Some 72-92% of adult cigarette smokers meet the criteria for dependence1. While nicotine is recognised as an addictive substance in the tobacco leaf, the risk of addiction to pure nicotine products is very low compared to cigarettes . Currently, it is being discussed in the public health community whether lowering the levels of nicotine in tobacco products would make people less addicted and accordingly reduce the consumption of tobacco products.
Tobacco additives were hardly used before 1970, but today they represent up to 10% of the cigarette weight. By altering the taste and smell of cigarettes the products are made more attractive and the smoke more palatable which leads to an increase of smoking initiation. At present, the role of additives in enhancing the addictiveness of tobacco products is not clear.
In order to make tobacco products more attractive, design features are introduced, e.g. package design and cigarette form. In addition, these features are used to undermine the effect of the maximum limits set by the Tobacco Products Directive 2001/37/EC on tar, nicotine, and carbon monoxide (CO) yields in cigarettes.
Article 13 of the Tobacco Products Directive (2001/37/EC)2 stipulates that Member States can keep or introduce, in accordance with the Treaty, more stringent rules concerning the manufacture, import, sale, and consumption of tobacco products which they deem necessary in order to protect public health. Member States may prohibit the use of ingredients which have the effect of increasing the addictive properties of tobacco products.
Article 12 of the Tobacco Products Directive invites the Commission to submit a proposal providing a common list of ingredients authorised for tobacco products, taking into account, inter alia, their addictiveness.
In its comments to the Green Paper Towards a Europe free from tobacco smoke: policy options at EU level3, the European Parliament invited the Commission to propose, by 2008 if possible, an amendment to the Directive including an evaluation and authorisation procedure for tobacco additives and an immediate ban on all additives that are addiction-enhancing4. In its 2nd Report on the implementation of the Tobacco Products Directive5 the Commission stresses the need for further work on the addictiveness of tobacco additives.
DG SANCO wishes to have a better understanding of the criteria based on which an additive can be considered (classified) as an addictive and/or attractive substance, the role of additives in tobacco products and the role of design features in the attractiveness and addictiveness of a tobacco product.
According to a report from WHO (2008), about 100 million people died in the 20th century from tobacco use. The number of deaths in 2007 due to tobacco related diseases was about 5.4 million and if current smoking patterns continue, more than 8 million deaths are expected to occur each year due to tobacco smoking by the year 2030. In the EU, about a third of the adult population are smokers. The number of deaths from smoking per year is currently about 500,000 in the EU and more than 1.5 million in the whole European region (WHO 2007a).
The vast majority of smokers use cigarettes, while other ways of smoking are less frequent (e.g. cigars, pipes, waterpipes). Apart from smoking tobacco, other tobacco forms (i.e. smokeless tobacco) may also have deleterious public health effects (SCENIHR 2008). In addition, exposure to tobacco smoke in the environment, so-called “passive smoking” or “second-hand smoking” is an important cause of excess mortality and morbidity. Passive smokers have a significantly increased risk for several diseases such as lung cancer (IARC 2004), respiratory diseases (Jaakkola and Jaakkola 2002a, Jaakkola and Jaakkola 2002b) and cardiovascular diseases (Whincup et al. 2004).
Table 1 shows that tobacco has a very high addictive potential in humans (Anthony et al. 1994, O’Brien and Gardner 2005) whereas numerous studies indicate that the risk of addiction to pure nicotine is low.
|Ever used a (%)||Dependence b (%)||Risk of addiction c (%)|
Table 1 Risk of addiction (adapted from Anthony et al. 1994, O’Brien and Gardner 2005)
a Column 1: Ever used – Prevalence of ever use in the total population
b Column 2: Dependence – Prevalence of dependence in the total population
c Column 3: Risk of addiction - Ratio of column 2 (dependence) and column 1 (ever used)
The addictiveness of nicotine is enforced by substances in tobacco leaves that inhibit the action of monoamine oxidase (MAO) in the body (Berlin and Anthenelli 2001). Apart from naturally occurring substances in tobacco leaves, a number of ingredients in the final product may create or increase dependence. The tobacco industry has admitted the use of 599 different cigarette additives in the United States (US), which are claimed to improve taste and reduce harshness of the smoke (Rabinoff et al. 2007). Current US-style cigarettes contain about 10% of additives by weight; mainly sugars, humectants, cocoa and liquorice. Most other additives are used in small amounts. As discussed later in this opinion, cigars, pipe tobacco and smokeless tobacco generally contain fewer additives than cigarettes. Tobacco used in waterpipes is characterised by a high content of water and various sugars.Certain flavours (e.g. candy and fruit) have been used largely to make tobacco products more appealing to children (called “young adults” by the tobacco industry). In order to decrease the appeal of cigarettes to children, the US Food and Drug Administration (FDA)6 banned the use of a number of flavours as additives in cigarettes in September 2009. Menthol is not one of the banned additives, but is currently being evaluated by the Tobacco Products Scientific Advisory Committee of the FDA. In other parts of the world (e.g. Canada, Australia, New Zealand), legal measures on additives are established or are in preparation. In Europe, some countries, such as Germany, United Kingdom, Austria, Romania and France, use positive and/or negative lists which respectively allow or prohibit the use of specific compounds as tobacco additives, whereas other countries do not have such a regulation.
It is the purpose of the present opinion to examine the criteria for classifying tobacco additives as addictive or attractive, and to evaluate their role for the creation or maintenance of dependence on tobacco products. This would serve as the scientific basis for regulation of the use of additives in order to reduce the toxicity and the addictiveness of the final tobacco product. An important question is whether some additives are addictive by themselves or if they act by increasing the addictiveness of nicotine. The different methods of assessing addictiveness of an additive, alone or in combination with other substances, will be reviewed. In addition to the interactions between additives and constituents of tobacco, the burning of tobacco creates other complex chemical substances that may be toxic or favour addiction. An example of this is aldehydes, such as acetaldehyde, formed by the pyrolysis of various sugars and polysaccharides in the tobacco (see section 220.127.116.11). The technical characteristics of tobacco products, in particular of cigarettes, may also influence their addictive potential. A number of additives favour attractiveness of tobacco products, and may thus promote smoking initiation. In this context special attention will be paid to how additives may be used to target specific groups.
A public call for information7 was launched in November 2009, giving all stakeholders the opportunity to submit relevant scientific information concerning tobacco additives. The information asked for concerned: 1) details about the manufacturing process of tobacco products; and 2) methods applicable for assessment of attractiveness. A number of organisations and major tobacco companies responded. The information received has been evaluated carefully and was in many cases useful for writing the opinion. A particular problem in the area of tobacco products is that a number of studies relevant for this opinion have never been published but exist as internal documents of the tobacco industry. Some of the documents contain sensitive information showing health risks associated with smoking. In 1992, 60 documents were destroyed by Imperial Tobacco Canada in order to avoid exposure of the company to liability or embarrassment. Hammond et al. (2009a) have recently reviewed the contents of these documents that were recovered at the British American Tobacco headquarters in the United Kingdom and were released in 1998 through court disclosure in a trial in Minnesota. The author concludes that most of the studies that were carried out by researchers employed by the industry were scientifically valid. They gave evidence that cigarette smoke was carcinogenic and addictive. Since then, a great number of industry documents have become publicly available and can be found in two searchable databases, http://tobaccodocuments.org and http://legacy.library.ucsf.edu. The collections continue to be updated and currently contain more than 60 million pages in over 11 million documents.
Furthermore, a tobacco documents bibliography is also available which includes papers and publications based on documented research, broadly classified into several groups. Some examples of publications based on research of industry documents appearing under the heading of “Ingredients and Design” illustrate the tobacco industry research and development strategy on issues including: smoker preferences (Chaiton et al. 2005); smoking behaviour and product design (Hammond et al. 2006); targeting consumer groups with specific psychological needs (Cook et al. 2003); research on nicotine (Hurt and Robertson 1998); addictiveness (Scharfstein 1999, Slade et al. 1995, Stevenson and Proctor 2008, Vagg and Chapman 2005); manipulation/free base nicotine (Wayne et al. 2006, Wayne and Carpenter 2009); flavoured cigarettes (Lewis and Wackowski 2006); menthol (Kreslake et al. 2008a, Wayne and Connolly 2004); youth targeting (Wayne and Connolly 2002); and particle size (Wayne et al. 2008a). Relevant publications are discussed in subsequent chapters of this opinion.
For the purpose of the present opinion, the health risks of tobacco products and additives have been investigated within different lines of evidence such as epidemiological studies, experimental studies in humans, experimental studies in animals, cell culture studies and in silico studies. In order to answer the questions in the Terms of Reference to this opinion, a weighted approach has been used, where data from all the available lines of evidence were integrated as appropriate. A more detailed description of how such weighting is performed is given in an earlier opinion of the SCENIHR (SCENIHR 2009). The primary sources for this opinion were original scientific reports published in peer-reviewed scientific journals. The secondary sources used were the stakeholder information mentioned above and reports and opinions of other Scientific Committees, as well as reports of various governmental bodies. In addition to the reports cited in the text and included in the list of references, various publications were noted but not considered appropriate for the purposes of developing the opinion.
Contents, ingredients, and additives
According to the terminology used in the WHO Framework Convention and the recommendation by the Scientific Advisory Committee in 2003, the term “contents” is used synonymously with the term “ingredients”. Consequently, it means all product components, the materials used to manufacture those components, residual substances from agricultural practices, storage and processing, substances that can migrate from packaging into the product, as well as what may be termed “additives” and “processing aids” in some countries and regions (WHO 2007b).
Based on the 2nd Report on the Application of the Tobacco Products Directive (EC 2007b), the current definition of “ingredients” in Article 2 (5) covers any substance or constituent used in the manufacture or preparation of a tobacco product and still present in the finished product even if in an altered form, including paper, filter, inks and adhesives. It does not cover the tobacco leaf itself or other natural or unprocessed tobacco plant parts.
For the purpose of this report, we consider that the WHO definition is the most useful, as some of the added ingredients (e.g. different forms of sugar) are already present in the tobacco leaves. Tobacco leaves may also in some cases contain various toxic substances such as cadmium or radioactive isotopes. The possible presence of residual substances from agricultural practices will not be addressed in this report. In order to avoid misunderstandings, the present report uses the term additives for added ingredients or substances. Additives are defined as any substance that is added, except water, during the course of manufacture of a tobacco product, including preservatives, humectants, flavours, and processing aids.
Natural or clean cigarettes are being marketed as having no chemicals or additives and the filters are made from natural cellulose. However, smoke from these cigarettes still contains all the carcinogens and toxins that come from the tobacco itself (Malson et al. 2002, McDaniel and Malone 2007).
Herbal cigarettes, although they may not contain tobacco, yield tar and carbon monoxide when smoked, and are thus also dangerous to health (Chen et al. 2007a, Gan et al. 2009).
Addiction is the commonly used term referring to what is technically known as “dependence” and is widely employed to connote severe substance dependence, as has been demonstrated to occur in tobacco users. Dependence has been defined by the WHO Expert Committee on Drug Dependence (WHO 2003) and The ICD-10 Classification of Mental and Behavioural Disorders: Clinical Descriptions and Diagnostic Guidelines (WHO 1992).
Addictiveness refers to the pharmacological potential of a substance to cause addiction. Abuse liability of a drug is the likelihood that its use will result in addiction (dependence) and it can be assessed in laboratories by methods referred to as abuse liability testing (Schuster and Henningfield 2003, Wayne and Henningfeld 2008b, WHO 2003).
The terms “dependence-causing” and “dependence potential” have been used as synonyms for “addictive” and “addictiveness”, respectively. In addition to the neurobiological characteristics of the substance itself, dependence potential is related to the dose, speed of absorption, metabolism, and to physical and chemical features of the formulation (WHO 2007b).
According to the WHO, the terms “attractiveness” or “consumer appeal” refer to factors such as taste, smell and other sensory attributes, ease of use, flexibility of the dosing system, cost, reputation or image, assumed risks and benefits, and other characteristics of a product designed to stimulate use (WHO 2007b). Physical product characteristics are often integrated with marketing (WHO 2007b). For example, a flavour such as “menthol”, “mint”, or “cherry”, which is intended to appeal to a target population, may be incorporated into the product name or descriptors and marketed to reach out to that population (WHO 2007b). Attractiveness is also related to nicotine dosing characteristics, which is why smokeless tobacco product companies may include products ranging from lower dosing and slower onsetting “starter” products to higher dose maintenance products (FDA 1995, FDA 1996).
Although the risk of dependence on any substance is partially related to the attractiveness and/or ease of use of the delivery system, these features are not typically evaluated in dependence-potential testing but rather are generally described as factors affecting “consumer appeal” or “attractiveness”. Addictiveness and attractiveness go hand in hand as the real world liability for abuse of and addiction to a tobacco product is to a large extent also related to the attractiveness of the tobacco product.
Attractiveness is powerfully determined by imagery and cultural associations that are cultivated by the tobacco industry and effects may therefore be indirect. Attractiveness is also influenced by product sensory characteristics using flavours, and product characteristics (as well as marketing) that are intended to reduce concerns or undesirable features (e.g. reduce concerns about cancer with “light” branding, and reduce noxious throat burn with various chemicals and “smoke smoothers”) (Wayne and Henningfield 2008b).
1 Henningfield JE, Zeller M. Could science-based regulation make tobacco products less addictive? Yale J Health Policy Law Ethics 2002; 3:127-38.
3 http://ec.europa.eu/health/ph_determinants/life_style/Tobacco/Documents/gp_smoke_en.pdf plus report on consultation: http://ec.europa.eu/health/ph_determinants/life_style/Tobacco/Documents/smoke_free_frep_en.pdf