence that ST causes any cancer. Hence, the evidence indicates that if there is any risk, it is small.
Other than oral cancer, the cancers that seem most likely to be associated with ST are laryngeal, esophageal (throat), and gastric (stomach), since those are the other parts of the body that come in contact with the tobacco. For these sites, along with a few others, there is plenty of scientific evidence that there is no substantially elevated risk. As with oral cancer, researchers have looked for an association and have failed to find one. For other cancer sites there are few or no studies, so we do not have any evidence one way or the other. However, studies that look at all cancers combined have failed to find an increase among smokeless tobacco users, so there cannot be a very big increase for any particular cancer.
Does smokeless tobacco cause heart disease or stroke? - The risk is probably quite small.
There may be some small risk, though there is no definitive evidence. Most studies of cardiovascular diseaseand smokeless tobacco have found no increased risk. However, because nicotine is a mild stimulant, it might increase the risk of certain cardiovascular outcomes, such as stroke. Many stimulants have been linked to some risk for fatal cardiovascular events. This does not prove that nicotine causes such risk, but it suggests that it is plausible.
The available evidence shows that if there is any risk from smokeless tobacco, it is low, less than a 20% increase (for comparison, smoking is estimated to roughly double this risk, a 100% increase). This is potentially more important than a similar percentage increase in oral cancer risk, since cardiovascular disease is a lot more common. If the risk of cardiovascular disease increases by as much as 20% then smokeless tobacco might cause 3% or 4% of the risk associated with smoking. That is still a lot better than smoking, of course, but worse than one percent. Fortunately, most of the evidence suggests that nicotine without the smoke is not actually quite that bad. But it would be useful to do more research to find out more. Unlike with oral cancer, there is not enough available research about cardiovascular disease for us to feel entirely comfortable in our conclusions.
What does this all mean for the total risk compared to smoking? - Using smokeless tobacco is much safer than smoking.
It means that the risk is tiny compared to smoking. You might hear other estimates but as far as we know, we are the only ones who have actually done the calculation. Our calculations show that if we take a worst case scenario and assess the risks for cancer and cardiovascular disease, the total risk from ST is only a few percent of that from smoking.
Due to the limitations of health science, we cannot be sure exactly how the risks compare. But we can be very sure that the total risk of dying from ST use is less than 1% of that from smoking. That is, for any plausible levels of risk for disease from ST, any values that are not clearly ruled out by the science, the total risk is less than 1/100th that from smoking. There is no legitimate scientific doubt that someone's risk drops by at least 99% by using ST instead of smoking.
So why do so many people say "smokeless tobacco is not a safe alternative to cigarettes"?
The main reason that this phrase is so common is that the U.S. and Canadian governments require a version of it to be printed on packages of ST products. This is unfortunate, since most people interpret the phrase "not safe" to mean "dangerous". The phrase "not a safe alternative to cigarettes" is understood as "just as bad as cigarettes".
Strictly speaking, the statement is true. If we interpret "safe" to mean "100% safe; creating no health risk at all" then smokeless tobacco is not "safe" (and neither is anything else. Chances are that somewhere, sometime, somehow, broccoli has killed someone). Smokeless tobacco may not be a "safe" alternative, but neither are any of the other products used to help people stop smoking. Certainly trying to quit and failing, or smoking for another year or two before quitting are not "safe" either. ST is much safer, however, and that is what matters.
Think about this: Driving safely and properly is not a safe alternative to speeding while drunk and not wearing a seatbelt. But it is very close. It might even reduce the risk by as much as 95%. Somehow, there seems to be no confusion about which of those options we think someone should choose. There should be no confusion about tobacco products either. Smokeless tobacco is much less harmful, and the fact that it is not 100% safe does not change this.
Are you saying that smokeless tobacco is harmless? - No.
As far as we know, no one is saying that. As we noted, it is possible that smokeless tobacco causes cancer or other deadly diseases at some very low level. We know that the nicotine has short-term cardiovascular effects which may be a little bit harmful. Nothing is 100% harmless, and this includes smokeless tobacco. This does not change the fact that it is a much less harmful alternative to smoking. (Reading list.)
What about all the chemicals in smokeless tobacco that I hear about?
Sometimes people who are opposed to harm reduction or any use of ST present a list of potentially harmful constituents. What they do not tell you is that many of those chemicals are in other plants, including the healthy vegetables that you eat. A little bit of all the metals on Earth,including cadmium, polonium, and others that are quite bad for you in large quantities, end up in everything. They are in tobacco, carrots, wheat, and even the water you drink. There are also organic molecules like formaldehyde in most every life form on Earth, including tobacco. Again, concentrated in large quantities, you would certainly want to stay away from some of these chemicals, but the amount that naturally occurs in plants (and in your own body) is not a problem. We are lucky that there are not a lot of anti-broccoli advocates out there trying to trick you into not eating it because it contains some cadmium.
One set of chemicals that are talked about a lot in the science are nitrosamines, or tobacco-specific nitrosamines (TSNAs). Nitrosamines are a class of chemicals that we are exposed to in food and through other pathways. Some of these are known to be carcinogens in some quantities. Whether the TSNAs might be carcinogens is the subject of debate. Fortunately, it does not matter whether these chemicals might cause cancer since we have evidence about whether ST itself causes cancer. The evidence shows that people who use ST do not have measurable increases in cancer rates or mortality. So it does not really matter what chemicals are there.
Think about this: If you met a thin person who always ate huge amounts of food, would you tell him that eating all that food is making him overweight? Obviously not, since you can directly observe that he is not overweight. All this talk about chemicals is like that. If all you knew about a person was that he always overeats, it might suggest that he would be overweight. Similarly, if you only knew about the chemicals and had never observed the actual health effect of the products, the chemical analysis could suggest that there might be a health effect. But once you observe that someone is thin (or observe that smokeless tobacco users do not have elevated disease rates), the suggestive evidence is no longer informative. It can only be used to trick you into thinking there is a health effect when the evidence actually says otherwise.
Doesn't smokeless tobacco contain fibreglass? - No.
The fiberglass thing is a complete myth. There is no evidence for this, though it gets repeated by many anti-ST activists. We have heard the claim that fiberglass creates little cuts which allows absorption of nicotine. However, nicotine is absorbed quite well through the intact and healthy inside surface of your mouth, so there is no reason to do this. Creating holes in the surface, thus causing bleeding, might well interfere with absorption.
The only legitimate reference we have been able to find for the use of fibreglass in tobacco products is in some cigarette filters.
I know someone (or read about someone) who got cancer from using smokeless tobacco. How do you explain that?
We offer our sympathy if someone in your life got cancer, but it wasn't necessarily caused by smokeless tobacco. The same is true for the various oral cancer victims whose stories are used in the anti-smokeless-tobacco literature. Just because two things coincide does not mean that one caused the other. These cases are almost certainly coincidences.
You can understand the reasoning behind this point without needing much math. In a large population (like North America), even if we are talking about something that relatively few people do (such as use ST) and a fairly rare disease (e.g., oral cancer), there will still be quite a few people who fall into both categories just by chance. For example, if 1-in-100 people are exposed to a substance, and 1-in-1000 people get a disease, then by chance (coincidence) alone we would expect about 1-in-100,000 people to both have that exposure and get the disease (1/100 x 1/1000 = 1/100,000).
That does not seem like a lot, until you realize that there are over 300,000,000 people in North America. So it is not strange then that you will find someone with oral cancer who used ST. There will be quite a few such people, just by chance. In our example above, 3000 people would have both the exposure and the disease by chance alone. That is why epidemiology (the science that explores whether a disease is caused by an exposure) always uses a "comparison group" (people who do not have the exposure) to see if those with the exposure have more cases of the disease. The evidence does not show that ST users have more cases of various diseases than the regular population, but of course they will have some cases.
Consider this analogy: Some people read a lot of mystery novels (the exposure); some people get pancreatic cancer. If we told a scary story about a mystery fan dying of pancreatic cancer, and posted gory pictures on the web, would that be evidence that the novels caused the disease?
Since oral cancer is usually caused by smoking, finding someone who never smoked who has oral cancer means looking for something unusual and misleading. It is as if someone collected pictures of people who had died in car crashes while wearing seatbelts and argued that seatbelts are bad for you. We know that wearing a seatbelt is much better than not wearing one, but there will still be people who die even when they choose the safer option. It is completely irresponsible to emphasize those people and discourage people from making the safer choice.
But what about cases where someone, perhaps a doctor, said that ST caused a particular disease? Well, there is a strong human tendency to look for a specific cause, especially for highly traumatic events, even when such a determination is impossible to make. Causes for some diseases are easy to see (if you break your arm while falling down, it is pretty easy to figure out why), but this is not true for cancer and cardiovascular disease. So, when people have been repeatedly told that ST is a terrible health hazard, it is not surprising that they end up thinking that it caused a disease.
For a few diseases and exposures, we can conclude that it is likely that the exposure caused the disease. If someone who has smoked for fifty years gets lung or oral cancer, we know that it is quite likely that this was caused by smoking. But we know that because when we compare smokers to non-smokers, the disease rates for smokers are much higher. If smokers have 9 times the risk of non-smokers for a particular disease, when one of them gets that disease there is a 90% chance it was caused by the smoking (and a 10% chance that it would have happened anyway).
As a final, more technical note, most of the oral cancer victims who are exploited in the anti-smokeless-tobacco propaganda suffered from tongue cancer at a fairly young age. To the extent that any evidence suggests there is a link between Western smokeless tobacco and oral cancer, the related cancer occurs at ages greater than 60, after many decades of use, (and it is not cancer of the tongue). In developed countries, almost all oral cancer occurs at age 60 or older. By contrast, there are a small number of young people, generally under 30, who get tongue cancer (in the range of one to two hundred per year in North America, which has increased over the last three or four decades). These cancers do not appear to be associated with smoking and heavy drinking nor with smokeless tobacco use. We do not know what causes these cancers, other than knowing that it does not seem to be any of the usual suspects. Of course, by coincidence, one or two of these victims every year will be smokeless tobacco users (and about 40 of them smokers, some heavy drinkers, and so on). Taking scary pictures of one of those unfortunate individuals provides absolutely no evidence about the dangers of smokeless tobacco, and for a health related initiative, is highly inappropriate.