On October 13 2017, the Norwegian Institute of Public Health was commissioned by the Ministry of Health and Care Services to update the 2014 report on health risks from snus use. «..The Ministry wants the report to be updated with subsequent relevant scientific publications on potential health damages from snus use, documented by various methods (epidemiological studies, clinical studies and animal studies). The report shall present updated figures on snus use in Norway, and assess exposure and risk. The Ministry also requests that the results should be discussed against the results of the global burden of disease project (GBD) and its estimates of burden of disease attributed to snus use in the population. The report should not include issues such as harm reduction and "gateway", topics that the Ministry has had thoroughly highlighted in other contexts. The Ministry also assumes that research funded by or which otherwise has links to the tobacco industry is not given weight. ”We have limited the report to the use of Swedish snus.
The executive summary first presents our main findings responding to the mandate of the Ministry of Health and Care Services. Subsequently we briefly present the development of snus use in Norway, and a review of the health damages related to the constituents of snus. We then discuss the findings in light of each other and we conclude with answers to the mandate's questions before we end by giving an account of the method used. The remainder of the document describes the different scientific approaches to the mandate's questions in detail.
Health risks from use of Swedish snus
Based on systematic reviews of research on people who use Swedish snus compared to people who do not use snus, knowledge about the properties of the constituents (TSNA and nicotine), animal studies on snus and evidence about adverse health effects caused by use of other tobacco products, we conclude as follows:
Use of Swedish snus among men probably increases the risk of oesophageal and pancreatic cancer.
It is possible that use of Swedish snus increases mortality after a diagnosis of cancer (all types of cancer assessed together), both when the cause of death is considered cancer-related and for all causes.
It is possible that use of Swedish snus increases the risk of cancer of the stomach and rectum.
It is uncertain whether use of Swedish snus affects the risk of cancer of the oral cavity and pharynx, the lungs, the upper part of the stomach (cardia), colon and anus.
It is uncertain whether use of Swedish snus affects mortality after colon and rectal cancer, and mortality after prostate cancer.
There are no studies investigating whether use of Swedish snus affects the risk of cancer in women
Use of Swedish snus among men probably increases the risk of high blood pressure.
It is uncertain whether of Swedish snus affects the risk of heart attack (myocardial infarction), stroke, atrial fibrillation and chronic heart failure.
Use of Swedish snus probably increases mortality after heart attack and stroke, and it increases the risk of reduced cellular function in the vessels.
It is possible that persons who stop using snus after a heart attack may halve the risk of dying during the next two years, compared with patients who continue to use snus.
It is possible that use of Swedish snus has immediate effects on cardiac function (reduced diastolic function).
There are no studies investigating whether use of Swedish snus affects the risk of cardiovascular disease in women.
It is possible that use of Swedish snus increases the risk of non-affective psychosis.
It is uncertain whether use of Swedish snus affects the risk of schizophrenia.
There are no studies investigating whether use of Swedish snus affects the risk of mental illnesses in women.
Type 2 diabetes, obesity and metabolic syndrome
High consumption (> four boxes of snus per week) of Swedish snus among men probably results in a large increase in the risk of type 2 diabetes and of metabolic syndrome.
It is uncertain if low consumption (<four boxes of snuff per week) of Swedish snus among men affects the risk of type 2 diabetes and metabolic syndrome.
It is possible that use of Swedish snus among men leads to weight gain and increases the risk of obesity.
Non-malignant changes in the oral cavity
The available studies are ambiguous regarding a possible association between caries occurrence and snus use.
Use of snus causes changes in the mucous membrane of the oral cavity.
Other health outcomes
It is possible that use of Swedish snus reduces the risk of Parkinson's disease.
It is uncertain whether use of Swedish snus affects the risk of inguinal hernia.
It is uncertain whether use of Swedish snus affects the risk of rheumatoid arthritis.
The use of Swedish snus probably increases the risk of premature births.
It is possible that use of Swedish snus during pregnancy increases the risk of stillbirths, being small for gestational age, reduced birth weight, cesarean section, short-term respiratory arrest and lip/palate defects, and levels of nicotine degradation products (cotinine) in the child's urine.
It is uncertain whether use of Swedish snus changes the risk of preeclampsia, deaths in the first week of life, and cardiac arrhythmias in the newborn.
Snus use in Norway
Almost 20% of adult men and just over 5% of adult women use snus daily. In addition, about 5% of men and just under 5% of women sometimes use snus. Use of snuff is most common among young adults; 25% of men and 15% of women aged 16-24 years use snus daily.
The use of snus increased in the late 1990s among men and approximately ten years later among women. The growth in the proportion of snus users has slowed among young adults of both sexes. However, the proportion is still increasing among men aged 25-49.
Snus users increasingly prefer portion snus rather than loose snus. In the period 2015-2017, almost 70% of men and almost all women responded that they only used portion snus. Eight percent of men and almost no women responded that they only used loose snus in the same amount of time.
The snus sold on the Norwegian market in 2015 contained on average more nicotine per gram than the one sold in 2005. At the same time, there is an ever greater variation in portion sizes. This means that the degree of nicotine exposure depends on the type of snus used.
There are few tobacco users who smoke and use snus at the same time. Of those who had ever used snus and who were interviewed in the period 2016-2018, 33% of men and 40% of women had no previous experience with smoking. This is an increase from the period 2004-2006; then 23% of men and 12% of women who had used snus had not smoked before.
A review of the literature on the global burden of disease (GBD) shows that there is not yet sufficient evidence to estimate the burden of disease related to snus use according to GBD's criteria. The latest available version of GBD (GBD 2017) did not include any estimates of the burden of disease caused by use of snus.
Health effects of constituents of snus
Tobacco specific nitrosamines (TSNA)
The tobacco-specific nitrosamines N-Nitrosonornicotine (NNN) and 4-(N-Methyl-N-nitrosamino)-1-(3-pyridyl)-1-butanone (NNK) as well as the NNK degradation product 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL) are important and potent carcinogens in tobacco. A number of studies have unequivocally shown that laboratory animals develop cancer after exposure to NNN, NNK and NNAL. In many cell types, including cells in the liver, lungs and oral cavity, these nitrosamines can be converted into reactive products of activation or detoxification enzymes. These metabolic transformation products can bind to DNA and cause irreversible damage. This can be an early step in cancer development. Both animal and human cells are capable of forming such DNA binding products. This indicates that the mechanism of cancer development in humans is similar to that seen in laboratory animals. International Agency for Research on Cancer (IARC) classifies NNK and NNN as carcinogenic to humans. This is based on evidence from studies in cell lines and animals, and population studies which have found a correlation between the degree of exposure to NNK and NNN and cancer risk in smokers.
Nicotine is one of the most important active substances in tobacco and has considerable addictive potential. Nicotine acts primarily by stimulating nicotinic cholinergic receptors and secondly by increased release of various signaling molecules in and outside the central nervous system. The nicotinic cholinergic receptors activate multiple cellular signaling pathways involved in physiological functions. However, nicotine activation of these receptors will also mediate many of the undesirable effects of nicotine.
Nicotine and addiction: Nicotine is highly addictive and the substance that triggers addiction in tobacco use. Withdrawal symptoms when trying to quit smoking are the typical signs that nicotine use leads to physical dependence. The use of nicotine also leads to development of nicotine tolerance. In some individuals, nicotine addiction can develop very quickly. There are indications that differences in nicotine dependence in the population may be related to genetic predisposition.
Nicotine and cancer: There is insufficient data from animal studies to conclude if nicotine is a complete carcinogen or not. Cell studies have shown that nicotine activates a number of signaling systems in the cells that are relevant in terms of whether nicotine can promote cancer development (promoter effect) when such development is first initiated by known carcinogens. The experimental studies in animals indicate that nicotine may be a promoter, that is, nicotine may promote already initiated cancer development. Animal studies have shown that nicotine can reduce the effects of cancer treatments by growth factor receptor blockade, chemotherapy and radiation therapy.
Lung development: Intake of nicotine during pregnancy will also expose the unborn child to nicotine. Several animal studies have shown that nicotine exposure during pregnancy has harmful effects on lung development and lung function after birth.
Immune system: Nicotine has a variety of immunological effects of both stimulant and suppressive nature. The effects are likely to be affected by the amount and duration of exposure, route of exposure, organ systems involved in the immune response, stage and type of disease, and the degree of autoimmune and inflammatory mechanisms involved. Gender, age and disease manifestations may alter the effects of nicotine.
Cardiovascular effects: Nicotine exposure increases heart rate and blood pressure. Nicotine can reduce insulin sensitivity, which in turn can increase the risk of atherosclerosis. Overall, nicotine may contribute to an increased risk of cardiovascular disease both in smokers and snus users.
Cognitive functions and mental health: Animal studies and human studies have shown that nicotine can alter cognitive functions. This is especially true when exposed to pregnancy through maternal nicotine use, as well as during childhood and adolescence where the brain is developing and thus extra vulnerable. Hypotheses have been made that nicotine use may alleviate mental disorders by enhancing cognitive functions (the self-medication hypothesis), while other data indicate that nicotine may be a risk factor for mental illness since it affects some of the same biological mechanisms underlying such disease. There are also studies indicating that there are common genes that provide vulnerability to nicotine addiction and mental illness.
Experimental animal studies on snus
Snus cause changes in the oral mucosa (snus-induced lesion) in rats, as well as an increased number of cancers outside the oral cavity compared with those found in animals not exposed to snus. In animal studies with mice, exposure to snus or nicotine alone seems to affect the ability of the blood vessels to repair and to heal wounds. Exposure to snus in animal studies affects the immune system by reduced white blood cells and altered levels of various inflammatory markers. Such changes can affect the body's defence and repair when exposed to illnesses or harmful effects.
We have prepared systematic reviews of available research on health risks from use of Swedish snus. Most studies were included in the report on snus from the Norwegian Institute of Public Health in 2014. The new studies have strengthened the conclusions of the previous report that the use of snus may increase the risks of some serious adverse health outcomes and some less serious adverse health outcomes, as shown in the results section.
The carcinogenic properties of the constituents, as shown in animal studies and cell studies, support the results of the population studies.
Available studies, on which we have based our conclusions, have varying design and length of follow-up. All follow-up studies identified the tobacco habits at the start of the studies, that is whether the participants used or had used snus, whether they smoked or had smoked, whether they neither smoked nor used snus, or both used snus and smoked (alternating or contemporary).
We rarely know how long the individuals who answered that they were using snus, had used snus before they were included in the study. Neither do we know how many of those who used snus at the start of the study, still used snus at the time of follow-up. Some may have quit, some may have changed to higher / lower consumption, and some may have switched to smoking or started to smoke in addition to using snus. Those who, at the start of their studies, said that they did not use snus or other tobacco products may also have started later, and may have used snus (or smoked or both) for many years before the risk analysis was conducted.
For some health outcomes (cancer of the pancreas, colon and rectum) we have a follow-up time of up to 35 years, and it is possible that some would have answered differently to questions about their tobacco habits at the end of the study than they did at the start of the study. For several of the health risks that we have considered, it takes a long time before the disease has progressed so far that the diagnosis is made. For example, more than half of those diagnosed with pancreatic, colon and rectal cancer are over 70 years of age. With increasing age, there is a greater chance of dying from other causes, making any risk associated with snus use more difficult to estimate in older age groups. Extended follow-up time will therefore have both strengths and weaknesses. We must also remember that people who choose to use snus may be different from those who choose not to use snus in many other respects. We do not have a complete overview of the factors that may be involved, and thus we do not have an overview of whether and how these factors may influence the same health outcomes.
There has been an increase in the use of Swedish snus in Norway in recent decades, especially among young adults and especially among young women. The majority of snus users are former smokers, but in the past 15 years there has been an increase in the proportion of snus users without prior smoking experience. In the period 2016-2018, 33% of men and 40% of women had not smoked before starting snus use. This is an increase from the period 2004-2006, when 23% of men and 12% of women who had used snus had never smoked.
Studies in cells and animals have shown that both nicotine and tobacco-specific nitrosamines have a variety of effects that may have undesirable health consequences of varying severity.
Research on the health risks associated with use of Swedish snus in humans mainly comes from observational studies. Even with the inherent limitations of such studies, we conclude that systematically summarized research shows that the use of Swedish snus:
- probably increases the risk of cancer of the esophagus and pancreas, and possibly increases the risk of cancer of the stomach and rectum
- possibly increases mortality after a cancer diagnosis (all types of cancer combined), both when the cause of death is considered cancer-related and for all causes
- probably increases the risk of high blood pressure
- probably increases the risk of dying the weeks after a heart attack or stroke, and may increase the long-term risk of dying after stroke
- may halve the risk of dying among those who stop using snus after a heart attack.
- may increase the risk of non-affective psychosis, weight gain and obesity
- high consumption of Swedish snus probably increases the risk of type 2 diabetes and metabolic syndrome
Women who use Swedish snus during pregnancy have:
- probably increased risk of premature births
- possible increased risk of stillbirths, being small for gestational age, reduced birth weight, cesarean section, short-term respiratory arrest and lip/palate defects, and levels of nicotine degradation products (cotinine) in the child's urine.
One study found that use of Swedish snus reduced the risk of getting Parkinson's disease. For many other health outcomes, there was too little information to conclude whether snus affects the risk of adverse health outcomes or in which direction.
We have described the prevalence of the use of snus in Norway. Furthermore, we have presented the effects (and possible mechanisms for these) of the important constituents nicotine and carcinogenic tobacco-specific nitrosamines as well as experimental animal studies of Swedish snus.
We have conducted systematic reviews of the health risks from using Swedish snus compared to not using tobacco products. Systematic literature searches have identified new studies in addition to the studies included in the National Institute of Public Health's report on snus from 2014. We have described the studies, and evaluated and presented the studies' internal validity (quality / risk of bias). If it had been possible we would have performed meta-analyses.
Since snus is known to contain carcinogens, it is unethical to conduct randomized controlled long-term trials. Therefore, observational studies, and preferably prospective cohort studies, are the best available studies we have on health risks associated with the use of snus. When we summarize health risks in accordance with international standards for systematic reviews, we grade our confidence in the effect estimates using the GRADE approach.
According to this approach, observational studies start and rarely achieve higher ratings than "low confidence" in the effect estimate. Therefore, most of the results on health risks in this report - which are from observational studies - are referred to by statements indicating that we have low confidence in the effect estimates (according to GRADE), and conclusions are similarly formulated with the words "It is possible that the use of snus leads to.." That does NOT mean that the cohort studies that have assessed the health risks of using Swedish snus are less trustworthy than cohort studies in general. For example, information we have about risks from smoking and risks from air pollution also comes from observational studies.
Some of the risk estimates show large effects and some other results have established a dose-response relationship (which means that a larger dose produces a larger response). In those cases, we have upgraded to "moderate confidence" in the effect estimate, with conclusions formulated as "it is likely that the use of snus increases the risk of …."
Many of the risk estimates for the use of Swedish snus have broad confidence intervals, including the possibilities of both a considerable reduction in risk and a considerable increase in risk. Then we are uncertain whether the use of snus will affect the outcome in question and if so in which direction, and we downgrade to very low confidence in the effect estimate. It is very important to distinguish between cases when we have such broad confidence intervals that we do not know whether or in which direction the outcome is affected, and when narrow confidence intervals indicate little or no effect. The final conclusion builds on the statistical results, as the assessment of whether something is important also depends on the outcome that is considered (for example, there is a difference between feeling nauseous and developing cancer). How many who are affected by an increased risk depends on the background risk in the population. We have explained our assessments in the relevant chapters. Here is a simplified table with the statements we have used.
GRADE – grading of our confidence in the effect estimates
RCTs that are conducted well, or well conducted observational studies which are upgraded twice
Exposure leads to…
RCTs which are downgraded; observational studies which are upgraded
It is likely /probable that the exposure leads to../ The exposure probably leads to…
Observational studies, or RCTS which are downgraded twice
It is possible that the exposure leads to../ The exposure possibly leads to.. may lead to..
Most estimates from observational studies will be in this category.
Observational studies which are downgraded or RCTs which are downgraded three times
We are uncertain if../ We do not know if the exposure affects…
RCT: randomised controlled trials
Criteria for downgrading (study limitations, inconsistent results across the studies, directness, precision, reporting bias)
Criteria for upgrading (large or very large effects, dose-response, assessment of confounders)
We have also given an account of the status of previous attempts to estimate the burden of disease from use of snus through the international project Global Burden of Disease (GBD).
Norway is currently the only country next to Sweden within the EU / EEA area where snus is allowed to be sold. At the same time that we have seen a decline in tobacco smoking in Norway, there has also been an increase in snus use. Chapter 15 in this report discusses what kind of data we need to assess the overall health consequences of increased snus use among non-smokers and among people who initially smoked.