NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

IARC Working Group on the Evaluation of Carcinogenic Risk to Humans. Smokeless Tobacco and Some Tobacco-specific N-Nitrosamines. Lyon (FR): International Agency for Research on Cancer; 2007. (IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, No. 89.)

Cover of Smokeless Tobacco and Some Tobacco-specific N-Nitrosamines

Smokeless Tobacco and Some Tobacco-specific N-Nitrosamines.

Show details

1Description of Smokeless Tobacco Practices

1.1. Historical overview

The tobacco plant is thought to have originated on the mainland between North and South America. Its cultivation probably dates back at least 5000 years; tobacco seeds were discovered during archaeological excavations in both Mexico and Peru around 3500 BC, which shows that tobacco was an article of value to the inhabitants (Voges, 1984).

American Indians were probably the first people to smoke, chew and snuff tobacco, as early as the 1400s (Christen et al., 1982). The Indians inhaled powdered tobacco through a hollow Y-shaped piece of cane or pipe by placing the forked ends into each nostril and the other end near the powdered tobacco. This instrument was called a ‘tobago’or ‘tobaca’. The word was later changed by the Spaniards to ‘tobacco’ (Christen et al., 1982).

1.1.1. Tobacco chewing

In 1499, Amerigo Vespucci found Indians on Margarita Island, off the coast of Venezuela, who chewed a green herb known as tobacco in order to quench their thirst, since it produced an increase in salivation; he also reported that the Indians chewed tobacco leaves to whiten their teeth and to alleviate hunger (Heimann, 1960; Stewart, 1967; Voges, 1984).

The practice of tobacco chewing was widespread in parts of Central and South America in the late 1500s (Voges, 1984). Columbus, in 1571, observed men in Veragua, later known as Costa Rica, who put a dry herb in their mouths and chewed it (Heimann, 1960). Use of plug tobacco was reported in Santo Domingo during the sixteenth century. Tobacco chewing seems to have been a common practice among the American Indians, especially when long distances had to be covered; it has been reported that tobacco was the support against hunger, thirst and fatigue when an American Indian would trek for 2 or 3 days with no other support. Several American tribes mixed either lime or finely-powdered and burned, fresh-or saltwater molluscs with their chewing tobacco (Curtis, 1935).

Among native Americans, chewing tobacco was thought to have several medicinal uses, such as to alleviate toothache, to disinfect cuts by spitting the tobacco juice and saliva mixture onto the wound, and to relieve the effects of snake, spider and insect bites (Axton, 1975).

By 1531, the Spaniards were growing tobacco commercially in the West Indies and maintained a monopoly over the European markets until 1575, at which time the Portuguese began to grow large quantities of the commodity. Tobacco was soon grown in Europe as both a decorative and medicinal plant. In 1559, Jean Nicot, in whose honour the genus Nicotiana was named, was ambassador to Sebastian, King of Portugal. He grew tobacco and promoted the product in Europe for its magic ‘cure-all’ properties. By the early seventeenth century, tobacco had become one of the major exports of the American colonies (Christen et al., 1982) and its use in various forms had spread throughout Europe, Turkey, Russia, Arabia, China, Alaska and the world (Axton, 1975). Portugese and Spanish sailing crews who were addicted to tobacco carried seeds, and planted them at ports.

When smoking was forbidden on British naval vessels because of the fire hazard, sailors turned to chewing tobacco and snuff. In Europe, tobacco was regarded as a prophylactic during the plague and, for those who did not like smoking, chewing was an alternative. Tobacco chewing was recommended for cleaning the teeth of women and children (Brooks, 1952). Chewing tobacco became popular in the USA only during the first half of the nineteenth century (Gottsegen, 1940). In spite of two centuries of pipe smoking and snuff use, by the mid-1850s, North Americans rejected the European practices in general, and British practices in particular, that entailed snuff boxes and formality; in addition, tobacco chewing was more convenient for Americans who trekked westward in their wagons. During the 1860s, tobacco was chewed in the form of either a plug or a twist. Of the 348 tobacco factories listed in the 1860 Census for Virginia and North Carolina, only seven manufactured smoking products (Heimann, 1960). American pioneers resorted to the use of a home-made sweet plug, so-named because the leaf was wadded into a hole in a log and laced with a sweetening agent (usually brandy or cane sugar), which, after removal of the fermented leaf, resulted in a tasty chew (Axton, 1975).

In 1797, Adam Clarke, a famous Methodist minister, appealed to all tobacco consumers and religious followers to avoid the use of tobacco for the sake of their health and their souls. This plea was also due to the fact that it had become unsafe to kneel when praying because chewers had made the floors unsanitary (Brooks, 1952).

During the latter part of the nineteenth century, the ‘germ theory of infection’ changed the course of chewing in America, and it was felt that expectorating on the floor and into a brass cuspidor could be a source of contamination and the spread of disease. By the 1890s, public outcry made tobacco chewing socially unacceptable behaviour and unlawful in most public places (Christen et al., 1982). Anti-spitting laws were passed in New York and Philadelphia, USA, in 1896 and in Toronto, Canada, in 1904 (Kozlowski, 1981).

The market for chewing tobacco passed its peak in 1890, when some 3 lb (about 1.5 kg) of plug, twist or fine-cut chewing tobacco were chewed annually per capita in the USA (Heimann, 1960). Nevertheless, chewing remained the dominant form of tobacco use in America until the expansion of the cigarette industry in 1918 (Maxwell, 1980). In 1945, cuspidors were removed from all federal buildings by order of the US District Court in Washington DC (Brooks, 1952). The apparent decline in tobacco chewing is exemplified by a memorandum of 14 September 1955 to the American Tobacco Company, stating, “It has become impossible to hire persons in the New York area to clean and maintain cuspidors • it will be necessary to remove them promptly from the premises” (Heimann, 1960). During the second half of the 1960s through to the 1970s, however, a resurgence in tobacco chewing occurred in the USA (Christen & Glover, 1981).

1.1.2. Snuff taking

The native populations of Brazil were the first people known to use snuff. Using a cup and a pestle made from rosewood, the tobacco leaves were ground into a powder and acquired the delicate aroma of the wood. The resulting snuff was placed in ornately decorated bone tubes, one end of which was plugged to preserve the fragrance (Curtis, 1935). The American Indians inhaled powdered tobacco through a hollow Y-shaped piece of cane or pipe by placing the forked ends into each nostril and the other end near the powdered tobacco (Christen et al., 1982).

Friar Ramón Pané, a Franciscan monk who travelled with Christopher Columbus on his second voyage to the New World in 1493, reported that the Caribbean Indians of the lesser Antilles used snuff (Christen et al., 1982). In Haiti, snuff powder was used by medicine men for clearing nasal passages and as an analgesic (Stewart, 1967). Friar Pané's return to Spain with snuff signalled the arrival in Europe of a practice that was to last for several centuries.

In 1519, Ocaranza found that Mexican Indians used tobacco powder to heal burns and wounds and, in 1525, Herrera observed that Mexican Indians held tobacco powder in their mouth to send them to sleep and reduce pain (Stewart, 1967).

The Dutch, who named the powdered tobacco ‘snuff’, were using the product by 1560 (Christen et al., 1982). By the early 1600s, snuff had become an expensive commodity and its use had spread throughout South America, China, Japan and Africa. The origin of the process terms ‘carotte’ and ‘rappee’ goes back to the 1600s when tobacco for snuff was prepared in the form of a carrot to be rasped in the quantity desired for use (Curtis, 1935). In 1620, the Royal Snuff Factory was established in Seville, and this became the centre of the manufacture and development of this product (Voges, 1984). Snuff use expanded through Japan to China (Ching Dynasty) in the 1650s: palace artisans produced exquisitely carved, inlaid enamelled or painted snuff bottles with a tiny spoon attached to the bottle stopper; a small portion of snuff was placed on the left thumbnail and inhaled through the nose. The Chinese believed that snuff cured pains in the eyes and teeth, alleviated throat ailments, constipation and cold symptoms, and promoted sweating (Christen et al., 1982).

By 1650, snuff use had also spread from France to England, Scotland and Ireland. The Irish called snuff ‘powder’ or ‘smutchin’; the Scots called it ‘sneeshin’ (Harrison, 1964). Jean Nicot is credited with introducing snuff to Catherine de Medici, Queen of France, to cure her headaches (Christen et al., 1982).

Snuff use reached a peak in England during the reign of Queen Anne (1702–14), and was called the ‘final reason for the human nose’. It was at this time that ready-made snuff became available in England. It continued to be popular during the reign of George III, and his wife, Charlotte (1760–1820), referred to as ‘Snuffy Charlotte’, had an entire room in Windsor Castle devoted to her snuff stock. Lord Nelson, the Duke of Wellington, Marie Antoinette, Disraeli, Alexander Pope and Samuel Johnson all used snuff (Harrison, 1964). In diplomatic intrigue, poisons were sometimes placed in snuff. The aristocratic popularity of snuff led to a minor art form, in that snuff boxes became symbols that reflected the wealth and rank of their owner. The dandy, Lord Petersham, was said to own an annual set of 365 snuff boxes (Christen et al., 1982).

The leading snuff supplier of the time provided King George IV with his own special blends, King's Morning Mix, King's Plain and King's Carotte (Ryan, 1980). Home-made snuff was common. The tightly-rolled tobacco leaves (carotte) were often soaked in cinnamon, lavender or almond oils; tobacco was dried and ground by means of an iron hand-grater that resembled a modern cheese-grater. The proper manner of inhaling snuff was to place a small quantity on the back of the hand and sniff it up the nostrils to induce a sneeze (Christen et al., 1982).

Although hundreds of varieties of snuff existed in Europe by the 1800s, these consisted of three basic types: Scotch snuff, which was a dry, strong, unflavoured and finely ground powder; Maccaboy, a moist and highly scented snuff; and Rapee, also known as Swedish snuff, a coarsely grated snuff (Heimann, 1960).

Snuff was introduced into Sweden in the middle of the seventeenth century, but its popularity among aristocrats reached a height during the eighteenth century, when use of nasal snuff became the highest fashion at the court of King Gustav III, among both men and women. The practice subsequently spread to the general Swedish population.

In many Swedish cities, snuff has been manufactured since the beginning of the eighteenth century. In Gothenburg, which is considered to be the centre of snuff produc-tion, manufacture started in about 1650 (Loewe, 1981). In 1795, Samuel Fiedler established a snuff mill in Gothenburg and began a small business, which later developed into three separate companies. At the end of the nineteenth century, the leading producer was Jacob Ljunglöf in Stockholm; his leading brand ‘Ettan’ became well known throughout Europe (Loewe, 1981). In 1914, the production of snuff in Sweden was taken over by the Swedish tobacco monopoly, which restored Gothenburg as its centre. A large factory was built around 1920, and expanded in 1979, for the production of snuff and chewing tobacco.

Since the beginning of the twentieth century, snuff has been used mainly orally in Sweden. In the 1950s and 1960s, use of moist snuff was prevalent predominantly among older men and was heading towards a ‘natural death‘: the median age of consumers in 1969–70 was over 40 years (Nordgren & Ramström, 1990). However, the development of new products and intensive advertisement and promotion by Swedish Match, the country's primary snuff manufacturer, starting in the late 1960s, led to a surge in the use of moist snuff among young men. By 1972–73, the median age of moist snuff users had dropped to 30 years (Nordgren & Ramström, 1990). More recently, Swedish Match has been representing its moist snuff products as less harmful tobacco products than cigarettes (Henningfield & Fagerström, 2001). As discussed later in this section, the prevalence of smokeless tobacco use continues to increase in Sweden, particularly among young men.

Commercially manufactured snuff made its way to North America in 1611 by way of John Rolfe, husband of Pocahontas. Rolfe introduced the better Spanish variety of tobacco to ensure the survival of the Jamestown Colony in Virginia. Although most of the colonists in America never fully accepted the English style of snuff use, American aristocrats used snuff, and Dolly Madison was known to distribute samples of snuff to White House guests. During the 1800s until the mid 1930s, a communal snuff box was installed for members of the US Congress. The colonists also found it more to their taste to place snuff in their mouths rather than to sniff it (Christen et al., 1982).

The first snuff mills in America were constructed in Virginia in about 1730 (Heimann, 1960). The snuff was made from New England tobacco and its quality was said to equal that of the native Scottish varieties (Robert, 1949). Pierre Lorillard, a Huguenot, esta-blished a snuff mill in New York in 1760 and carefully guarded the secret of the ingredients and blends of his products (Christen et al., 1982).

Between 1880 and 1930, the production of snuff in the USA increased from 4 million lb (1.8 million kg) to more than 40 million lb (18 million kg) per year (Garner, 1951). By 1945, the American Snuff Company in Memphis, TN, claimed to be the largest snuff manufacturer in the world (Christen et al., 1982). Snuff was made predominantly from dark, air-and fire-cured leaves. Stems and leaves were aged in hogsheads and conditioned before being cut into strips of 1–2 in (2.5–5 cm) in width. The chopped leaves underwent further fermen-tation for about 2 months, during which time the tobacco lost its creosote-like odour and became more aromatic. It was next dried by passing it through steam-heated containers and then ground to a fine powder in a revolving steel drum. The powder was passed over silk cloth that contained as many as 96 threads per in (38 per cm). The coarse residue was returned to the mill for additional grinding before being packed into 100-lb (45-kg) bags for storage prior to repacking in smaller containers for retail sale. The dry and moist snuffs were used for dipping and placing in the mouth. Rappee or French snuff was used for inhaling, and Maccaboy snuff was both sucked and inhaled (Garner, 1951).

The use of smokeless tobacco products in the USA was widespread throughout the nineteenth century. Dental snuff was advertised to relieve toothache; to cure neuralgia, bleeding gums and scurvy; and to preserve and whiten teeth and prevent decay (Christen et al., 1982). With the advent of anti-spitting laws, loss of social acceptability and increased popularity of cigarette smoking, its use declined rapidly during the twentieth century.

Beginning in the mid-1970s, the US Tobacco Company (later renamed the US Smokeless Tobacco Co.), the leading manufacturer of smokeless tobacco products in the USA, developed new products, new images and an aggressive marketing campaign to expand its market (Connolly et al., 1986; Connolly, 1995). The marketing campaign included a ‘graduation’ marketing strategy that was designed to recruit new, young users with low-dose nicotine ‘starter’ moist snuff products and move them to higher-dosage products as they developed tolerance and addiction to nicotine (Connolly, 1995). The result was a ninefold increase in the prevalence of snuff use among young adult men (< 24 years old) between 1970 and 1987 (Giovino et al., 1994; Giovino, 1999). The United States Smokeless Tobacco Company continues to market its products for young men (Myers, 2003) and, in recent years, has also been marketing products for smokers as an alternative tobacco product, particularly for use when faced with smoking restrictions (Henningfield et al., 2002).

Tobacco was introduced into South Asia in the 1600s as a product to be smoked and was gradually used in many different forms (Bhonsle et al., 1992; Gupta & Ray, 2003). The chewing of betel quid (pan) was a popular practice that existed for over 2000 years and extended eastwards as far as the South Pacific Islands. After its introduction, tobacco soon became a new ingredient in betel quid, which has become the most commonly used form of smokeless tobacco in South Asia (Gupta & Ray, 2003; IARC, 2004a).

In Sudan, the introduction of toombak is historically attributed to a Koranic (Islamic) teacher, who came from Egypt, Timbuktu in Mali, Morocco, Turkey or Arabia, and dates back several centuries (Idris et al., 1998a). Another popular name for toombak is sute, which means ‘sniffing of the product’ in the local language, and indicates nasal usage when it was first introduced.

1.1.3. Attitudes and beliefs regarding smokeless tobacco use

The use of tobacco, including smokeless tobacco, has been controversial since its introduction. Therefore, a history of smokeless tobacco use is not complete without a discussion of the attacks on tobacco by various groups. In 1590 in Japan, tobacco was prohibited, and users lost their property or were jailed. James VI of Scotland, who became King James I of England and Ireland in 1603, was a strong anti-smoking advocate and increased taxes on tobacco by 4000% in an attempt to reduce the quantity imported into England. In 1633, the Sultan Murad IV of Turkey made any use of tobacco a capital offence, punishable by death from hanging, beheading or starvation, and maintained that tobacco caused infertility and reduced the fighting capabilities of his soldiers. The Russian Czar Michael Fedorovich, the first Romanov (1613–45), prohibited the sale of tobacco, and stated that users would be subject to physical punishment; persistent users would be killed. A Chinese law in 1638 threatened that anyone who possessed tobacco would be beheaded (Christen et al., 1982).

During the mid 1600s, Pope Urban VIII banned the use of snuff in churches, and Pope Innocent X attacked its use by priests in the Catholic Church. Other religious groups banned snuff use: John Wesley (1703–91), the founder of Methodism, attacked its use in Ireland; similarly, the Mormons, Seventh-Day Adventists, Parsees and Sikhs of India, Buddhist monks of Korea, members of the Tsai Li sect of China, and some Ethiopian Christian sects forbade the use of tobacco (Christen et al., 1982).

In Bavaria, Germany, in 1652, tobacco was available only on a doctor's prescription; Frederick the Great, King of Prussia, prevented his mother, the Dowager Queen of Prussia, from using snuff at his coronation in 1790. Louis XV, ruler of France from 1723 to 1774, banned the use of snuff from the Court of France (Christen et al., 1982).

In 1761, John Hill, a London physician and botanist, concluded that nasal cancer could develop as a consequence of snuff use. He reported five cases of ‘polypusses, a swelling in the nostril that was hard, black and adherent with the symptoms of an open cancer’ (Redmond, 1970).

1.2. Manufacture and use of smokeless tobacco products

Smokeless tobacco is consumed without burning the product, and can be used orally or nasally. Oral smokeless tobacco products are placed in the mouth, cheek or lip and sucked (dipped) or chewed. Tobacco pastes or powders are used in a similar manner and applied to the gums or teeth. Fine tobacco mixtures are usually inhaled and absorbed in the nasal passages. Table 1 lists smokeless tobacco products according to their mode of use.

Table 1. Classification of smokeless tobacco products by mode of use.

Table 1

Classification of smokeless tobacco products by mode of use.

Smokeless tobacco products are used throughout the world (National Cancer Institute/Centers for Disease Control, 2002; Gupta & Ray, 2003). Table 2 presents an overview of their use by WHO region. It is worth noting that some products are known to be used by immigrants from certain regions where a product is used to other regions.

Table 2. Use of smokeless tobacco products by WHO region.

Table 2

Use of smokeless tobacco products by WHO region.

There are many different botanical classifications for tobacco plants. The genus Nicotiana is classified into three main subgenera, N. rustica, N. tabacum and N. petuniodes. Smokeless tobacco products use N. tabacum, and sometimes N. rustica. In the USA, tobacco is also classified by the curing method (e.g. flue-cured, air-cured, dry air-cured tobacco) and by production areas (Virginia, North Carolina, Tennessee, Wisconsin) (Tso, 1990).

1.2.1. Oral use

Oral use of smokeless tobacco is practised in Africa, North America, South-East Asia, Europe and the Middle East, and consists of placing a piece of tobacco or tobacco product in the mandibular groove and either chewing or sucking it for a certain period of time: a ‘chaw‘, which refers to a portion of tobacco the size of a golf ball, is generally chewed, whereas a ‘quid’ is usually a much smaller portion and is held in the mouth rather than chewed (Pindborg et al., 1992).

(a) Betel quid with tobacco

Betel quid with tobacco, commonly known as paan or pan, consists of four main ingredients: (i) betel leaf (Piper betle), (ii) areca nut (Areca catechu), (iii) slaked lime and (iv) tobacco. Of these, tobacco is the most important ingredient for regular users. Betel quid can be prepared by the vendor or at home. Various tobacco preparations are used in unprocessed, processed or manufactured forms. Tobacco may be used in raw, sun-dried or roasted form, then finely chopped or powdered and scented. Alternatively, tobacco may be boiled, made into a paste and scented with rosewater or perfume. The final product is placed in the mouth and chewed. Betel quid with tobacco is used in Central, East, South and South-East Asia, in the western Pacific and in migrant communities arising therefrom (Bhonsle et al., 1992; Gupta & Ray, 2003). Exposure to and the health effects of betel quid with or without tobacco are described in detail in a previous monograph (IARC, 2004a).

(b) Chimó

Chimó is specific to Venezuela. It contains tobacco leaf, sodium bicarbonate, brown sugar, ashes from the Mamón tree (Melicocca bijuga), and vanilla and anisette flavourings. The ingredients vary according to the region within Venezuela. Tobacco leaves are crushed and boiled for several hours, during which starch and fibre are discharged. The remaining portion becomes a concentrated product: 10 kg of tobacco yield 1 kg of ‘pasta’. For maturation, chimó is then placed in natural containers or ‘taparas‘ (the dried fruit from the Tapara tree) or is wrapped in banana leaves. The matured paste is ‘seasoned’ with the ingredients listed above. Finally, it is packaged in small tins or candy-like wrapped cylinders. A small amount of chimó is placed between the lip or cheek and the gum and left there for some time, usually 30 min. The mixture of chimó and saliva is spat out.

(c) Creamy snuff

Creamy snuff consists of finely ground tobacco mixed with aromatic substances, such as clove oil, glycerin, spearmint, menthol and camphor, salts, water and other hydrating agents. It is often used to clean teeth. The manufacturer recommends letting the paste linger in the mouth before rinsing. Creamy snuff is manufactured commercially and marketed as a dentifrice, and is commonly used as such by women in South Asia.

(d) Dry snuff

In Europe and the USA, tobacco (primarily Kentucky and Tennessee tobacco) is fire-cured, then fermented and processed into a dry, powdered form. The moisture content of the finished product is less than 10%. Dry snuff is packaged and sold in small metal or glass containers. Typically, in the USA, a pinch (called a ‘dip’) is held between the lip or cheek and gum. In Europe, it is commonly inhaled into the nostrils (see Section 1.2.2 Nasal use).

In India, dry snuff was once commonly used nasally, but is now used mainly orally. It is frequently prepared at home by roasting coarsely cut tobacco on a griddle and then powdering it. This pyrolysed snuff-like preparation, mainly used in Goa, Maharastra, Gujarat and eastern parts of India, is widely used by the poorer classes as a dentifrice (applied to the teeth and gums), especially by women, but tends to be used many times a day, due to its addictive properties. It is known as bajjar or tapkir/tapkeer.

In many regions of the world, dry snuff is used both orally and nasally.

In northern Africa, dry snuff is known as naffa, tenfeha or nufha.

(e) Gudhaku

Gudhaku is a paste made of powdered tobacco and molasses. It is available commercially and is stored in a metal container. Gudhaku is applied to the teeth and gums with the finger, predominantly by women in India in the States of Bihar, Orissa, Uttar Pradesh and Uttaranchal.

(f) Gul

Gul contains tobacco powder, molasses and other ingredients and is manufactured commercially. It is applied to the teeth for the purpose of cleaning and then to the gums many times during the day. Gul is used in South Asia, including the Indian Subcontinent.

(g) Gutka

Gutka is manufactured commercially and consists of sun-dried, roasted, finely chopped tobacco, areca nut, slaked lime and catechu mixed together with several other ingredients such as flavourings and sweeteners. The product is sold in small packets or sachets. It is held in the mouth, sucked and chewed. Saliva is generally spat out, but is sometimes swallowed. Gutka is used in South Asia, including the Indian Subcontinent, and by Asian expatriates in several parts of the world, especially Canada, the United Kingdom and the USA (IARC, 2004a).

(h) Iq'mik

Fire-cured tobacco leaves are mixed with punk ash, which is generated by burning a woody fungus that grows on the bark of birch trees. The separate ingredients are available at grocery stores and retail outlets, but are generally combined by the user before use. Users pinch off a small piece and chew the iq'mik. The user may pre-chew the iq'mik and place it in a small box for later use by others, including children and sometimes teething babies. Iq'mik is used by native Americans in the northwestern parts of North America.

(i) Khaini

Khaini is made from sun-dried or fermented coarsely cut tobacco leaves. The tobacco used for khaini is from N. rustica and/or N. tabacum. The tobacco leaves are crushed into smaller pieces. A pinch of tobacco is taken in the palm of the hand, to which a small amount of slaked lime paste is added. The mixture is then rubbed thoroughly with the thumb. Khaini is usually prepared by the user at the time of use, but is also available commercially. It is held in the mouth and sucked or chewed. Areca nut may sometimes be added to khaini by the user. Khaini is used in South Asia, including the Indian Sub-continent.

(j) Khiwam

Khiwam (or qimam) consists of tobacco extract, spices and additives. The tobacco used for khiwam is from N. rustica and/or N. tabacum. Tobacco leaves are processed by removing their stalks and stems, then boiling and soaking them in water flavoured with spices (e.g. saffron, cardamom, aniseed) and additives such as musk. The resulting pulp is mashed, strained and dried into a paste. The paste is placed in the mouth and chewed. Khiwam may also be used in betel quid (IARC, 2004a). It is used in South Asia, including the Indian Subcontinent.

(k) Loose-leaf

Loose-leaf tobacco is manufactured commercially and consists of loose cigar tobacco leaves from Pennsylvania and Wisconsin that are air-cured, stemmed, cut or granulated, and loosely packed to form small strips of shredded tobacco. Most brands are sweetened and flavoured with liquorice, and are typically sold in pouches weighing about 3 oz. Loose-leaf tobacco is high in sugar content (approximately 35%). A piece of tobacco 0.75–1 in in diameter is placed between the cheek and lower lip, typically toward the back of the mouth. It is either chewed or held in place. Saliva is spat or swallowed. Loose-leaf is used in Europe and North America.

(l) Maras

In Turkey, a type of smokeless tobacco called maras is widely used in the southeastern region, especially in the cities of Kahramanmaras and Gaziantep. First, sun-dried leaves of the tobacco plant species N. rustica L. — known locally as ‘crazy tobacco’ — are powdered and mixed with the ash of wood, in particular oak, walnut or grapevine, in 1:2 or 1:3 proportions (tobacco and oak, respectively). Then, water is sprinkled onto the mixture for humidification. A small amount of the mixture (approximately 1 g) is applied between the lower labial mucosa and gingiva for 4–5 min. This procedure is repeated many times during the day; some people even sleep with the powder in their mouth.

(m) Mawa

Mawa is a mixture of small pieces of sun-cured areca nut with crushed tobacco leaves and slaked lime. The resulting mixture is about 95% areca nut by weight. It is placed in the mouth and chewed for 10–20 min. Mawa is used in South Asia, including the Indian Subcontinent.

(n) Mishri

Mishri is made from tobacco that is baked on a hot metal plate until toasted or partially burnt, and then powdered. It is applied to the teeth and gums as a dentifrice, usually twice a day and more frequently in some cases. Users then tend to hold it in their mouths. Mishri is used in South Asia, including the Indian Subcontinent.

(o) Moist snuff

The tobacco is either air- or fire-cured, then processed into fine particles (‘fine-cut’) or strips (‘long-cut’). Tobacco stems and seeds are not removed. The final product may contain up to 50% moisture. Moist snuff is sold either loose or packaged in small, ready-to-use pouches called packets or sachets. A pinch (called a dip) or a pouch is placed and held between the lip or cheek and gum. Saliva may be swallowed or, more commonly, spat out. Moist snuff is used in Europe and North America, and is the most common form of smokeless tobacco in the USA (see Section 1.4.2).

Swedish-type moist snuff (snus) consists of finely ground dry tobacco (Kentucky and Virginia tobacco), mixed with aromatic substances, salts (sodium chloride), water, humidifying agents and chemical buffering agents (sodium carbonate). A pinch (called a dip) is placed between the gum and upper lip. The average user keeps snuff in the mouth for 11–14 h per day. In Sweden, the portions come in two doses, regular and ‘mini-portions’ (1.0 g and 0.5 g tobacco, respectively), or loose.

(p) Naswar

Naswar (or nass) is a mixture of sun-dried, sometimes only partially cured, powdered local tobacco (N. rustica), ash, oil, flavouring agents (e.g. cardamom, menthol), colouring agents (indigo) and, in some areas, slaked lime. It is made by pouring water into a cement-lined cavity to which lime is added, followed by tobacco. Colouring and flavouring agents are then added. The ingredients are then pounded and mixed with a heavy wooden mallet. The type of oil varies by region. Water is added and the mixture is rolled into balls. It is then usually placed under the tongue (in the floor of the mouth) and then sucked. Naswar is used widely in Afghanistan, Iran, Pakistan and the central Asian Republics, and in South Africa.

(q) Plug chewing tobacco

Plug is the oldest form of chewing tobacco. It is produced from the heavier grades of Burley and bright tobacco or cigar tobacco leaves harvested from the top of the plant. Once the stems are removed, the leaves are immersed in a mixture of liquorice or sugar, pressed into a plug, covered by a wrapper leaf and re-shaped into bricks or flat blocks. Moist plug tobacco has at least 15% moisture content; plug or ‘firm plug’ tobacco has less than 15% moisture content. Sugar content is approximately 25%. Moist plug is chewed, or held between the cheek or lower lip and gum. Saliva is spat or swallowed. Moist plug is used primarily in North America.

(r) Red tooth powder

Red tooth powder is a fine tobacco powder that is red in colour and contains many additional ingredients including herbs and flavouring agents. It is manufactured commercially and marketed as a herbal product. Red tooth powder is used in South Asia as a dentifrice.

(s) Shammah

Shammah is a mixture of powdered tobacco, lime, ash, black pepper, oils and flavourings. The greenish-yellow powder is placed in the buccal or lower labial vestibule of the mouth. The user spits out insoluble debris. It is used in the Middle East, including some parts of southern Saudi Arabia and Yemen.

(t) Snuff

Two types of snuff are used orally: dry snuff and moist snuff; these are discussed under (d) and (o), respectively. Dry snuff may also be used nasally (see Section 1.2.2(a)).

(u) Tobacco chewing gum

Tobacco chewing gum was developed by the company Swedish Match in 2003 and marketed under the brand name ‘Fire’ as an alternative tobacco product and test marketed in Tokyo, Japan.

(v) Tobacco tablets

Tobacco tablets were introduced on the market in 2002 in the form of 10-piece blister card. They are made of compressed powdered tobacco, mint and eucalyptus and melt in the mouth. Each tablet contains approximately 1.3 mg nicotine (Nguyen et al., 2002). Tobacco tablets are also known by the brand names Ariva® and Cigalett®.

(w) Toombak

Toombak is a moist tobacco product used primarily in Sudan. It consists of tobacco (N. rustica and/or N. glauca) and sodium bicarbonate. Tobacco leaves are harvested and left in a field to dry uniformly. The leaves are then tied into bundles, sprinkled with water and stored for a couple of weeks at 30–45 °C to allow fermentation. They are then ground and matured for up to 1 year. After maturation, toombak vendors (in toombak shops) place the product in bowls and gradually add sodium bicarbonate until the mixture is approximately four parts of tobacco to one part of sodium bicarbonate. The mixture is blended by hand and constantly tested with the tips of the fingers until it becomes moist and hardened. The toombak is then placed in an air-tight container shortly before sale.

Toombak is rolled into a ball that weighs about 10 g and is called a saffa. The saffa is held between the gum and the lip or cheek, or under the tongue on the floor of the mouth. It is sucked slowly for 10–15 min. Male users periodically spit, while female users typically swallow the saliva generated. The user usually rinses his/her mouth with water after the saffa is removed. Commercial names for toombak include El-Sanf (of high quality), Wad Amari (accrediting the person who was believed to have introduced it) and Sultan El-Khaif (the power to improve one's state of mind) (Idris et al., 1998a).

(x) Tuibur

Tuibur (or hidakphu) is tobacco water for oral use. Tobacco smoke is passed through water and the water is used for gargling or sipping. Tuibur is commonly used in the northeastern states of India (Manipur, Mizoram, Sikkim, Tripura) (Mahanta et al., 1998).

(y) Twist/roll chewing tobacco

Twist/roll chewing tobacco is hand-made by commercial manufacturers. Dark, air- or fire-cured leaf Burley tobacco is treated with a tar-like tobacco leaf extract and flavours, and twisted into rope-like strands that are dried. The product is sold by the piece in small (about 50 g) or larger sizes based on the number of leaves in the twist. Twist/roll is used in North America.

(z) Zarda

Zarda consists of tobacco, lime, spices and vegetable dyes. Tobacco leaves are broken up and boiled with lime and spices until dry. The mixture is dried and coloured with vegetable dyes. Zarda is generally chewed mixed with finely chopped areca nuts and spices. It is often used as an ingredient in betel quid. Zarda is commonly used in India and the Middle East, and is known as dokta in West Bengal.

1.2.2. Nasal use

(a) Dry snuff

Tobacco (primarily Kentucky and Tennessee tobacco) is fire-cured, then fermented and processed into a dry, powdered form. The moisture content of the finished product is less than 10%. It is packaged and sold in small metal or glass containers. In Europe, dry snuff is commonly inhaled into the nostrils. In many regions of the world, it is used both orally and nasally.

(b) Liquid snuff

Liquid snuff was reported to be used by the Nandi tribe in East Africa. It is used nasally (Hou-Jensen, 1964).

1.3. Chemical composition of smokeless tobacco

1.3.1. General overview

The type of tobacco used in a particular product has a decisive influence on its chemical composition. That of leaf tobacco varies with genetic make-up, environmental conditions and every step of production and handling (Tso, 1990). The classification of leaf tobacco commonly used in smokeless tobacco products is primarily based on curing methods (e.g. air-, flue- and fire-cured tobacco) and tobacco types (e.g. Burley, Wisconsin, Pennsylvania air-cured tobacco, dark fire-cured tobacco, Virginia flue-cured tobacco).

The first summary of chemical components found in tobacco and tobacco smoke was prepared by Stedman in 1968. Since then, frequent additions have been made to the list and, in 1988, the number of compounds identified in tobacco totaled 3044 (Roberts, 1988). The latter count has not been confirmed by independent research. Moreover, Roberts (1998) does not list many of the constituents that are currently known to be present in tobacco (e.g. volatile N-nitrosamines, tobacco-specific N-nitrosamines, N-nitrosamino acids). Hoffman et al. (2001) expanded the list to include 23 N-nitrosamines and 28 pesticides, which brought the number to 3095 constituents in tobacco. The identification of each single compound is an arduous task and requires a vigorous confirmation protocol that uses state-of-the-art instrumentation as well as synthesis.

During preparation for product manufacture, tobacco leaves, stems and other ingredients are blended to achieve a specific nicotine content, pH, taste, flavour and aroma. These features are critical for acceptance of the product by the user. For cigarettes, it has been demonstrated that the type of tobacco blend significantly affects these features as well as the toxicity of the product (Abdallah, 2003; Baker & Smith, 2003). The pH strongly influences the concentration of unprotonated nicotine, the bioavailable form of nicotine (Djordjevic et al., 1995; Henningfield et al., 1995; Richter & Spierto, 2003), while the nitrite content influences the levels of nitrosamines in the product (Fischer et al., 1989; Burton et al., 1994; Hoffmann et al., 1995).

A choice of 60 N. tabacum species and 100 varieties of tobacco can be blended. However, the majority of commercial tobacco products use N. tabacum species, which are grown in North America and throughout the world. The alkaloid content in N. tabacum species varies greatly. From a random examination of 152 cultivated varieties, a range of alkaloid content between 0.17 and 4.93% was found. Tobacco types, plant parts, cultural practices, degree of ripening and fertilizer treatment are among some prominent factors that determine the level of alkaloids in Nicotiana plants. Every step in tobacco production that affects plant metabolism influences the level of alkaloid content to a certain degree. Cured tobacco lines can contain between 0.2 and 4.75% nicotine by weight, depending on plant genetics, growing conditions, degree of ripening, fertilizer treatment and leaf position on the stalk (Tso, 1990; Stratton et al., 2001).

N. rustica species is cultivated in some parts of eastern Europe, Asia Minor and Africa, and the cured leaves may contain up to 12% nicotine. In greenhouse-grown plants, N. rustica can accumulate up to 5.3 mg nicotine/g tobacco (98.2% of total alkaloids) and in field-grown plants up to 24.8 mg nicotine (97.1% of total alkaloids) (Sisson & Severson, 1990). Toombak, which contains N. rustica tobacco, was reported to contain the highest levels of nicotine (up to 102.4 mg/g dry wt) and nicotine-derived tobacco-specific nitrosamines ever measured in consumer products (Idris et al., 1991; Prokopczyk et al., 1995).

The chemical composition of tobacco undergoes substantial changes during growing, curing, processing and storing (Burton et al., 1983, 1989a,b; Peele et al., 1995; Walton et al., 1995; Wiernik et al., 1995; Peele et al., 2001; Bush et al., 2001). The purpose of curing is to produce a dried leaf of suitable physical properties and chemical composition. At the beginning of curing, a tobacco leaf is metabolically active and continues to live until biochemical processes are arrested by thermal effects or desiccation. In curing, the starch content of the leaves declines drastically, while the amount of reducing sugars increases by 100%. Protein and nicotine contents decrease slightly. The bulk of the processed tobacco leaf before fermentation consists of carbohydrates (about 50%) and proteins. Fermentation of cured tobacco causes the contents of carbohydrates and polyphenols in the leaves to diminish. Other major components are alkaloids (0.5–5.0%), which include nicotine as the predominant compound (85–95% of total alkaloids), terpenes (0.1–3.0%), polyphenols (0.5–4.5%), phytosterols (0.1–2.5%), carboxylic acids (0.1–0.7%), alkanes (0.1–0.4%), aromatic hydrocarbons, aldehydes, ketones, amines, nitriles, N- and O-heterocyclic hydrocarbons, pesticides, alkali nitrates (0.01–5%) and at least 30 metallic compounds (Brunnemann & Hoffmann, 1992; IARC, 2004b).

Because of the disappearance of carbohydrates and polyphenols during fermentation, heavy casings [additives applied during processing] such as molasses, liquorice and fruit extracts are added to tobacco to meet the consumer's requirements (e.g. they improve taste, flavour and aroma, and prolong shelf-life). Many smokeless tobacco formulations use plant extracts or chemicals as flavouring agents (Mookherjee & Wilson, 1988; Roberts, 1988; Sharma et al., 1991). Tobacco additives may include methyl or ethyl salicylate, β-citronellol, 1,8-cineole, menthol, benzyl benzoate, eugenol and possibly coumarin, among others (LaVoie et al., 1989; Stanfill et al., 2006). Eugenol (ranging from < 0.00005 to 25 706 µg/g in Dentobac Creamy Snuff sold in India; Gupta, 2004) and menthol are used to numb the throat and facilitate tobacco use (Ahijevych & Garrett, 2004; Wayne & Connolly, 2004). Ascorbic acid is added to tobacco as an antimicrobial agent whereas the addition of sodium propionate serves as a fungicide. Other additives, such as ammonia, ammonium carbonate and sodium carbonate, are applied to control nicotine delivery by raising pH and subsequently the level of unprotonated nicotine which is the form of nicotine that is most readily absorbed through the mouth into the bloodstream (Djordjevic et al., 1995; Henningfield et al., 1995). However, the formulation of most of the additives, including flavours, remains a trade secret.

1.3.2. Carcinogenic compounds in smokeless tobacco

To date, 28 carcinogens have been identified in smokeless tobacco (Table 3; adapted from Brunnemann & Hoffmann, 1992). The major and most abundant group of carcinogens are the non-volatile alkaloid-derived tobacco-specific N-nitrosamines (TSNA) and N-nitrosoamino acids. Other carcinogens reportedly present in smokeless tobacco include volatile N-nitrosamines, certain volatile aldehydes, traces of some polynuclear aromatic hydrocarbons such as benzo[a]pyrene, certain lactones, urethane, metals, polonium-210 and uranium-235 and -238 (see Brunnemann & Hoffmann, 1992 for review).

Table 3. Chemical agents identified in smokeless tobacco products.

Table 3

Chemical agents identified in smokeless tobacco products.

There are three major types of nitroso compounds in smokeless tobacco: (a) non-volatile TSNA, including 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) and N′-nitrosonornicotine (NNN); (b) N-nitrosamino acids, including N-nitrososarcosine (NSAR), 3-(methylnitrosamino)propionic acids (MNPA) and 4-(methylnitrosamino)butyric acids (MNBA); and (c) volatile N-nitrosamines, including N-nitrosodimethylamine (NDMA), N-nitrosopyrrolidine (NPYR), N-nitrosopiperidine (NPIP) and N-nitrosomorpholine (NMOR).

TSNA are present in fresh green tobacco leaves in N. tabacum species, at levels of up to 0.39 µg/g NNN and 0.42 µg/g NNK in the top leaves of tobacco (flue-cured type) grown in the USA (Djordjevic et al., 1989a), up to 0.035 µg/g NNN and 0.0115 µg/g NNK in N. tabacum grown in India (Bhide et al., 1987a) and up to 46.1 µg/g NNN and 2.34 µg/g NNK in N. rustica species grown in India (Bhide et al., 1987a). However, TSNA are formed primarily during tobacco curing, fermentation and ageing, from their alkaloid precursors (namely, nicotine, nornicotine, anatabine and anabasine) and from nitrite/nitrate. The nitrate or nitrite content, the mode of curing and the various steps of processing are therefore the determining factors for the yields of TSNA in tobacco (Burton et al., 1989a,b; Fischer et al., 1989; Chamberlain & Chortyk, 1992; Djordjevic et al., 1993a; Burton et al., 1994; Peele et al., 2001; Li & Bush, 2004). NNN, N′-nitrosoanatabine (NAT) and N′-nitrosoanabasine (NAB) are formed primarily from the corresponding secondary amines in the early stages of tobacco processing; some NNN and the majority of NNK are formed from the tertiary amine nicotine at the later stage of tobacco curing and fermentation (Spiegelhalder & Fischer, 1991).

In addition to these three groups of compounds, smokeless tobacco contains N-nitrosodiethanolamine (NDELA), which is formed from diethanolamine, a residual contaminant in tobacco. In 1981, the levels of NDELA were up to 224 ng/g in chewing tobacco and up to 6840 ng/g in fine-cut moist snuff. Treatment of Burley leaves with the sucker growth inhibitor maleic hydrazide significantly increased the hydrazine content. Although a tolerance of 80 ppm for maleic hydrazide was established in at least three European countries and the USA, concentrations up to 269 ppm were reported for the flue-cured tobacco harvested in Georgia, USA, in 1990 (Sheets, 1990). As the use of maleic hydrazide–diethanolamine as a sucker growth-controlling agent was gradually reduced, the concentration of NDELA decreased to less than 100 ng/g in 1990 (Brunnemann & Hoffmann, 1991).

Polycyclic aromatic hydrocarbons (PAHs) originate primarily from polluted air and perhaps from fire-curing of some tobaccos.

Formaldehyde, acetaldehyde and crotonaldehyde, which are themselves probable or known human carcinogens, probably contribute to the carcinogenic potential of smokeless tobacco. It is known that tobacco contains a large spectrum of alkyl aldehydes that contribute to its aroma and are formed from amino acids and sugars by heating during tobacco processing (Coleman & Perfetti, 1997).

The α- and β-angelica lactones have been reported in natural tobacco (Weeks et al., 1989). A minor group of polyphenols in tobacco are coumarins, of which scopoletin is the major representative. The presence of urethane in fermented Burley tobacco (up to 400 ng/g) is not unexpected since the fermentation of food and beverages leads to the formation of this compound. Both air- and flue-cured tobaccos contain hydrazines.

Radioactive polonium-210, which decays to radon, originates from soil that is fertilized with phosphates rich in radium-226 (Tso et al., 1966).

1.3.3. Smokeless tobacco products

(a) Nicotine, pH and unprotonated nicotine

All smokeless tobacco products contain nicotine as a major constituent, which is addictive (Henningfield et al., 1997; Hatsukami & Severson, 1999). The level of unprotonated nicotine affects the rate and degree of nicotine absorption (see Section 4.1).

Djordjevic et al. (1995) analysed 17 brands of moist snuff purchased in Westchester County, New York (USA) in 1994. In addition, samples of the five leading brands were purchased in six areas of the USA (Alameda, CA; Boston, MA; Denver, CO; Lansing, MI; Lexington, KY; Westchester, NY) and analysed separately to determine geographic variations. The nicotine content in 17 brands ranged from 0.47% dry wt (in Hawken Wintergreen) to 3.43% (in Skoal Long Cut Mint), which corresponds to 3.4 mg/g and 14.5 mg/g, respectively; the pH ranged from 5.39 (in Skoal Bandits Classic) to 7.99 (in Kodiak Wintergreen); unprotonated nicotine ranged from 0.23% of total nicotine (in Skoal Bandits Classic) to 48.3% (in Kodiak Wintergreen). The average values for the five best-selling brands of moist snuff in the USA in 1994 are summarized in Table 4.

Table 4. Nicotine content and pH of the five leading brands purchased at different locations in the USA.

Table 4

Nicotine content and pH of the five leading brands purchased at different locations in the USA.

Similar findings were reported by Henningfield et al. (1995) for products purchased at three locations (Baltimore, MD; Boston, MA; Lansing MI; Table 4). Both studies show that nicotine-dosing capability varies remarkably between products and that it is governed predominantly by nicotine content and pH level.

The Centers for Disease Control and Prevention (CDC) carried out an analysis of 18 smokeless tobacco products (eight brands of moist snuff and 10 of loose-leaf chewing tobacco) (Richter & Spierto, 2003). Among moist snuff brands, Timber Wolf Long Cut Straight contained the highest amount of nicotine (13.54 mg/g) followed by Copenhagen snuff and Skoal (12.71 mg/g and 12.94 mg/g, respectively). Consistent with the findings by Djordjevic et al. (1995), the highest pH was measured for Kodiak Wintergreen (pH, 8.28), which also had the highest quantity of unprotonated nicotine (64.5%; 5.81 mg/g). The lowest pH and amount of free nicotine were reported for Hawken Wintergreen (pH, 5.35; 0.20% free nicotine or 0.01 mg/g).

Another CDC study (CDC, 1999a) also reported that Copenhagen snuff and Kodiak Wintergreen had the highest pH (8.18 and 8.35, respectively) and the highest concentration of unprotonated nicotine (6.23 and 5.83 mg/g tobacco, respectively); Skoal Bandits Straight and Hawken Wintergreen had the lowest pH (5.52 and 5.24, respectively) as well as the lowest concentration of unprotonated nicotine (0.025 and 0.007 mg/g tobacco, respectively).

In 1996, Massachusetts enacted a tobacco product disclosure law which required manufacturers of cigarettes and smokeless tobacco products to disclose the ingredients and nicotine content by brand for average consumers. The Massachusetts Department of Public Health (MDPH) promulgated regulations in 1996 that required cigarette and smokeless tobacco manufacturers to file annual reports on nicotine yield by brand (MDPH, 2004). The requirements for reporting on smokeless tobacco were based on federal rules published by the CDC, adopted in 1996 and revised in 1999 (CDC, 1999b). Unlike Massachusetts, where disclosure of nicotine is a public record, data reported to the CDC remain private. Annual reports submitted by all smokeless tobacco manufacturers who sold products in Massachusetts from 1997–2003 contributed the most comprehensive data base on the levels of total nicotine (expressed as % and mg/g adjusted for moisture), tobacco pH and the levels of unprotonated nicotine (expressed as % of total nicotine and mg/g dry wt) in smokeless tobacco. Tables 57 list the pH, and total and unprotonated nicotine content of individual brands of, respectively, chewing tobacco, dry snuff and moist snuff sold in the Commonwealth of Massachusetts in 2003; Table 8 presents the mean values for each type of tobacco product.

Table 5. Chemical composition of chewing tobacco sold in Massachusetts (USA) in 2003.

Table 5

Chemical composition of chewing tobacco sold in Massachusetts (USA) in 2003.

Table 7. Chemical composition of moist snuff sold in Massachusetts (USA) in 2003.

Table 7

Chemical composition of moist snuff sold in Massachusetts (USA) in 2003.

Table 8. Ranges of pH and nicotine concentration in smokeless tobacco products sold in Massachusetts (USA) in 2003.

Table 8

Ranges of pH and nicotine concentration in smokeless tobacco products sold in Massachusetts (USA) in 2003.

Table 6. Chemical composition of dry snuff sold in Massachusetts (USA) in 2003.

Table 6

Chemical composition of dry snuff sold in Massachusetts (USA) in 2003.

On average, moist snuff contained the highest percentage of moisture (mean, 52.6%; range, 21.58–55.77%) and nicotine (mean, 2.58% dry wt; range, 0.49–3.7%) (Table 8). Dry snuff had the lowest moisture content (mean, 8.2%; range, 5.38–23.9%) but middle range of nicotine (mean, 1.82%; range, 1.14–2.69%). Chewing tobacco had the lowest nicotine content (mean, 1.22%; range 0.45–4.65%). Moist snuff had, on average, the highest pH (7.43 versus 6.36 and 5.82 in dry snuff and chewing tobacco, respectively). Because of the high pH, the levels of unprotonated nicotine in moist snuff averaged 3.52 mg/g product (range, 0.03–8.57 mg/g); this is fivefold higher than that in dry snuff and 32-fold higher than that in chewing tobacco. The highest concentration of unprotonated nicotine was reported for Longhorn Fine Cut Natural, which is marketed by Swedish Match North America (Table 7).

Regular and comprehensive reporting on the chemical composition of smokeless tobacco products to the MDPH enables analysis of trends in chemical composition over time and comparison of the levels of specific constituents between different brands or types of products. The trends for pH and nicotine content (both total and unprotonated) in the four leading brands of moist snuff in the USA (Copenhagen Fine Cut, Skoal Straight Fine Cut, Kodiak Wintergreen and Timberwolf Fine Cut Wintergreen) (Maxwell Tobacco Facts Book, 2002) from 1997 to 2003 are presented in Figures 13 (MDPH, 2004). Nicotine content (% dry wt) in three of the brands did not change notably between 1997 and 2003, while it increased steadily in Timber Wolf from 2.8 to 3.6% during the same period.

Figure 1

Figure 1

The pH of leading US moist snuff products (1997–2003)

Figure 3

Figure 3

The unprotonated nicotine content (% of the total nicotine) in leading US moist snuff brands (1997–2003)

The pH values for Copenhagen fine cut were constant between 1997 and 2003 (Figure 2), while the pH of Skoal Fine Cut Straight dropped significantly during the same period. Of the four brands, Kodiak has had the highest pH since 1999, and the pH of the Timber Wolf fluctuated between 7.6 and 8.0. The latter observation underlines the importance of monitoring the composition of all products rather than using one brand as a proxy for different types of smokeless tobacco product, or sub-brands of a brand family.

Figure 2

Figure 2

The nicotine content (% dry weight) in leading US moist snuff products (1997–2003)

As shown in Figure 3, the levels of unprotonated nicotine were the highest in Kodiak Wintergreen, and increased from 35.19% total nicotine in 1997 to 60.27% total nicotine in 2003. This pattern parallels the trend in pH. On average, the levels of unprotonated nicotine in Copenhagen and Skoal brand families decreased steadily overtime. However, for the individual brands, this trend was only true for Skoal Fine Cut Straight, and not for Copenhagen Fine Cut, similar to the observation regarding pH. The percentage of unprotonated nicotine for Timberwolf also parallels the pH (Figures 2 and 3). As a result of the constant interplay of pH, nicotine content and moisture in tobacco products, the levels of unprotonated nicotine vary from product to product and from year to year.

In summary, the data from the MDPH show that pH and unprotonated nicotine content are brand- and company-specific. pH appears to be the primary determinant of nicotine absorption (Tomar & Henningfield, 1997). Among the 562 components reported on the list of additives for smokeless tobacco products (House of Representatives, 1994), several salts (e.g. ammonium, sodium and potassium salts) may alter the pH of smokeless tobacco. Moreover, smokeless tobacco contains components that are intended to control delivery of nicotine to the body (Food and Drug Administration, 1996). However, exposure of users to tobacco toxins does not depend only on their concentration in a particular product but also how the product has been used. Smokeless tobacco users who dip or chew eight to 10 times a day may be exposed to the same amount of nicotine as individuals who smoke 30–40 cigarettes a day (DHHS, 1986). Lemmonds et al. (2005) examined the relationship between topographical measures of oral smokeless tobacco and biomarkers of exposure to tobacco and carcinogens. The major finding of the study was that frequency and duration measures of smokeless tobacco use are significantly correlated with total cotinine, a major metabolite of nicotine. Fifty-four male snuff users of 2.8 tins/week (6.1 dips/day) excreted on average 20.1 nmol cotinine/mg creatinine (or 3.3 nmol cotinine per dip) in urine compared with 27 nmol cotinine/mg creatinine excreted by smokers who consumed on average 27.9 cigarettes/day (or 1.07 nmol creatinine per cigarette) (Hecht et al., 2005). Thus, snuff dippers are exposed to 3.08-fold higher amounts of nicotine than cigarette smokers. This high exposure to nicotine needs to be taken into consideration when recommending nicotine replacement therapy to those who contemplate quitting snuff use. Moreover, it has been shown that increasing the nicotine concentration in the presence of alcohol significantly increases the penetration of NNN across the oral mucosa (Du et al., 2000).

The latest information on the chemical composition of 14 varieties of smokeless tobacco products used in India, including pH and nicotine content, was made available in a report to the WHO South-East Asian Regional Office (Gupta, 2004; Table 9). Some products had a pH of up to 10.1 and a nicotine content of up to 10.2 mg/g.

Table 9. Chemical composition of smokeless tobacco products used in India.

Table 9

Chemical composition of smokeless tobacco products used in India.

Ayo-Yusuf et al. (2004) reported on the pH and nicotine content of moist snuff products consumed in South Africa. The pH ranged from 7.1 to 10.1, the nicotine content from 0.8 to 1.6% wet wt [11.6–29.3 mg/g dry wt, as adjusted for moisture content] and from 10.1 to 99.1% in the unprotonated form.

A new product that is on the market, tobacco tablets, also referred to as Ariva® or Cigalett®, contain 1.3 mg nicotine per tablet and have a pH of 8.4. The ‘buffering capacity’ of Ariva® is sufficient to control the acidic pH of human saliva (Nguyen et al., 2002).

(b) Tobacco-specific N-nitrosamines (TSNA)

Hoffmann et al. (1995) provided the most comprehensive insight into the levels of major tobacco carcinogens in the leading brands of moist snuff sold in the USA. The purpose of the study was threefold: (a) to determine the concentrations of major carcinogenic TSNA and N-nitrosamino acids in each of the five most popular brands of moist snuff (Table 10); (b) to analyse quantitative differences in selected snuff components (e.g. NNK and NNN) between two major categories of moist snuff: a category that comprised those brands known to have high levels of unprotonated nicotine (Copenhagen, Skoal Fine Cut and Kodiak) versus a category that comprised those brands known to have low levels (Hawken and Skoal Bandits); and (c) to compare the concentration of nicotine, NNN, NNK and total TSNA between these two categories. Concentrations (mean ± standard deviation [SD]) of nicotine, NNN, NNK and total TSNA in the two categories were as follows: nicotine, 11.6 ± 1.06 mg/g versus 6.96 ± 3.62 mg/g (p = 0.0017); NNN, 7.74 ± 1.70 µg/g versus 4.17 ± 1.35 µg/g (p < 0.0001); NNK, 1.23 ± 0.68 µg/g versus 0.61 ± 0.41 µg/g (p = 0.012); and total TSNA (including NNN, NNK, NAB and NAT), 14.3 ± 3.82 µg/g versus 6.3 ± 2.56 µg/g (p < 0.001). In another study, moist snuff with a high pH and high unprotonated nicotine content, purchased in 2000, contained 15.4 µg/g dry wt NNN and 2.5 µg/g dry wt NNK (Brunnemann et al., 2002). The brand Conwood's Grizzly contained 70.8 µg/g NNN and 10.1 µg/g NNK (Brunnemann et al., 2004).

Table 10. Levels of tobacco-specific N-nitrosamines and N-nitrosamino acids in the five leading brands sold in the USA, 1994.

Table 10

Levels of tobacco-specific N-nitrosamines and N-nitrosamino acids in the five leading brands sold in the USA, 1994.

Table 11 shows an international comparison of the concentrations of two carcinogenic TSNA, NNN and NNK, as well as of tobacco pH (as determined in an aqueous tobacco suspension). The ranges for all three measures are wide and are product-specific and country-specific. The highest values of pH were measured in naswar from Uzbekistan (Brunnemann et al., 1985), toombak from Sudan (Idris et al., 1998a) and new moist snuff brands recently introduced in South Africa (Ayo-Yusuf et al., 2004). The highest concentrations of NNN and NNK were measured in some moist snuff brands in the USA (135 and 17.8 µg/g tobacco, respectively). However, values as high as 3085 and 7870 µg/g dry wt tobacco, respectively, have been measured in home-made toombak.

Table 11. International comparison of the pH and concentration ranges of N-nitrosonornicotine (NNN) and 4-(N-methyl-N-nitrosamino)-1-(3-pyridyl)-1-butanone (NNK) in smokeless tobacco products (µg/g tobacco).

Table 11

International comparison of the pH and concentration ranges of N-nitrosonornicotine (NNN) and 4-(N-methyl-N-nitrosamino)-1-(3-pyridyl)-1-butanone (NNK) in smokeless tobacco products (µg/g tobacco).

Although there has been a decline in the concentrations of nitrosamines in some smokeless tobacco products in Sweden and the USA since the 1980s (Djordjevic et al., 1993b; Brunnemann et al., 2004; Österdahl et al., 2004), the trend may not apply to other products and countries. For example, the concentrations of NNN and NNK in the two leading snuff brands in the USA were reduced significantly by 70–90% from 1980 to 1992, based on dry weight (Djordjevic et al., 1993b). However, samples of a new brand of moist snuff introduced on the US market in the 1990s contained very high amounts of NNN and NNK (up to 57.1 and 16.4 µg/g dry wt, respectively) (Hoffmann et al., 1991). Moreover, the commercial brand Conwood's Grizzly, purchased in the USA in 2003, contained 70.8 µg/g dry wt NNN and 10.1 µg/g dry wt NNK (Brunnemann et al., 2004). In Sweden, the concentrations of NNN and NNK in moist snuff decreased, respectively, from 3.8 and 0.8 µg/g in 1983 to 0.49 µg/g and 0.19 µg/g wet wt in 2002 (87% and 76%decrease, respectively; Österdahl et al., 2004). Values for NNN and NNK of up to 3085 and 7870 µg/g, respectively, were reported in toombak (Idris et al., 1991, 1998a). The latest report by Stepanov et al. (2006) shows the wide range of TSNA concentrations in 19 brands of new and conventional smokeless tobacco products purchased in retail stores in the USA or online from Snus Worldwide, Sweden. Levels of NNN ranged from 0.019 µg/g wet wt in Ariva® hard snuff to 4.5 µg/g in Skoal Long Cut; those of NNK ranged from 0.032 µg/g in Revel to 1.6 µg/g in Copenhagen Long Cut; and those of NAT ranged from 0.12 µg/g in Ariva® to 4.1 µg in Skoal Long Cut Straight. Stepanov et al. (2005) also reported a wide range of TSNA concentrations in smokeless tobacco products from India: NNN concentrations ranged from not detected in supari and a sample of a tooth powder to 76.9 µg/g wet wt in khaini; those of NNK ranged from not detected to 28.4 µg/g in khaini.

In recent years, the Swedish Match Company has developed a new method for manufacturing oral snuff that uses select blends of tobacco as well as a new processing method. Instead of the dark fire-cured tobacco commonly used in US snuff, Swedish Match uses tobacco with a low nitrate content, which itself reduces TSNA levels. In addition, the tobacco is processed in a heated closed system that resembles pasteurization of milk, which eliminates bacteria that may be indirectly responsible for the formation of the nitrosamines (Parsons et al., 1986; Gothia, 2004). The company also encourages retailers to refrigerate packages to prevent the formation of TSNA during storage (see below).

In 2001, the MDPH initiated a study aimed at comparing traditional snuff brands with PREPs (Stratton et al., 2001). The study found that the levels of NNN, NNK, NAT and NAB in moist snuff produced by the new manufacturing process (Swedish Match brand Ettan) were up to 45 times lower than those in leading products manufactured under standard processes in the USA (Table 12).

Table 12. Levels of tobacco-specific N-nitrosamines (TSNA) in the five leading brands in the USA versus PREP.

Table 12

Levels of tobacco-specific N-nitrosamines (TSNA) in the five leading brands in the USA versus PREP.

In Sweden, all moist snuff brands on the market in 2002 contained low amounts of TNSA: NNN, 0.15–0.61 µg/g wet wt; and NNK, 0.03–0.36 µg/g wet wt. NNN concentrations in moist snuff decreased consistently from 1983 to 2002 from 3.8 to 0.49 µg/g wet wt and those of NNK from 0.80 to 0.19 µg/g wet wt (Österdahl et al., 2004).

Levels of TSNA in new oral snuff brands do not always parallel nicotine content (see Table 7 for the nicotine content and Table 12). For example, Taxi, a very high nicotine-delivery product manufactured by Swedish Match for the South African market, contains low levels of TNSA: NNN, 2.07 µg/g dry wt; and NNK, 0.29 µg/g dry wt (Brunnemann et al., 2004).

(c) N-Nitrosamino acids

The amino acids present in tobacco, and probably also the proteins with secondary amino groups, are amenable to N-nitrosation. Since 1985, numerous studies have reported the presence of nitrosamino acids in smokeless tobacco products. Levels of N-nitrosoamino acids in smokeless tobacco products worldwide are presented in Table 13. To date, 11 N-nitrosamino acids have been identified in smokeless tobacco: NSAR, N-nitrosoazetidine-4-carboxylic acid (NAzCA), MNPA, MNBA, N-nitrosoproline (NPRO), N-nitrosohydroxyproline (NHPRO), N-nitrosopipecolic acid (NPIC), N-nitrosothiazolidine- 4-carboxylic acid (NTCA), N-nitroso-2-methylthiazolidine-4-carboxylic acid (MNTCA), 4-(methylnitrosamino)-4-(3-pyridyl)butyric acid (iso-NNAC) and 2-(methylnitrosamino)-3-phenylpropionic acid (MNPhPA) (Ohshima et al., 1985; Tricker & Preussmann, 1988; Djordjevic et al., 1989b; Tricker & Preussmann, 1989, 1991; Hoffmann et al., 1995). Of these, the following have been established as carcinogens in experimental animals: NSAR, MNPA, MNBA and NAzCA. The concentration of the nitrosamino acids depends on the nitrate or nitrite content of the tobacco; in addition, they are formed during prolonged storage, particularly under adverse conditions of temperature and relative humidity (Djordjevic et al., 1993a).

Table 13. Comparison of the major carcinogenic N-nitrosamino acids in smokeless tobacco (µg/g dry wt) across countries.

Table 13

Comparison of the major carcinogenic N-nitrosamino acids in smokeless tobacco (µg/g dry wt) across countries.

The highest concentrations of N-nitrosamino acids in moist snuff purchased in the USA were found in Skoal Bandits Straight and Hawken Wintergren (13.45 and 11.56 µg/g, respectively) and the lowest in Kodiak (5.7 µg/g), which is opposite to the trend observed for TNSA (Hoffmann et al., 1995).

(d) Volatile N-nitrosamines

Volatile N-nitrosamines are formed from volatile amines and nitrosating agents. The levels of volatile N-nitrosamines in smokeless tobacco products worldwide are presented in Table 14. The highest amounts were found in moist snuff (NDMA up to 265 ng/g dry wt and NPYR up to 860 ng/g dry wt; see also Table 3). The presence of NMOR (see IARC, 1987) indicates contamination with morpholine either from additives or from diffusion of containers coated with morpholine-containing wax (Brunnemann et al., 1985; Brunnemann & Hoffmann, 1991).

Table 14. Comparison of the major carcinogenic volatile N-nitrosamines in smokeless tobacco (ng/g dry wt) across countries.

Table 14

Comparison of the major carcinogenic volatile N-nitrosamines in smokeless tobacco (ng/g dry wt) across countries.

(e) Other carcinogenic compounds

In smokeless tobacco products from the USA, the levels of benzo[a]pyrene ranged from < 0.1 to 63 ng/g in moist snuff (Hoffmann et al., 1986) and up to 90.5 ng/g in dry snuff (Brunemann & Hoffmann, 1992; Table 3). Bhide et al. (1984a) reported on the whole range of PAHs in Indian smokeless tobacco products such as mishri and snuff: benzo[a]pyrene, 7.6–66 ng/g; benzofluoranthenes (b + j + k), 35–231 ng/g; indeno[1,2,3-cd]pyrene, 4.3–24 ng/g; benz[a]anthracene, 19–79 ng/g; chrysene and triphenylene, 37–192 ng/g; benzo[e]pyrene, 10–110 ng/g; pyrene, 60–169 ng/g; fluoranthene, 55–218 ng/g; and benzo[ghi]perylene, 5.6–17 ng/g.

Hoffmann et al. (1987) reported the levels of select volatile aldehydes in smokeless tobacco products: formaldehyde, 3.9–6.8 µg/g in moist snuff and 1.6–7.4 µg/g in dry snuff; acetaldehyde, 2.4–7.4 µg/g in moist snuff and 1.4–3.9 µg/g in dry snuff; and crotonaldehyde, 1.0–2.4 µg/g in moist snuff and 0.2–0.6 µg/g in dry snuff.

Uranium was reported in five samples of Indian snuff at a concentration of about 3 pCi/g tobacco (Sharma et al., 1985). Hoffmann et al. (1987) reported 0.16–1.22 pCi/g polonium-210 in commercial moist snuff and 0.23–0.39 pCi/g in commercial dry snuff in the USA.

(f) Effect of storage conditions on the levels of N-nitrosamines

The effect of storage conditions on the formation of TSNA in smokeless tobacco was studied in moist and dry snuff and in chewing tobacco.

In a study of the effects of ageing and storage on the levels of TSNA, N-nitrosamino acids and volatile N-nitrosamines in commercial moist snuff from the USA, it was found that during storage at 4 °C none of these compounds increased significantly (Djordjevic et al., 1993a). However, at higher temperatures, the levels of N-nitrosamines and nitrite in the moist snuff increased significantly over time. After 8 weeks of storage at 37 °C, the levels of NNN and NNK had risen threefold (from 6.24 to 18.7 µg/g), those of the N-nitrosamino acids MNPA and MNBA had risen 5.2-fold (from 3.13 to 16.3 ppm) and those of volatile N-nitrosamines had risen 10-fold (from 0.02 to 0.2 µg/g); moist snuff stored for 8 weeks at 37 °C contained 0.0386 µg/g NDMA, 0.0714 µg/g NPYR and 0.0176 µg/g NMOR. The concentration of 4-(methylnitrosoamino)-1-(3-pyridyl)-1-butanol (NNAL), a metabolite of NNK, doubled during storage at 37 °C from 0.29 to 0.65 µg/g. In a study conducted by the MDPH (Connolly, 2001), the effect of ageing of snuff was examined over 2, 4 and 6 months. Levels of total TSNA, including NNN, NNK, NAT and NAB, in the leading US brand Copenhagen increased 137%. No significant changes were observed in TSNA levels in Ettan, the Swedish Match moist snuff brand, when subjected to storage under adverse conditions. An earlier study revealed that levels of both NNN and NNK in moist snuff increased 21 and 12-fold, respectively, within the first 24 weeks of storage; in contrast, levels of nicotine decreased 1.3-fold during the same period. Concentrations of NNN and NNK in chewing tobacco and dry snuff during 24 weeks of storage increased 1.5- and 1.8-fold, respectively (Andersen et al., 1989).

1.3.4. Kentucky (KY) reference smokeless tobacco products

For research purposes, a series of reference smokeless tobacco products was developed and manufactured by the Tobacco and Health Research Institute (1987) at the University of Kentucky, Lexington, KY (USA) in the late 1980s. Each reference product, i.e. moist snuff, dry snuff and loose-leaf chewing tobacco, was custom made to mimic the chemical composition of commercial products in the respective category. However, specific flavourings and additives, including those used by manufacturers to influence levels of unprotonated nicotine, were not included in KY reference products. KY reference smokeless tobacco products contain the following ingredients:

Loose-leaf chewing tobacco (1S1): Wisconsin air-cured tobacco, 17.4%; Pennsylvania air-cured tobacco, 15.47%; crushed Burley tobacco stems, 5.8%; glycerin, 3.75%; sucrose, 23.01%; dextrose, 1.7%; maltose, 1.3%; other corn syrup solids, 6.21%; salt, 1.6%; sodium propionate, 0.28%; water, 23.48%.

Dry snuff (1S2): dark-fired tobacco, 22.75%; fire-cured Virginia tobacco, 19.66%; air-cured stems, 33.03%; fire-cured stems, 15.2%; salt, 0.36%; water, 9.0%.

Moist snuff (1S3): dark-fired tobacco, 25.73%; air-cured tobacco, 7.83%; Burley stems, 3.73%; sodium carbonate, 0.51%; sodium chloride, 7.4%; water, 54.80%.

As the blending recipe for KY reference products shows, loose-leaf chewing tobacco and moist snuff contain about 30% of tobacco by weight whereas dry snuff contains 75% of tobacco. The chemical composition of these reference products is shown in Table 15. In addition to data on nicotine, total nitrogen, nitrate nitrogen, total sugars, reducing sugars, moisture, pH, ash, potassium, sodium and calcium (Tobacco and Health Research Institute, 1987), the levels of selected TSNA and N-nitrosamino acids are also presented (Djordjevic et al., 1989b; Brunnemann et al., 2002).

Table 15. Chemical composition of Kentucky reference smokeless tobacco products.

Table 15

Chemical composition of Kentucky reference smokeless tobacco products.

1.3.5. Pesticide residues

Maximum allowable limits for pesticides on tobacco (e.g. maleic hydrazide, chlordane, dichlorodiphenyltrichloroethane, dichlorodiphenyldichloroethylene, dieldrin, endrine, heptachlor) in Germany, Italy, Spain and the USA are summarized by Sheets (1990).

1.4. Production, consumption and prevalence of use of smokeless tobacco products

This section presents data on sales, consumption and prevalence of use of smokeless tobacco products. Where possible, data are presented separately for each product type. In some countries and surveys, consumption was not measured or reported separately and thus overall consumption or prevalence of use of smokeless tobacco is reported. In most countries, surveys do not specify which type of snuff is used, but the overwhelming majority of snuff is of the moist variety and is taken orally.

Data on prevalence of smokeless tobacco use among youths in South America (Section 1.4.2(c)), South Asia (Section 1.4.3) and Africa (Section 1.4.4) rely primarily on the Global Youth Tobacco Survey (GYTS). The GYTS project was developed by WHO and the CDC in the USA. It is an international surveillance project designed to enhance the capacity of countries to monitor tobacco use among youths, and to guide the implementation and eva-luation of tobacco prevention and control programmes. The GYTS has been completed in 120 countries. It uses a two-stage cluster sample survey design that produces representative samples of students in grades associated with the ages of 13–15 years.

The prevalence measures used in this study included: current cigarette smoking — defined as ‘The percentage of students who smoked cigarettes on 1 or more days during the past 30 days’ and current other tobacco use — defined as ‘The percentage of students who had used any form of tobacco products other than cigarettes during the past 30 days’. Thus, other tobacco products include smokeless tobacco products as well as other smoking products.

1.4.1. Europe

Trends on sales of chewing tobacco in six European countries and of snuff in 13 countries are given in Tables 16 and 17, respectively. For three of these countries (Austria, Finland and France), the reports combined sales of chewing tobacco and snuff.

Table 16. Sales of chewing tobacco in selected European countries (tonnes).

Table 16

Sales of chewing tobacco in selected European countries (tonnes).

Table 17. Sales of snuff in selected European countries (tonnes).

Table 17

Sales of snuff in selected European countries (tonnes).

For many countries, no data were available on the consumption of smokeless tobacco products. For most countries included in Tables 16 and 17, no additional information was available on the use of smokeless tobacco other than annual sales, and those countries are not listed separately in this section, which includes a discussion of available data for those countries for which data on the prevalence of smokeless tobacco use were available. Estimates of annual per-capita consumption and prevalence of use of smokeless tobacco in these countries are given in Table 18 and 19, respectively.

Table 18. Estimated per-capita consumption of smokeless tobacco in selected European countries (g per person aged ≥ 15 years).

Table 18

Estimated per-capita consumption of smokeless tobacco in selected European countries (g per person aged ≥ 15 years).

Table 19. Prevalence (%) of daily use of moist snuff in three Nordic countries.

Table 19

Prevalence (%) of daily use of moist snuff in three Nordic countries.

(a) Denmark

Sales of chewing tobacco in Denmark have been declining since the early 1900s (Table 16). In 1995, snuff and chewing tobacco comprised 0.5% of all tobacco sales by weight in Denmark (Forey et al., 2002). There are few recent reports on prevalence of use. Among employed men who participated in the Copenhagen Male Study in 1985–86 (mean age, 63 years; range, 53–74 years), an estimated 3.5% reported chewing tobacco or using snuff without smoking (Suadicani et al., 1997).

(b) Finland

Per-capita consumption of moist snuff was relatively constant in Finland from 1970 to 1987 at 6–8 g per person aged 15 years and older, after which it increased to 22–29 g per person for the period 1988–94 (Wicklin, 2005) (Table 18). A 1987 survey of 14–18-year-olds found that use of snuff varied widely among regions in Finland: the proportion of boys who had tried snuff ranged from 17% in eastern and central Finland to 41% in Lapland, and regular use of snuff was reported by 2% of boys in eastern and central Finland, 4% in western Finland, 7% in Uusimaa and 10% in Lapland (Karvonen et al., 1993). Although the proportion of girls who had tried snuff ranged from 5 to 15% among the regions, regular use was reported by no more than 1% in any region.

A survey of 793 first- and second-year students in four senior high schools in the Turku region (mean age, 16.6 years) was conducted in December 1994, before the ban on snuff sales was enacted in Finland on 1 March 1995 (Merne et al., 1998). A cross-sectional survey was conducted in the same schools 1 year later, in December 1995. The study showed a prevalence of snuff use of 9% in 1994 (19% of the boys and 1% of the girls) and of 8% in 1995 (sex-specific prevalence not reported). Of students who reported the use of snuff before the ban, 10% reported to have quit because of the ban, 20% reported reducing their use, 12% reported switching to cigarettes and 5% reported switching to other drugs.

In 2002, 1.2% of adult men in Finland used snuff daily, and prevalence of daily use was highest among men aged 25–34 years (2.3%) (Patja & Vartianen, 2003). Occasional use of snuff was reported by 6% of men and boys aged 15–24 years. Among women, 1.1% used snuff occasionally and 0.6% used it daily. Among 16-year-olds, 3.3% of boys used snuff daily and 9% reported occasional use; 1% of girls used snuff occasionally but none reported daily use.

(c) Norway

Data on use of snuff in Norway has been collected by Statistics Norway since 1985 (Kraft & Svendsen, 1997). Most recent data from national surveys indicate that, in 2004–05, 10% of boys and men aged 16–74 years used snuff: 5% used it daily and 5% occasionally (Directorate of Health and Social Affairs, 2006a).

Among boys and men aged 16–24 years, the prevalence of daily or occasional snuff use increased from 9% in 1985 to 15% in 1994 (Kraft & Svendsen, 1997), and to 33% in 2004–05, the highest of any age group (Table 20) (Directorate of Health and Social Affairs, 2006a). Between 1983 and 2001, the prevalence of daily smoking by boys and men aged 16–24 years remained relatively constant at 28–32% (Directorate of Health and Social Affairs, 2006b), which suggests that the rise in snuff use among young men in Norway was not accompanied by a decline in smoking. The prevalence of occasional or daily use of snuff in men aged 16–44 years has increased steadily since 1988 and has more than doubled between 1985 and 2003 (Directorate of Health and Social Affairs, 2006a). Among men aged 65–74 years, the prevalence of daily or occasional snuff use declined from 12 to 6% between 1985 and 1994 (Kraft & Svendsen, 1997) and was 1% in 2004–05 (Directorate of Health and Social Affairs, 2006a).

Table 20. Prevalence (%) by age of men who use snuff in Norway, 2004–2005.

Table 20

Prevalence (%) by age of men who use snuff in Norway, 2004–2005.

The National Council on Tobacco and Health of Norway conducts surveys of tobacco use among Norwegian lower secondary school youths corresponding to the ages of 13–15 years (Braverman et al., 2001; Directorate of Health and Social Affairs, 2003, 2006c). The surveys are administered at 5-year intervals, and items on snuff were added in 1985. Experimental use and current use of snuff declined between 1985 and 1990 for boys and girls and then increased slightly between 1990 and 2000 (Table 21). In 2000, 15.7% of boys and 1.9% of girls aged 13–15 years reported current snuff use. The prevalence of current use increased with increasing grade in school in both sexes.

Table 21. Prevalence (%) by sex, grade and year of survey of youths in Norway who have tried snuff or use it currently.

Table 21

Prevalence (%) by sex, grade and year of survey of youths in Norway who have tried snuff or use it currently.

Cross-sectional surveys of military personnel in the late 1980s found a very high prevalence of snuff use relative to the general male population. A 1986 survey of Norwegian Army conscripts found that 33% used snuff (10% daily and 23% occasionally); 82% of snuff users also smoked (Schei et al., 1990). Similarly, prevalence was relatively high among military officers in a 1989 survey, with 23% reporting current use (15% daily and 8% occasionally) (Schei, 1992).

(d) Sweden

After declining from 6500 tonnes to 2500 tonnes between 1920 and 1967, annual sales of moist snuff (snus) in Sweden increased back to 6800 tonnes in 2002 (Table 17). Accordingly, per-capita consumption of moist snuff between 1970 and 2004 increased steadily from 393 to 922 g per person (Table 18) (Wicklin, 2005).

The most recent official Swedish national survey on the prevalence of moist snuff use among adults was conducted in 2004–05. In 1996–97, 20.0% of men and 0.9% of women aged 16–84 years used moist snuff daily and 5.4% of men and 1.1% of women used it occasionally (Wicklin, 2006) (Table 22). The prevalence of daily moist snuff use among men increased from 16.7% in 1988–89 to 20.0% in 1996–97 and 22% in 2004.

Table 22. Prevalence (%) of use of snus among persons aged 16–84 years in Sweden, 1980–1997 (SCB/ULF surveys).

Table 22

Prevalence (%) of use of snus among persons aged 16–84 years in Sweden, 1980–1997 (SCB/ULF surveys).

The most recent age-specific official data on moist snuff use among men in Sweden (Wicklin, 2006) show that, in 2004, the prevalence of daily moist snuff use was highest among men aged 35–44 years (29%) and lowest among men aged 75–84 years (6%). This pattern is slightly different from that seen in 1988–89 and 1996–97 (Table 23), when the prevalence was highest among men aged 25–34 years and was lowest among men aged 55–64 years and 65–74 years, respectively. Between 1989 and 2004, the prevalence among men aged 16–24 years remained at 21–23% and increased among men aged 35–64 years. Moist snuff use was most prevalent among skilled and unskilled workers of all occupational groups; the survey showed some regional variations of moist snuff use.

Table 23. Prevalence (%) by selected demographic characteristics of daily use of snus among men aged 16–84 years in Sweden, 1988–89 and 1996–97 (SCB/ULF surveys).

Table 23

Prevalence (%) by selected demographic characteristics of daily use of snus among men aged 16–84 years in Sweden, 1988–89 and 1996–97 (SCB/ULF surveys).

Although unofficial trends of the prevalence of moist snuff use in Sweden are also available from the mail-based TEMO surveys conducted by the Statistical Bureau VECAHB and sponsored by the Swedish Match Company (Wicklin, 2006). The sample size in 2004 was about 12 000, but response rates were not reported; 20.4% of men aged 16–75 years reported daily use of snuff and 4.0% reported occasional use (Table 24). Daily moist snuff use was reported by 3.4% of women in that age range and 2.9% of women reported using it occasionally.

Table 24. Prevalence (%) of use of snus by sex in Sweden, 1970–2004 (TEMO surveys).

Table 24

Prevalence (%) of use of snus by sex in Sweden, 1970–2004 (TEMO surveys).

Data on daily use of moist snuff among young people in Sweden have been collected since 1981 during the School Children's Drug Habits surveys conducted by the Swedish Council for Information on Alcohol and Other Drugs (Wicklin, 2006). The prevalence of daily snuff use in 2003 was 2% among 12–13-year-old boys and was not reported for girls at that age. Among students aged 15–16 years, daily use of moist snuff remained relatively constant among boys aged 15–16 years, in the range of 11–14% until 1998, with a possible trend toward increasing moist snuff use in more recent years. Moist snuff use among 15–16-year-old girls remained relatively constant over time and was in the range of 0–2%.

(e) United Kingdom

The use of chewing tobacco is relatively rare in the general population of the United Kingdom, although use of various forms of oral tobacco is common in some immigrant communities in the form of chewing betel quid with tobacco.

1.4.2. North and South America

Data on sales of snuff and chewing tobacco are available for Canada and the USA (Table 25). The USA are the leading producer of snuff worldwide, and have experienced substantial increases in sales of snuff in recent decades, from 10 840 tonnes in 1980 to 33 520 tonnes in 2003 (209% increase) (Forey et al., 2002; Department of Agriculture, 2003).

Table 25. Sales of chewing tobacco and snuff in North America (tonnes).

Table 25

Sales of chewing tobacco and snuff in North America (tonnes).

Estimates of annual per-capita consumption of smokeless tobacco are available for the USA only (Table 26).

Table 26. Estimated per-capita consumption of smokeless tobacco in the USA (g per person aged ≥ 15 years).

Table 26

Estimated per-capita consumption of smokeless tobacco in the USA (g per person aged ≥ 15 years).

(a) Canada

Recent Canadian national data on consumption of smokeless tobacco and prevalence of use are reported only in aggregate and not by product type (Table 25). Sales of smokeless tobacco products in Canada have remained relatively constant from 1989 to 2003 other than some fluctuation in 2000 and 2001 (Tobacco Control Programme, 2004). In 1992–97, chewing tobacco generally accounted for about 20–30% of the smokeless tobacco market by weight; the majority of the market was snuff (Wyckham, 1999).

A 1994 survey on smoking in Canada found that about 1% of the male population aged 15 years and older used smokeless tobacco products currently, which was unchanged from the 1986 prevalence (Wyckham, 1999). In 1986, use of chewing tobacco was slightly more prevalent (0.7%) than that of snuff (0.4%) among men aged 15 years and older, with a prevalence of 1.8% for those over 65 years of age (Millar, 1989). Use of chewing tobacco was slightly more prevalent among men in the Atlantic region (2.0%) and Prairies (1.1%) than in other regions. More recently, the Canadian Tobacco Use Monitoring Surveys enquired whether respondents had ever tried chewing tobacco, pinch or snuff; in 1999–2003, 13–15% of men and 2–3% of women aged 15 years and older reported ever trying these products (Tobacco Control Programme, 2004).

Relatively high use of chewing tobacco and other smokeless tobacco products has been reported among some native populations in some localities, among college athletes and among some young people who use other forms of tobacco. A survey of native Canadians in northern Saskatchewan found that 15% of boys and men and 7% of girls and women aged 7–21 years used chewing tobacco and 23% and 14%, respectively, used snuff (Hoover et al., 1990). Similar findings were reported from a 1987 survey of 5–19-year-olds in the Canadian Arctic, in which 11% of boys and 2% of girls currently used chewing tobacco and 13% and 5%, respectively, used snuff; the prevalence was more than 10 times higher among Dene or Métis and Inuit children than among non-native children (Millar, 1990). In a 1989 random telephone survey of boys and girls aged 11–19 years in northeastern Ontario, 4.5% of respondents reported current use of chewing tobacco and 1.1% used snuff (Blackford et al., 1994). In a longitudinal panel survey in Calgary, the prevalence of smokeless tobacco use was 1.1% in grade 6 (boys, 1.6%; girls, 0.7%), 2.2% in grade 7 (boys, 3.1%; girls, 1.1%) and 4.2% in grade 8 (boys, 6.9%; girls, 1.6%) (Abernathy & Bertrand, 1992).

A survey of 754 athletes at 10 English Canadian universities found that, among men, smokeless tobacco was used by 47.2% of hockey players, 36.2% of football players, 22.0% of soccer players, 12.0% of volleyball players and 6.8% of track or cross-country athletes (Spence & Gauvin, 1996). Use among female university athletes in some sports was relatively high compared with the general Canadian population, including track or cross-country (6.2%), basketball (4.0%), soccer (3.3%) and volleyball (2.4%). A 1987 survey of Ontario students aged 13–19 years found that smokeless tobacco use was uncommon in the general student population (2.6% of boys, 0.6% of girls), but the prevalence was relatively high among students who currently smoked (9.6% of boys, 1.9% of girls) (Adlaf & Smart, 1988).

(b) USA

Information on sales and per-capita consumption of chewing tobacco and snuff in the USA between 1920 and 2003 are presented in Tables 25 and 26, respectively.

In 2000, 4.4% of men and 0.3% of women in the USA were current users of smokeless tobacco products (Table 27). Current use was more common among men aged 18–44 years (5.0–5.8%) than among men aged 45 years and older (2.8–3.1%). Non-Hispanic white men were more likely to be current users (5.5%) than were men in other racial or ethnic groups (0.8–2.2%), although the sample size was insufficient to permit meaningful national estimates for some racial and ethnic groups that may have high levels of use, such as American Indians. The prevalence of smokeless tobacco use was higher among men with a high school education or less (5.6–5.7%) than among those with at least some post-high school education (3.4%). Prevalence of smokeless tobacco use was higher among men in the South (6.7%) than in all other geographic regions (2.2–4.4%), and was much higher among men who lived outside of metropolitan statistical areas (9.0%) than among urban men (3.3%) (Tomar, 2003a).

Table 27. Prevalence by selected demographic characteristics of current use of smokeless tobacco in the USA, 2000 (National Health Interview Survey).

Table 27

Prevalence by selected demographic characteristics of current use of smokeless tobacco in the USA, 2000 (National Health Interview Survey).

Based on combined unpublished data from the January and May 2000 Current Population Survey Tobacco Use Supplements, the prevalence of smokeless tobacco use among adult men in 2000 was highest in West Virginia (13.9%), Montana (13.1%), Wyoming (13.3%), Mississippi (9.4%) and Tennessee (9.2%) and lowest in Massachusetts (0.2%), Rhode Island (0.5%), New Jersey (0.6%), Connecticut (1.0%) and Hawaii (0.8%).

Similar to the pattern observed among adults, adolescent smokeless tobacco users in the USA are predominantly boys. In 1986–87, 6.1% of boys and 0.1% of girls aged 12–17 years reported current use of smokeless tobacco (Tomar et al., 1997). All recent national surveys of young people report the prevalence of ‘smokeless tobacco’ use, which includes snuff or chewing tobacco, and generally do not enquire about the products separately. The prevalence of current smokeless tobacco use among male high-school students has declined from 20.4% in 1993 (Kann et al., 1995) to 11.0% in 2003 (Grunbaum et al., 2004) (Table 28). Use of smokeless tobacco ranged from 9.1 to 13.3% for boys in grades 9 to 12, while among high-school girls it ranged from 1.0 to 3.8% across grades. The prevalence was substantially higher among male non-Hispanic white students (13.2%) than among male Hispanic (6.1%) or non-Hispanic black students (4.1%) (Grunbaum et al., 2004). A nationally representative cohort study conducted in the early 1990s estimated that, each day, 2200 young people in the USA first try smokeless tobacco and about 830 become regular users (Tomar & Giovino, 1998).

Table 28. Prevalence by sex, race or ethnicity and grade of use of smokeless tobacco among high school students in the USA, 2003 (Youth Risk Behavior Survey).

Table 28

Prevalence by sex, race or ethnicity and grade of use of smokeless tobacco among high school students in the USA, 2003 (Youth Risk Behavior Survey).

(i) Chewing tobacco

Detailed data for production of the three major forms of chewing tobacco in the USA during 1981–2003 are presented in Table 29. Loose-leaf chewing tobacco remained the predominant form throughout that period, and comprised 94% of the chewing tobacco market by weight. However, production declined for all chewing tobacco products during that time. Per-capita consumption followed the same trend, declining by 68% from 273 g per person aged 15 years or older in 1980 to 86 g per person in 2002 (Table 26).

Table 29. Production of chewing tobacco in the USA, by major category.

Table 29

Production of chewing tobacco in the USA, by major category.

Use of chewing tobacco in the USA is primarily practised by men although there are examples of subpopulations of women in which use is relatively prevalent, particularly some American Indian and Alaskan Native communities (Schinke et al., 1987; Lanier et al., 1990). Prevalence of tobacco chewing appears to be declining in the USA after having reached a peak of 4.1% in 1987 (Table 30). In 2000, current use of chewing tobacco was reported by 2.5% of men and 0.1% of women; it tended to be slightly higher for men aged 25–34 years than in other age groups. Table 31 presents more detailed characteristics of the prevalence of use of different types of smokeless tobacco among men in 2000 in the USA (unpublished data from the 2000 National Health Interview Survey). Use of chewing tobacco was more prevalent among non-Hispanic white men than among other racial or ethnic groups, among men with less than a high school education than among more educated men, and among rural men than among urban men. Approximately one-half of the men who reported current use of chewing tobacco used those products on a daily basis.

Table 30. Prevalence (%) by sex and age of current use of chewing tobacco or snuff among adults in the USA.

Table 30

Prevalence (%) by sex and age of current use of chewing tobacco or snuff among adults in the USA.

Table 31. Prevalence (%) by selected demographic characteristics of current use of smokeless tobacco among men aged 18 years and older in the USA, 2000 (National Health Interview Survey).

Table 31

Prevalence (%) by selected demographic characteristics of current use of smokeless tobacco among men aged 18 years and older in the USA, 2000 (National Health Interview Survey).

(ii) Snuff

Moist snuff is the predominant form of snuff sold in the USA. It comprised 95% of the snuff market by weight in 2001 (Federal Trade Commission, 2003). Sales of dry snuff declined steadily from 3678.7 tonnes in 1986 to 1526.2 tonnes in 2001, while moist snuff sales increased gradually from 16 391.0 to 28 980.0 tonnes during that period (Tables 25 and 32). Except for a slight decline in the mid-1980s, per-capita consumption of snuff (moist and dry) in the USA has increased every year since 1981 except for some decline in 1986–89 (Table 26).

Table 32. Sales of snuff in the USA, by category (tonnes).

Table 32

Sales of snuff in the USA, by category (tonnes).

Trends in the prevalences of use of smokeless tobacco between 1970 and 2000 by sex and age are given in Table 30. This includes only individuals who do not smoke cigarettes, but who may smoke cigar or pipes.

In 2000, current snuff use was highest among men aged 18–24 years (3.6%) and was more prevalent than chewing tobacco use among men aged 18–44 years (Table 31). Snuff use was more prevalent among men in southern regions of the USA than in other regions, among men with a high school diploma or equivalent than among those with a higher education, among non-Hispanic whites than among other racial or ethnic groups and among men who resided outside of metropolitan statistical areas (i.e. primarily rural areas) than those who lived in metropolitan areas. About 60–65% of men who were current snuff users used those products on a daily basis, except for the youngest age group (18–24 years) (unpublished data from the 2000 National Health Interview Survey).

(iii) Population groups with a high prevalence of use

There are groups within the USA with exceptionally high prevalences of use of smoke-less tobacco. A review of studies of professional baseball players conducted between 1987 and 1998 reported a prevalence of smokeless tobacco use of 35–46%, including both chewing tobacco and snuff (Greene et al., 1998), although snuff is used much more commonly than chewing tobacco among this group (75%–90% of current users). In another study of professional baseball players conducted in 1988 (Ernster et al., 1990), 42% of players reported current use of any type of smokeless tobacco and, among users, 43% reported using both. In a study conducted in 1999, 31% of professional baseball players reported current use of smokeless tobacco, 82% of whom were using snuff (Cooper et al., 2003).

High rates of smokeless tobacco use have also been reported among college athletes (Levenson-Gingiss et al., 1989; Walsh et al., 1994; Hannam, 1997; Green et al., 2001). More than 20% of National Collegiate Athletic Association student athletes reported current use of smokeless tobacco in 1996, with a range of 6–55% among male teams and 1–22% among female teams (Green et al., 2001). Those studies of college athletes that examined product type reported that exclusive snuff use was more common than exclu-sive chewing tobacco use, but dual use of products was common (Levenson Gingiss & Gottlieb, 1991; Walsh et al., 1994; Chakravorty et al., 2000).

Elevated use of smokeless tobacco has also been reported among high-school athletes compared with non-athletes (Davis et al., 1997; Melnick et al., 2001; Castrucci et al., 2004). Similar to patterns among college athletes, high-school baseball players who used smokeless tobacco were much more likely to use exclusively snuff (40%) or to use both snuff and chewing tobacco (52%) than to use exclusively chewing tobacco (8%) (Walsh et al., 2000).

There are indications that the prevalence of smokeless tobacco use is also relatively high among military personnel in the USA and ranges from 15 to 46% (Ballweg & Bray, 1989; Forgas et al., 1996; Kenny et al., 1996; Grasser & Childers, 1997; Chisick et al., 1998; Kao et al., 2000). The few studies that examined product type reported at least a 3:1 ratio of use of snuff to use of chewing tobacco (Kenny et al., 1996; Grasser & Childers, 1997).

(c) South America

Limited information is available from the GYTS on the prevalence of the use of smokeless tobacco and non-cigarette tobacco for selected countries in Latin America (Table 33).

Table 33. Prevalence (%) of use of smokeless and non-cigarette tobacco products among 13–15-year-old schoolchildren in Latin America (Global Youth Tobacco Survey).

Table 33

Prevalence (%) of use of smokeless and non-cigarette tobacco products among 13–15-year-old schoolchildren in Latin America (Global Youth Tobacco Survey).

In a cross-sectional survey in schools in Venezuela, chimó was used by 13.5% of boys in grades 6–9 (∼13–16 years old), including 10% of boys in grade 6 (Granero et al., 2003).

1.4.3. South Asia

The prevalence of smokeless tobacco use is high in South Asia. Also, new forms of smokeless tobacco have been emerging over the last few decades to entice new con-sumers. Increasing use has been reported not only among men, but also among children, teenagers, women of reproductive age and immigrants of South Asian origin wherever they have settled (Gupta, 1992). In the WHO South-East Asia Region, over 250 million people use smokeless tobacco products, which represents 17% of the total population; of those, 95% live in India (82%) or Bangladesh (13%) (Sinha, 2004).

Data from national or sub-national surveys or data from studies with large sample sizes are presented in this section.

(a) Bangladesh

Zarda, khiwam and gul are manufactured in Bangladesh and are also imported from India (Sinha, 2004).

In Bangladesh, 20–30% of women in rural areas are estimated to use smokeless tobacco, predominantly as part of a betel quid (Islam & Al-Khateeb, 1995). Among 638 respondents in a community-based intervention study on non-communicable diseases, 26% reported chewing tobacco products. Among users, 85% chewed daily and 15% occasionally (Table 34) (Rahman et al., 2001).

Table 34. Prevalence of chewing in Dhaka Metropolitan City, Bangladesh.

Table 34

Prevalence of chewing in Dhaka Metropolitan City, Bangladesh.

Among 11 409 respondents in a baseline community-based health behaviour surveillance study conducted in rural and urban areas, 169 (1.5%) reported current use of gul (urban 2%, rural 0.5%); application of gul was reported most frequently (5.2%) by urban women of lower socioeconomic classes. In addition, four people reported use of snuff (Rahman et al., 2004).

A cross-sectional survey conducted among tobacco users in selected population groups in Bangladesh in 2003 showed use of treated tobacco leaf by 41.9%, raw dried tobacco leaf by 17.4% and powdered tobacco leaf by 3.9% (Table 35).

Table 35. Type of tobacco product used among tobacco users in Bangladesh, 2003.

Table 35

Type of tobacco product used among tobacco users in Bangladesh, 2003.

Among rickshaw pullers, 42.7% reported applying gul and 45% used betel quid with tobacco (Sinha, 2004).

(b) Bhutan

Tobacco consumption in Bhutan has changed from smoking to other forms such as oral use. Despite a total ban of sales of tobacco in Bhutan, packages of zarda used to be on sale in the Thimphu vegetable market. Many people, including young boys and monks, chew zarda and scented khaini (Sinha, 2004). A recent study showed that 8% of people in Bhutan chew or sniff tobacco (7% women, 10% men). Smoking prevalence is estimated to be 1% (Ugen, 2003).

(c) India

India is one of the major producers of chewing tobacco in Asia. A specific feature of tobacco production in India is the variety in the types of tobacco produced. The presence of a strong domestic demand for tobacco product for chewing and application to a relatively large extent influences the cultivation of tobacco for these uses. Tobacco used for chewing and application is grown in Tamil Nadu, Uttar Pradesh, Bihar, West Bengal and Orissa (Reddy & Gupta, 2004).

In 2002, 40.6% of the tobacco production was used in cigarettes, 33.3% in bidi pro-duction and 12.4% was used for smokeless forms of chewing, snuffing and applied tobacco (Table 36; Reddy & Gupta, 2004). Between 1976 and 1994, chewing tobacco production represented between 11% and 19% of total tobacco production, but production has increased substantially since 1995 (Table 37). In 2002, 65 million kg of chewing tobacco and 10 million kg of snuff tobacco were produced in India (Table 36). This increase was accompanied by a huge growth in the export of both chewing tobacco (9-fold increase between 1995 and 2005) and snuff tobacco (18-fold increase during the same period) (Table 38; Reddy & Gupta, 2004; Tobacco Board, 2006).

Table 36. Tobacco production by type of tobacco in India, 2002.

Table 36

Tobacco production by type of tobacco in India, 2002.

Table 37. Production of smokeless tobacco (in millions of kg) in India (derived estimates) and percentage of total tobacco production.

Table 37

Production of smokeless tobacco (in millions of kg) in India (derived estimates) and percentage of total tobacco production.

Table 38. Exports of tobacco from India by product (in tonnes).

Table 38

Exports of tobacco from India by product (in tonnes).

Large variations in the prevalences and patterns of smokeless tobacco use occur across the country. Apart from regional preferences due to different socio-cultural norms, the preference for smokeless rather than smoked tobacco is inversely related to education and income (Gupta, 1996). Per-capita consumption of smokeless tobacco has increased among the lower socioeconomic classes between 1961 and 2000 in both rural and urban areas (data from the National Sample Survey Organization, cited in Gupta & Ray, 2003).

Six sets of data may allow an estimation of the prevalence of smokeless tobacco use in India: (a) large sub-national cross-sectional and cohort studies, (b) the National Family Health Survey, (c) the WHO sub-national study, (d ) the National Sample survey on household consumer expenditure, (e) the Global Youth Tobacco Survey and ( f ) the Sample Registration system (unpublished). The last set of data is not discussed here.

(i) Sub-national cross-sectional and cohort studies

It has been estimated that approximately one-third of women and two-thirds of men in India use tobacco in one form or another (WHO, 1997). In prevalence surveys in 10 rural areas in eight states of India, smokeless tobacco was used by 3–53% of men and 3–49% of women (Table 39). In these areas, 2–26% of men and 0–4% of women also reported both smoking and smokeless tobacco use (Gupta & Ray, 2003). In a cross-sectional and cohort study in Mumbai, the prevalence of smokeless tobacco use in 1992–94 was 57.1% among women and 45.7% among men (Gupta, 1996). In another cross-sectional survey in a suburb of Trivandrum, Kerala, where residents were mostly of lower socioeconomic status, chewing practices were reported by 26.8% of men (n = 25 453) and 26.4% of women (n = 34 441), mainly in the form of betel quid with tobacco (Sankaranarayanan et al., 2000).

Table 39. Prevalence (%) of use of various types of tobacco in 10 areas in eight states of India.

Table 39

Prevalence (%) of use of various types of tobacco in 10 areas in eight states of India.

(ii) National Family Health Survey

In the National Family Health Survey-2 conducted in 1998–99, 315 597 individuals aged 15 years or older from 91 196 households were sampled (Rani et al., 2003). Among the study population, 20% (28.1% of men and 12.0% of women) reported chewing tobacco/pan masala; however, the prevalence may be underestimated by almost 11% for men and 1.5% for women because of the use of household informants. The prevalence of chewing tobacco/pan masala varied significantly (7–60%) between states (Table 40). Chewing of tobacco/pan masala was relatively more common (> 16%) in the central, eastern, western (except Goa) and northeastern states (except Tripura) compared with the northern and southern states. The prevalence of chewing tobacco/pan masala was significantly higher in rural, poorer and less educated populations compared with urban, wealthier and more educated populations in both men and women. The socioeconomic gradients (household wealth, education) had more impact for women than for men. The prevalence of chewing tobacco/pan masala was higher among tribal populations than among other communities (Table 41). In a multivariate analysis, the older population (≥ 25 years) had a greater likelihood of chewing tobacco compared with the younger population (15–24 years). Muslim women were more likely to chew tobacco than Hindu women, and the Sikh religion emerged as one of the strongest predictors among women for not chewing tobacco. The differentials by state of residence also persisted in the multivariate analysis. No significant association was observed between urban or rural residence and chewing of tobacco/pan masala among men after controlling for other characteristics. However, rural women were less likely to chew tobacco than urban women (Rani et al., 2003).

Table 40. Prevalence of chewing in India by state and by sex (National Family Health Survey, 1998–99).

Table 40

Prevalence of chewing in India by state and by sex (National Family Health Survey, 1998–99).

Table 41. Prevalence (%) of chewing of tobacco/pan masala in India (National Family Health Survey, 1998–99).

Table 41

Prevalence (%) of chewing of tobacco/pan masala in India (National Family Health Survey, 1998–99).

(iii) WHO Sub-national Study

In a WHO study (Chaudhry, 2001), 35 288 respondents in Karnataka and 29 931 res-pondents in Uttar Pradesh (aged ≥ 10 years) were surveyed. Tobacco was predominantly used in smokeless form among women of all ages and among men under 30 years of age, both in urban and rural areas. The overall prevalence of current use of smokeless tobacco was 13.9% in Karnataka (13.4% among men, 14.4% among women) (Table 42) and 17.5% in Uttar Pradesh (24.3% among men, 6.6% among women) (Table 43). In Karnataka, the prevalence of use of smokeless tobacco was higher among women compared with men in the age groups above 40 years. In Uttar Pradesh, the proportion of men who used smokeless tobacco was higher than that of women in all age groups. In both regions, prevalence of smokeless tobacco use by women increased with age; for men, prevalence was highest in the age groups 25–29 years and above 70 years. Trends were similar in urban and rural areas. The prevalence of smokeless tobacco use was generally lower among educated women, especially in Karnataka. Clear-cut trends in reduced prevalence with increasing education were not observed in all age groups among men. Muslim men in Karnataka showed a higher overall prevalence compared with Hindus, while in Uttar Pradesh, a higher proportion of Hindu men compared with Muslims used smokeless tobacco. The reverse trend was observed among women in the two states [data for other religions were based on too few numbers to be reliable]. Variations in prevalence according to family income did not follow any specific trend, but the prevalence was comparatively lower in both states among women with higher family income (Chaudhry, 2001).

Table 42. Prevalence by rural/urban area, age and sex of current use of smokeless tobacco in Karnataka, India (WHO Sub-national Study, 2001).

Table 42

Prevalence by rural/urban area, age and sex of current use of smokeless tobacco in Karnataka, India (WHO Sub-national Study, 2001).

Table 43. Prevalence by rural/urban area, age and sex of current use of smokeless tobacco in Uttar Pradesh, India (WHO Sub-national Study, 2001).

Table 43

Prevalence by rural/urban area, age and sex of current use of smokeless tobacco in Uttar Pradesh, India (WHO Sub-national Study, 2001).

(iv) National Sample Survey Organisation

The National Sample Survey Organisation conducted its fifth quinquennial nation-wide survey of household consumer expenditure in India during 1993–94. Interviews were conducted in 115 354 households in 6951 villages and in 4650 urban blocks. Pre-valence of use of chewing tobacco was 11.2% and 6.3% among men in rural and urban areas, respectively, and 3.9% and 2.0%, respectively, for women (Table 44) (National Sample Survey Organisation, 1998). The prevalence of tobacco use was underestimated because only one household respondant answered for all inhabitants of the household.

Table 44. Prevalence (%) by rural/urban area and sex of use of tobacco in various forms in India (National Sample Survey Organisation).

Table 44

Prevalence (%) by rural/urban area and sex of use of tobacco in various forms in India (National Sample Survey Organisation).

Comparison of the data from 1987–88 and 1993–94 (National Sample Survey Organisation, 1998; Gupta & Sankar, 2004) revealed no significant change in overall use of smokeless tobacco during this period (Table 44). Other reports suggest that there has been a shift towards use by younger people and at a very early age. For example, the prevalence of mawa use rose from 4.7% in 1969, mainly among older women, to 19% in 1993–94, mainly among younger generations (Gupta, 2000); in a survey conducted among rural children in three states, snuff was ever used by 38% of boys and 12% of girls aged 10–20 years (Krishnamurthy et al., 1997); a comparative study of the prevalence of tobacco use in a rural area in Bihar showed that the prevalence of total tobacco use remained the same between 1967 and 2000, but that there had been a remarkable shift towards the use of smokeless tobacco (Sinha et al., 2003a).

(v) Global Youth Tobacco Survey (GYTS) and Global School Personnel Survey (GSPS)

The Global School Personnel Survey (developed by WHO/CDC) is a cross-sectional survey that employs a cluster sample design to produce a representative sample of school personnel drawn from the same schools that were selected for GYTS. All school personnel (including non-teaching staff) in the selected schools were eligible to participate (Sinha et al., 2002).

In the eight northeastern states of India, daily use of smokeless tobacco among school personnel varied from 8.9 (Sikkim) to 49.4% (Mizoram) among men and from 1.6 (Manipur) to 80.3% (Mizoram) among women (Table 45) (Sinha et al., 2003b). In five of the eight states, the prevalence of daily use of smokeless tobacco among men and women was similar. In the eastern region, daily use of smokeless tobacco among school personnel varied from 7.8 (West Bengal) to 58.7% (Bihar) in men and from 1.0 (West Bengal) to 53.4% (Bihar) in women (Sinha et al., 2002, 2003b; Sinha & Gupta, 2004a; Sinha & Roychoudhury, 2004). The prevalence of use of each type of products is detailed in Table 46.

Table 45. Prevalence (%) by state of current use of smokeless tobacco among school personnel in the northeastern and eastern regions of India (Global School Personnel Survey, 2001).

Table 45

Prevalence (%) by state of current use of smokeless tobacco among school personnel in the northeastern and eastern regions of India (Global School Personnel Survey, 2001).

Table 46. Prevalence (%) by type of tobacco product of smokeless tobacco use by school personnel in the northeastern states of India (Global School Personnel Survey, 2001).

Table 46

Prevalence (%) by type of tobacco product of smokeless tobacco use by school personnel in the northeastern states of India (Global School Personnel Survey, 2001).

Smokeless tobacco use among students varied between states from 2.8 (Goa) to 55.6% (Bihar) (Table 47). Among boys, it varied from 2.7 (Delhi) to 57.6% (Bihar) and, among girls, from 2.1 (Goa) to 49.2% (Bihar). In 11 of 13 states, prevalences of smokeless tobacco use among boys and girls were similar; boys in Meghalaya and Uttaranchal reported significantly more smokeless tobacco use than girls (Arora et al., 2001; Sinha & Gupta, 2002a,b; Sinha et al., 2003c; Pednekar & Gupta, 2004; Sinha & Gupta, 2004b; Sinha et al., 2004a).

Table 47. Prevalence (%) by state of current use of smokeless tobacco among students in India (Global Youth Tobacco Survey).

Table 47

Prevalence (%) by state of current use of smokeless tobacco among students in India (Global Youth Tobacco Survey).

The use of tobacco products as dentrifice among students aged 13–15 years varied widely between states (Table 48). The prevalence among boys compared with that among girls was notably higher in Orissa and Uttaranchal, marginally higher in nine states and marginally lower in three states. Of the specific products, tobacco toothpaste (creamy snuff) and tooth powder (lal dant manjan) were common in all 14 states; the prevalence of use ranged from 2 to 32% and from 2 to 29%, respectively. Gul was used in eight states and the prevalence of its use ranged from 2 to 6%. Other dentifrice products containing tobacco were: mishri and bajjar in Goa and Maharashtra; gudhaku in Bihar, Orissa, Uttar Pradesh and Uttaranchal; and tobacco water (tuibur) in Manipur, Mizoram, Sikkim and Tripura (Sinha et al., 2004b).

Table 48. Prevalence (%) of application of tobacco products as dentifrice in 14 states in India (Global Youth Tobacco Survey, 2000–2002).

Table 48

Prevalence (%) of application of tobacco products as dentifrice in 14 states in India (Global Youth Tobacco Survey, 2000–2002).

The current use of smokeless tobacco among the participants of the GSPS (Sinha et al., 2003b) and GYTS (Sinha et al., 2003c) surveys in eight of the states is detailed below.

Arunachal Pradesh

In Arunachal Pradesh, betel quid was the most popular form of smokeless tobacco among men (73.6%) and women (51.4%). Khaini was used exclusively by men (19.8%), while tuibur (32.6%) and snuff (15.2%) were used exclusively by women (Sinha et al., 2003b).

Current use of smokeless tobacco was reported by 37.2% of students (35.0% of boys, 40.2% of girls), whereas smoking was reported by 22.8% (31.8% of boys, 8.3% of girls). Smokeless tobacco use exclusively in the form of chewing was reported by 55.2% and use exclusively in the form of application was reported by 28.8%. The remainder used several forms of smokeless tobacco. Among chewers, gutka was the most popular product (49.8%), followed by tamol and a tobacco mixture (31%). Among appliers, 79.7% applied tobacco toothpaste, 12.3% applied red tooth powder and 8% applied gul (Sinha et al., 2003c).

Assam

In Assam, the most popular form of smokeless tobacco use among men was betel quid (75.5%), followed by khaini (7.9%) and gutka (7.8%). Among women, betel quid (36.3%) was commonest, followed by tuibur (35.7%), gul (13.5%) and gutka (13.4%). Gul and tuibur were used primarily by women (Sinha et al., 2003b).

Current use of smokeless tobacco was reported by 25.3% of students (29.3% of boys, 20.4% of girls). Current smoking was reported by 19.7% of students (28.6% of boys, 8.9% of girls). Smokeless tobacco use exclusively in the form of chewing was reported by 48.5% and that exclusively in the form of application by 18.8%. The remainder used several forms of smokeless tobacco. Among chewers, gutka was the most popular product (54.4%), followed by tamol and a tobacco mixture (28.9%). Among appliers, 58.5% applied tobacco toothpaste, 25% applied red tooth powder and 16.3% applied gul (Sinha et al., 2003c).

Manipur

In Manipur, betel quid (54.7%) was the most popular form of smokeless tobacco among both men and women. The prevalence of gutka use among men was higher (24.1%) than that among women (2.6%), while the prevalence of khaini use among women was higher (29.1%) than that in men (8.9%). Tuibur was used predominantly by women (27.5%) (Sinha et al., 2003b).

Current smokeless tobacco use was reported by 46.1% (51.5% of boys, 40.1% of girls), whereas smoking was reported by 26.8% (40.8% of boys, 10.7% of girls). Smokeless tobacco use exclusively in the form of chewing was reported by 53.2% and that exclusively in the form of application by 31.9%. The remainder used several forms of smokeless tobacco. Among chewers, gutka (23.7%) was the most popular product (17.9% of boys, 30.2% of girls), followed by tamol and a tobacco mixture (18.1% overall, 28.0% of boys, 6.8% of girls). Among appliers, 18.3% of boys and 32.6% of girls applied tobacco toothpaste (Sinha et al., 2003c).

Meghalaya

In Meghalaya, betel quid (55.4%) was the most popular form of smokeless tobacco, followed by tuibur (13.1%), gul (12.0%) and khaini (9.1%). Gul and tuibur were used primarily by women (Sinha et al., 2003b).

Current smokeless tobacco use was reported by 35.3% (43.0% of boys, 26.8% of girls), whereas smoking was reported by 21.4% (32.1% of boys, 9.9% of girls). Smokeless tobacco use exclusively in the form of chewing was reported by 55.2% (62.1% of boys, 47.7% of girls) and that exclusively in the form of application by 22.9% (28% of boys, 17.6% of girls). The remainder used several forms of smokeless tobacco. Chewing was mainly in the form of gutka (19.4%), tamol with tobacco (9.2%, > 80% of boys) and tamol without tobacco (21%). Tobacco was applied by 18.2% as tobacco toothpaste and by 3.9% as red tooth powder (Sinha et al., 2003c).

Mizoram

In Mizoram, among smokeless tobacco users, 24.8% used gutka, 22.3% used khaini, 20.2% used betel quid, 16.4% used gul and 9.3% used snuff. The use of gutka and snuff was reported slightly more frequently among women while that of betel quid and gul was more frequent among men (Sinha et al., 2003b).

Current smokeless tobacco use was reported by 42.9% (45.7% of boys, 40.1% of girls), whereas current smoking was reported by 34.5% (40.7% of boys, 28.2% of girls). Smokeless tobacco use exclusively in the form of chewing was reported by 60.7% and that exclusively in the form of application by 25.0%. The remainder used several forms of smokeless tobacco. Among chewers, use of gutka (20%) was reported to be the most popular, followed by pan with tobacco (12.9%). Among appliers, the majority preferred tobacco toothpaste (11.8%) (Sinha et al., 2003c).

Nagaland

In Nagaland, betel quid (69.3%), khaini (15.9%) and gutka (8.3%) were the most pre-valent forms of smokeless tobacco used. Betel quid was more common among women, whereas khaini was used almost exclusively by men (Sinha et al., 2003b).

Current smokeless tobacco use was reported by 49.9% (52.5% of boys, 47.2% of girls), whereas smoking was reported by 29.6% (38.7% of boys, 19.7% of girls). Smokeless tobacco use exclusively in the form of chewing was reported by 62.4% and that exclusively in the form of application by 40%. The remainder used several forms of smokeless tobacco. Among chewers, 28.1% reported chewing gutka and 8% reported use of pan with tobacco. Among appliers, the predominant form was tobacco toothpaste (31.7%) (Sinha et al., 2003c).

Sikkim

In Sikkim, gutka was the preferred (34.4%) form of smokeless tobacco, followed by khaini (18.9%), betel quid (15.7%), gul (15.0%) and tuibur (12.4%). Gutka, khaini and tuibur were used mainly by men while betel quid and gul were used primarily by women (Sinha et al., 2003b).

Current smokeless tobacco use was reported by 37.7% (42.5% of boys, 31.8% of girls), whereas smoking was reported by 23.6% (32.9% of boys, 12.1% of girls). Smokeless tobacco use exclusively in the form of chewing was reported by 48.3% and that exclusively in the form of application by 11.3%. Among chewers, tamol and tobacco mixture were reported to be the most popular (52.3%), followed by gutka (33.5%). Among appliers, 69.2% applied tobacco toothpaste, 21.4% applied red tooth powder and 9.4% applied tuibur (Sinha et al., 2003c).

Tripura

In Tripura, betel quid was the most popular (54.9%) form of smokeless tobacco, followed by gutka (21.0%) and khaini (10.7%). Betel quid was more popular among men while khaini was more popular among women (Sinha et al., 2003b).

Current smokeless tobacco use was reported by 35.1% (39.7% of boys, 29.4% of girls), whereas smoking was reported by 21.2% (28.6% of boys, 12.4% of girls). Smokeless tobacco use exclusively in the form of chewing was reported by 57.5% and that exclusively in the form of application by 28.8%. Among chewers, gutka was the most popular (21.3%), followed by tamol with tobacco (10.5%; 17.0% of boys, 2.6% of girls) and tamol without tobacco (23.0%; 23.7% of boys, 22.1% of girls). Thus, boys equally used tamol with tobacco or without tobacco, whereas girls preferred tamol without tobacco. Among appliers, the majority preferred tobacco toothpaste (25%) (Sinha et al., 2003c).

(vi) Type of tobacco used by sex and region

Bhonsle et al. (1992) reviewed available data from the 1970s on the prevalence of smokeless tobacco use by type of tobacco. Khaini use among men ranged from < 0.5% (Andhra Pradesh) to 44% (Bihar); that among women ranged from < 0.5 (Gujarat, Kerala) to 10% (Jharkhand). Chewing tobacco leaf varied among men from < 0.5 (Bihar, Goa, Gujarat, Jharkhand) to 9% (Andhra Pradesh) and among women from < 0.5 (Gujarat, Jharkhand) to 2% (Andhra Pradesh, Kerala). Applied tobacco (bajjar and gudhaku) was used by 1% of men and by 14–16% of women in Gujarat and Jharkhand (Tables 49 and 50; Bhonsle et al., 1992). Among 6271 school children in Goa (western India), 731 were tobacco users. Of these, 56% of boys and 66% of girls used mishri and almost half in both groups used creamy snuff (Table 51) (Vaidya et al., 1992). Among 9097 adults (≥ 15 years) in a rural site in Bihar (eastern India), one third (32.7%) used smokeless tobacco, of whom 11.4% used khaini and 18.9% used tooth powder that contained tobacco (Table 52) (Sinha et al., 2003a).

Table 49. Prevalence of use of smokeless tobacco and other chewing products among men in selected states in India.

Table 49

Prevalence of use of smokeless tobacco and other chewing products among men in selected states in India.

Table 50. Prevalence of use of smokeless tobacco and other chewing products among women in selected states in India.

Table 50

Prevalence of use of smokeless tobacco and other chewing products among women in selected states in India.

Table 51. Prevalence of use of different types of tobacco among schoolchildren in Goa, India.

Table 51

Prevalence of use of different types of tobacco among schoolchildren in Goa, India.

Table 52. Prevalence of use (%) of different types of tobacco among adults (≥ 15 years) in Bihar, India.

Table 52

Prevalence of use (%) of different types of tobacco among adults (≥ 15 years) in Bihar, India.

A population-based cross-sectional survey was conducted in the city of Mumbai among 99 598 individuals aged 35 years and older during 1992–94 (Gupta, 1996). A high percentage of women used tobacco (57.5%), almost solely in the smokeless form. About one fifth (20%) of the population (26.5% of women; 10.3% of men) used mishri alone and 3.7% (1.1% of women; 7.5% of men) used tobacco leaf and lime (Table 53).

Table 53. Prevalence of use of smokeless tobacco and other chewing products in Mumbai, India.

Table 53

Prevalence of use of smokeless tobacco and other chewing products in Mumbai, India.

Among 539 patients who entered hospital in Kerala and were recruited as controls for a case–control study, seven reported use of nasal snuff (Sankaranarayanan et al., 1989a).

(d) Indonesia

In Indonesia, smokeless tobacco is used mainly as part of a betel quid and mostly in rural areas. Betel quid with tobacco and chewing tobacco were identified as smokeless tobacco products used by a small number of respondents both in Jakarta and Sukabumi. Of 5899 tobacco users, less than 0.5% (22 persons) had used chewing tobacco (Sinha, 2004).

(e) Malaysia

A cross-sectional survey was conducted to document the use of smokeless tobacco among Kadazan women in a rural area in the state of Sabah, East Malaysia (Gan, 1995). Of the 472 women interviewed, 328 chewed; 60% of all women included tobacco as an ingredient in their chew, while 10% did not. Tobacco with lime was used by 2.3% of women and tobacco only by 1.1%. Women with a low education were more likely to be chewers. The chewing practice was usually acquired during the teenage years and was perceived mainly as a cultural norm. The majority of tobacco chewers (46.3%) used three or four fresh preparations per day. Tobacco use increased with increase in age (Table 54).

Table 54. Prevalence of tobacco chewing by Kadazan and Bajaus women in Sabah, Malaysia.

Table 54

Prevalence of tobacco chewing by Kadazan and Bajaus women in Sabah, Malaysia.

In a similar survey among the indigenous people of Sabah State, 845 Bajaus (414 men, 431 women) were interviewed (Gan, 1998). Of these, 74.4% of men smoked compared with 3.3% of women and 77% of women used smokeless tobacco compared with 4.3% of men. Tobacco was commonly used in the form of a betel quid. Among chewers, half (51.2%) used fewer than five quids per day. Only nine women used tobacco without areca nut (tobacco only, 1.4%; tobacco with lime, 0.7%). The prevalence of smokeless tobacco use was significantly lower among the better educated and increased with increasing age (Table 54).

(f) Myanmar

Zarda is manufactured in Myanmar and is also imported from India (Sinha, 2004).

The WHO Sentinel Prevalence Survey of Tobacco Use in Myanmar (WHO SEARO, 2001) covered a sample of 6600 individuals (2903 men, 3697 women) in the Hinthada district from the Delta region and the Pakkuku township from the Dry zone region. Among current tobacco users, two-thirds reported smoking and one-third reported chewing. Among chewers, most chewed tobacco with areca nut (31%) and only 2% chewed raw tobacco. Among the respondents, 21.2% (33.8% of men, 11.2% of women) reported ever use of smokeless tobacco and 14.9% reported current use. Current smokeless tobacco use was nearly three times more prevalent among men than among women both in rural and urban areas. Use of smokeless tobacco was not reported by any respondent aged 10–14 years.

In the GYTS conducted in 2004, smokeless tobacco use was reported by 10.8% of students aged 13–15 years. Boys reported significantly more smokeless tobacco use than girls (18.1% versus 3.6%) (Kyaing, 2005).

(g) Nepal

Several smokeless tobacco products — khaini, gutka and zarda — are consumed in Nepal. Although they are fairly new to the hill population, they are becoming increasingly popular in all parts of the country. Between 1996 and 1999, imports of khaini and zarda into Nepal, mostly from India, increased 72-fold (Karki et al., 2003).

Studies on the economics of tobacco use in Nepal revealed that there are no national or sub-national data from Nepal. A prevalence of 9.4% for khaini use and of 31.6% for smoking has been reported from a survey of 6000 people aged 10 years or over (Karki et al., 2003).

A cross-sectional survey was conducted in Dharan municipality (eastern Nepal) in 2001–2002 (Niraula, 2004). A representative sample of 2340 women aged 15 years and above was selected. Of these, 12.9% were cigarette smokers and 14.1% were smokeless tobacco users. The prevalence of tobacco chewing increased from 6.0% in the 15–24-year age group to peak at 25.3% in the 35–44-year age group, after which it decreased gradually. Tobacco chewing was more common among women who were involved in business (30.5%) than among others. Muslims were more likely and Christians were less likely to use tobacco than Buddhists (Table 55).

Table 55. Prevalence by sociodemographic characteristics of tobacco chewing among women (≥ 15 years) in Dharan, Nepal, 2002.

Table 55

Prevalence by sociodemographic characteristics of tobacco chewing among women (≥ 15 years) in Dharan, Nepal, 2002.

Nearly one student in 10 (9.3%) aged 13–15 years from the GYTS survey in Nepal reported current smokeless tobacco use. The prevalence of use among boys was significantly higher than that among girls (11.8% versus 5.6%) (Pandey & Pathak, 2001).

Among secondary school students of the sub-metropolitan city of Pokhara, ever use of gutka and khaini was reported by 41.2% and 3.0%, respectively. Smokeless tobacco use was more frequent among boys than girls (56.4% versus 31.2%). Non-governmental school students were more likely to use smokeless tobacco than governmental school students (Table 56; Paudel, 2003).

Table 56. Prevalence (%) of ever use of tobacco by type of product among 2032 secondary school students in Nepal.

Table 56

Prevalence (%) of ever use of tobacco by type of product among 2032 secondary school students in Nepal.

(h) Pakistan

Tobacco chewing alone, tobacco chewing with pan and tobacco chewing with smoking was reported by 2.2, 14.8 and 0.5% respondents, respectively, in a population sample of 10 749 people in Karachi (Mahmood et al., 1974) (Table 57). In a survey conducted in 1980 among 990 residents in Karachi, about 60% of men and 38% of women used tobacco; of these, about 11% of men and 31% of women chewed tobacco either on its own (1–2%), with pan (6.4–27%) or in association with smoking (2.2–2.5%) (Mahmood, 1982).

Table 57. Prevalence (%) of use of different tobacco products in Karachi, Pakistan, 1967–72.

Table 57

Prevalence (%) of use of different tobacco products in Karachi, Pakistan, 1967–72.

(i) People's Republic of China

China is the largest producer of tobacco in Asia (Shafey et al., 2003). No additional information on China was available to the Working Group.

(j) Sri Lanka

In the WHO Sentinel Prevalence Survey of Tobacco Use in Sri Lanka (cited in Sinha, 2004), a total sample population of 5886 people (49.3% men, 50.7% women) was inves-tigated. Current use of smokeless tobacco products was mainly a rural phenomenon (seven times more prevalent among men and six times more prevalent among women; Table 58), and prevalence among men was almost twice that among women. The trend indicated a decrease in the current use of smokeless tobacco with education and economic level and an increase with increasing age.

Table 58. Prevalence by selected sociodemographic characteristics of current use of smokeless tobacco in Sri Lanka, 2001.

Table 58

Prevalence by selected sociodemographic characteristics of current use of smokeless tobacco in Sri Lanka, 2001.

(k) Uzbekistan

In a survey conducted in Samarkand Oblast, all men aged 55–69 years who were residents in one local authority district were invited to participate. Of 1569 men, 636 (41%) reported naswar use and 259 (17%) were cigarette smokers (Zaridze et al., 1985).

1.4.4. Africa

The two major tobacco producing countries in Africa are Malawi and Zimbabwe (Shafey et al., 2003). The most widely grown type of tobacco in Zimbabwe is flue-cured Virginia, while Malawi is the largest producer of Burley tobacco in Africa. (Burley tobacco accounts for just under 15% of global tobacco production.) Worldwide, approximately 11–12 million farmers cultivate tobacco, about 18 000 of whom are in Zimbabwe and 375 000 in Malawi (Jaffee, 2003). Exports of tobacco leaf increased continuously in both countries between 1980–82 and 1997–99, by 130% in Malawi and by 69% in Zimbabwe (Jaffee, 2003). According to different sources, Zimbabwe produced 210 000–230 000 tonnes of tobacco leaves and Malawi 99 000–125 000 tonnes in 2000 (Jaffee, 2003; Shafey et al., 2003). Tobacco accounts for over 30% of the export revenue of Zimbabwe and 75% of that of Malawi (Shafey et al., 2003).

Data from countries that participated in the GYTS are presented in Table 59 (Global Youth Tobacco Survey Collaborating Group, 2003).

Table 59. Prevalence (%) of use of non-cigarette tobacco among students aged 13–15 years in African countries, 1999–2002 (Global Youth Tobacco Survey).

Table 59

Prevalence (%) of use of non-cigarette tobacco among students aged 13–15 years in African countries, 1999–2002 (Global Youth Tobacco Survey).

(a) Kenya

A small study conducted among five ethnic groups in Kenya assessed differences in smokeless tobacco use related to gender and generation. In four of the five groups, little or no difference was observed in the prevalence of smokeless tobacco use between generations (except for the Gikuyu) or between sexes (except for the Luo) (Kaplan et al., 1990).

(b) South Africa

In South Africa, smokeless tobacco is more commonly used through the nose and less commonly orally (Ayo-Yusuf et al., 2004). Between 1992 and 1995, the consumption of snuff in South Africa increased by about 30% from 1.1 million kg to 1.5 million kg (Tobacco Board, 1992, 1994/95, cited in Ayo-Yusuf et al., 2004).

A national household survey provided cross-sectional data on a representative sample of the population of South Africa (Table 60). Of 13 826 participants (5753 men, 8073 women), 6.7% (0.9% of men, 10.2% of women) used snuff or chewed tobacco. Smokeless tobacco was used predominantly by African and coloured women and the prevalence increased with age to peak at 28.9% and 9.7%, respectively, in women older than 64 years. The age-standardized prevalence of use of smokeless tobacco in South Africa was higher for Africans (8.4%) than for other ethnic groups (coloureds, 1.9%; whites, 0.8%; Asians, 0.2%) (Steyn et al., 2002).

Table 60. Prevalence of use of smokeless tobacco in South Africa, 1998.

Table 60

Prevalence of use of smokeless tobacco in South Africa, 1998.

A telephone survey of 300 tobacco users each in Seshego (black area) and Pietersburg (white area) revealed that 3% of the white tobacco users used snuff while almost half (46.7%) of the blacks used snuff (Peltzer, 1999). The typical form of using snuff in this survey was sniffing (86.7%); placing snuff in the mouth was practiced by men only (13.3%).

In a sample of 330 grade-10 and 382 grade-12 students from a rural population (age range, 13–23 years), 4.0% of boys and 8.4% of girls were current snuff users (Peltzer, 2003). Twenty-four (3.4%) of the participants were current snuff users only, 31 (4.3%) were current smokers only and 17 (2.4%) currently used both snuff and cigarettes (Table 61). The preferable mode of taking snuff was by sniffing, followed by mouth only and both sniffing and by mouth. The prevalence of smokeless tobacco use was not significantly different between grade-10 and grade-12 students (Peltzer, 2003).

Table 61. Status of snuff use and age at start by sex among students in South Africa.

Table 61

Status of snuff use and age at start by sex among students in South Africa.

A structured questionnaire was administered by means of face-to-face interviews to 30 randomly selected households in a rural population (125 adults over the age of 30 years). Of the respondents, 20.8% were active oral snuff dippers (Table 62). No significant difference was observed in the prevalence between sexes (p > 0.05). None of the snuff dippers chewed or smoked tobacco. Among the snuff dippers, the vast majority (85%) placed the snuff in the lower labial sulcus; the remainder placed it in the lower buccal sulcus. About half of them prepared the snuff themselves and the other half acquired it commercially. The mean age of snuff dippers was 62.7 years (range, 36–95 years) and mean duration of use was 21.5 years; dipping lasted for about 2 h per day with an average of 35 min per dip (Ayo-Yusuf et al., 2000).

Table 62. Prevalence and pattern of snuff dipping in a rural population in South Africa.

Table 62

Prevalence and pattern of snuff dipping in a rural population in South Africa.

(c) Sudan

Idris et al. (1994) conducted a cross-sectional survey on the use of toombak in a random population sample in the Nile State in northern Sudan. In a preliminary report on 2000 households with 5500 adults, about 40% of the men dipped toombak, and 9% were both cigarette smokers and toombak dippers. Toombak was particularly prevalent (> 45%) among men aged 40 years or older. Among women, toombak use was popular only in the older age groups; up to 10% of women aged 60 years and over used toombak (Idris et al., 1994).

A later report included results from 4535 households with 21 594 individuals aged 4 years and over (Table 63). In 60% of all households at least one member used toombak. The prevalence of toombak use in the entire population aged 4 years or older was 12.6%. The prevalence of toombak use was low (1.7%) among children and adolescents (4–17 years) and was highest in the oldest age group (70 years and older). Among the adult population aged 18 years and older, the prevalence of toombak use was significantly higher among men (34.1%) than among women (2.5 %), and was significantly higher in rural than in urban areas (35.4% versus 23.5% in men). The highest rates of toombak use were found in the rural areas among men aged 30 years and older (mean, 46.6%; range, 45.3–47.1%) (Idris et al., 1998b).

Table 63. Prevalence (%) by age and area of residence of current use of toombak among men and women in Sudan, 1992.

Table 63

Prevalence (%) by age and area of residence of current use of toombak among men and women in Sudan, 1992.

(d) Tunisia

A cross-sectional study of a representative national sample of 5696 subjects aged 25 years and over was conducted in 1996, in which data were collected by means of a questionnaire. Tobacco use was reported by 30.4% of the respondents; 5.8% consumed ‘traditional’ tobacco, which was defined as tobacco in the form of snuff (neffa), chewing tobacco and/or a waterpipe. In this geographical area, neffa is the predominant form of snuff used. Use of ‘traditional’ tobacco was influenced by age, sex, level of education and rural or urban environment (Table 64). The proportion of men who only consumed ‘tradi-tional’ tobacco increased from 2.4% in the 25–34-year age group to 20.4% in the ≥ 55year age group; the corresponding values for women were 0.1% and 14.3%, respectively. The consumption of ‘traditional’ tobacco was more widespread in rural than in urban areas and was relatively high among poorly educated men from economically deprived backgrounds (Fakhfakh et al., 2002).

Table 64. Prevalence (%) by socioeconomic characteristics and age of use of ‘traditional’ tobacco in 5696 subjects in Tunisia, 1996.

Table 64

Prevalence (%) by socioeconomic characteristics and age of use of ‘traditional’ tobacco in 5696 subjects in Tunisia, 1996.

(e) Other countries

In Algeria, an estimated 90% of tobacco production is used for the manufacture of snuff, and 24% of all tobacco consumed is in the form of snuff (WHO, 1997).

In Libya, apporoximately 140 tonnes of chewing tobacco are consumed every year (WHO, 1997).

Shammah is a traditional form of chewing tobacco that is used very commonly in southern Saudi Arabia and in Yemen (Hannan et al., 1986; Ibrahim et al., 1986).

In Lesotho, according to a 1992 survey in rural areas, prevalence of smokeless tobacco use was 2.7% for nasal snuff and 0.3% for oral snuff in those aged 15–29 years, 19.6% for nasal snuff and 2.1% for oral snuff in those aged 30–44 years and 28.5% for nasal snuff and 8.7% for oral snuff in those aged 45 years and over (WHO, 1997).

In Swaziland, Zambia and Zimbabwe, snuff taking is common in rural areas, particularly among older persons (WHO, 1997).

1.4.5. Association between smokeless tobacco use and cigarette smoking

Because the use of smokeless tobacco or cigarettes are both associated with nicotine delivery and addiction, interrelationships between smokeless tobacco use and smoking may help to explain long-term historical patterns and trends in the use of these products by populations of various cultures. Some observations suggest that certain smokeless tobacco products may serve as an effective method to quit smoking (Kozlowski et al., 2003). Others have attributed the decline in smoking that has occurred in Sweden since the early 1980s to the expansion of moist snuff use in that country (Bates et al., 2003). However, some researchers have suggested that smokeless tobacco may actually serve as starter product for nicotine addiction among young people in the USA, which could lead to subsequent smoking (Tomar, 2003b), and is rarely used as a smoking cessation strategy (Tomar & Loree, 2004); others have questioned whether snuff played any significant role in reducing smoking in Sweden (Tomar et al., 2003; Lambe, 2004). The interrelationship between smokeless tobacco use and smoking, together with recommendations by tobacco manufacturers or those who advocate that tobacco users switch from one product type to another, may have significant implications for exposure to carcinogens among individuals and populations. Sweden and the USA provide the only two examples of nations in which commercial moist snuff products are widely promoted, available and used, and from which there are available epidemiological data to examine the interrelationship between the use of moist snuff and cigarette smoking.

(a) Data from Sweden

A number of reports indicate that dual use of moist snuff and cigarettes is fairly prevalent in Sweden. In 1985–87, 47% of all male snuff dippers were also smokers compared with 36% of non-snuff users who were smokers (Nordgren & Ramström, 1990). More recent, official national data on the prevalence of dual use could not be located, although a Swedish survey of current and former smokers commissioned by the Swedish Cancer Society and Pharmacia AB in 2000 found that 19.8% of male current smokers also used moist snuff (Gilljam & Galanti, 2003). A census of ninth grade students (aged 15–16 years) in the County of Stockholm found that 14.3% of boys were exclusively smokers, 5.7% were exclusively snuff dippers and 13.8% used both cigarettes and snuff (Galanti et al., 2001a), that is, 71% of boys who used snuff also smoked and 49% of boys who smoked also used snuff.

Some data indicate that snuff use may be a precursor to smoking among young men in Sweden. In a cohort study conducted in the County of Stockholm that began in 1997, 2883 students in the fifth grade were recruited and followed-up 1 year later (Galanti et al., 2001b). At baseline, 22% of boys and 15% of girls had ever smoked and 8 and 3%, respectively, had ever used oral moist snuff. One year later, the overall prevalence of smoking had increased markedly, as had the transition to more advanced stages of smoking, especially among girls. The authors concluded that, in most cases, experimentation with oral snuff among boys marked the transition to cigarette smoking.

The extent to which snuff use may account for the decline in smoking in Sweden during the past few decades is unclear. Ramström (2000) reported that, in national surveys of the Swedish population in 1987 and 1988, respondents who had ever used tobacco were asked whether their primary tobacco use was smoking or snuff dipping. Among men aged 18–34 years, 43% were ever daily smokers; of these, 21.5% were former smokers and 21.5% were current daily smokers. Fifty-one per cent of women of the same age were ever daily smokers: 18.5% were former smokers and 32.5% were current daily smokers. From this observation, the author concluded that “Since the one major difference between men and women in Sweden is the widespread use of snuff among men and virtually no snuff use among women, it seems probable that male snuff use has kept down onset of smoking and increased smoking cessation” (Ramström, 2000). Similarly, the review by Foulds et al. (2003) cited ecological data on trends of sales of snuff and cigarettes, unadjusted data on prevalence of smoking and male snuff use and sequential cross-sectional surveys from a study in northern Sweden [the Working Group noted that this study was funded by the smokeless tobacco industry] (Rodu et al., 2002) as being “strongly sugges-tive of snus having a direct effect on the changes in male smoking and health”. [Most conclusions that suggest that snuff played a significant role in reducing cigarette smoking are based largely on ecological or cross-sectional studies.]

Several studies in Sweden examined the possible contribution of snuff to quitting smoking. In a 1-year cohort study of 12 507 persons aged 47–68 years at baseline in 1992–94, Lindström et al. (2002a) examined predictors of smoking cessation or change to intermittent (non-daily) smoking among 3550 daily smokers. At baseline, 7.0% of all men and 0.4% of all women were snuff users. At the 1-year follow-up, 7.2% of daily smokers had quit and 6.5% had become intermittent smokers. In a multiple logistic regression analysis that controlled for sex and other demographic characteristics, daily smokers who remained so were less likely than the total population to be snuff users at baseline (odds ratio, 0.67; 95% confidence interval [CI], 0.51–0.87); daily smokers who became intermittent smokers were more likely than the general population to be snuff dippers at baseline (odds ratio, 1.94; 95% CI, 1.07–3.51); and daily smokers who quit smoking did not differ from the total population in their use of snuff at baseline (odds ratio, 1.1; 95% CI, 0.54–2.26). The study did not report changes in snuff use during the period of follow-up. The authors concluded that sex differences in snuff consumption could provide “… a substantial, although not major, fraction of the explanation for why there has been an increase in smoking cessation in recent years among men but not among women, although we believe that other social and work-related factors may be even more important”. Another analysis of the same cohort focused on intermittent smokers at baseline (Lindström et al., 2002b), who accounted for 4.8% of the cohort of 699 people. At the 1-year follow-up, 59.9% of inter-mittent smokers were still intermittent smokers (intermittent/intermittent), 15.9% had become daily smokers (intermittent/daily) and 19.2% had stopped smoking (intermittent/topped). Among intermittent/intermittent, 11.5% were snuff users at baseline, as were 9.5% of intermittent/daily, 9.0% of intermittent/stopped and 3.0% of the total cohort which included daily smokers, former smokers and never smokers. In multivariate logistic regression modelling, snuff use was a moderately strong correlate of intermittent smoking compared with the reference group regardless of smoking status at follow-up: odds ratios were 3.40 (95% CI, 1.70–6.81) for intermittent/daily, 4.22 (95% CI, 3.00–5.94) for intermittent/intermittent and 3.20 (95% CI, 1.79–5.71) for intermittent/stopped. The investigators did not report changes in snuff use during the follow-up and did not explicitly compare changes in smoking status as a function of snuff use at baseline. From these two studies, it may be concluded that: (a) snuff use may have been more common among intermittent smokers aged 45–69 years than among the rest of the adult population of that age group, but did not seem to be associated with subsequent cessation or prevent transition to daily smoking; and (b) snuff use was less common among daily smokers who remained daily smokers than among the general population, but was associated only with their transition to intermittent smoking and not with smoking cessation at the 1-year follow-up.

In a similar study, 5104 persons aged 16–84 years were interviewed in 1980–81 and then followed up in 1988–89 (Tillgren et al., 1996). The cohort included 1546 daily smokers, 418 men who were daily snuff users and 129 men who used both snuff and cigarettes. At follow-up, 5% of male smokers had switched to snuff and 2% had started using snuff in addition to cigarettes, and 5% of non-tobacco users had started using snuff. Among male exclusive snuff users, 26% had quit all tobacco use and 10% had taken up cigarettes in addition to (5%) or instead of (5%) snuff. Among male dual product users, 56% either had continued dual product use or exclusively smoked, 31% exclusively used snuff and 13% had quit all tobacco use.

Rodu et al. (2003) reported findings from the MONICA cohort study: persons aged 25–64 years at study entry and who joined the cohort in 1986, 1990 and 1994 were followed up until 1999. Among all 308 men who smoked at entry to the study, 19% exclu-sively used snuff and 24% used no tobacco product at follow-up. Among 195 male smokers who had never used snuff at entry [63% of all male smokers], 57% were still exclusively smokers at follow-up, 8% had switched to snuff, 6% used both cigarettes and snuff and 29% used no tobacco product. Among 423 women who smoked at entry to the study, 3% exclusively used snuff and 27% used no tobacco products at follow-up.

The most recent evidence that snuff may be a factor in the decline in smoking in Sweden over the past 20 years derives from cross-sectional studies. Gilljam and Galanti (2003) reported results from a survey in 2000 of 1000 former and 985 current daily smokers aged 25–55 years. Among men, more former smokers than current smokers had ever used snuff (54.7% versus 44.8%; p = 0.003) or currently used snuff (28.9% versus 19.8%; p = 0.002). Among men, snuff had been used at the most recent attempt to quit smoking by 28.7% of former smokers and 23.0% of current smokers (p = 0.072). The study found that having used snuff at the most recent attempt to quit was associated with an increased likelihood of abstinence among men (odds ratio, 1.54; 95% CI, 1.09–2.20). [The authors did not report an association between snuff use and cigarette smoking separately for women, but it could be calculated from the data reported in the tables. Snuff use was much less common among women than among men and did not differ between current and former smokers: 13.1% of women reported ever using snuff, including 14.1% of current smokers and 12.1% of former smokers, and 2.9% of women were current snuff users, including 2.5% of current smokers and 3.3% of former smokers. Use of snuff at the most recent attempt to quit smoking was reported by 4.8% of female current smokers who had attempted to quit and 4.5% of female former smokers.] These findings suggest that snuff use may be associated with smoking cessation among Swedish men but not women.

(b) Data from the USA

Many cross-sectional studies in the USA have reported moderate-to-strong degrees of association between concurrent smoking and use of smokeless tobacco in the adolescent population (Lichtenstein et al., 1984; Ary et al., 1987; Jones & Moberg, 1988; Murray et al., 1988; Olds, 1988; Ary et al., 1989; Colborn et al., 1989; Glover et al., 1989; Peterson et al., 1989; Riley et al., 1989; Sussman et al., 1989; Severson, 1990; Lee et al., 1994; Hatsukami et al., 1999; Coogan et al., 2000; Ringel et al., 2000). These studies, however, used a wide range of definitions of tobacco use and were often unable to establish a temporal relationship with the initiation of use of each tobacco product. Relatively few reports of longitudinal investigations into the relationship between smoking and smokeless tobacco have been published.

Some longitudinal studies found that the use of smokeless tobacco was predictive of the onset of or increase in cigarette smoking (Ary et al., 1987; Dent et al., 1987; Ary et al., 1989; Haddock et al., 2001), while others reported that smoking was predictive of initiation of experimentation with or regular use of smokeless tobacco (Ary et al., 1987; Dent et al., 1987; Ary, 1989; Sussman et al., 1989; Tomar & Giovino, 1998).

Two recent cohort studies suggest that use of smokeless tobacco may be a predictor of subsequent smoking among young men in the USA. In a cohort study of 7865 Air Force recruits with a mean age of 19 years at baseline, Haddock et al. (2001) considered regular smokeless tobacco use to be use of these products at least once per day; the 1-year measure of smoking outcome was defined as any smoking within the preceding 7 days. Among current smokeless tobacco users, 27% initiated smoking, compared with 26.3% of former smokeless tobacco users and 12.9% of never users. After adjustment for demographic characteristics among recruits who had never been daily smokers, current users (odds ratio, 2.33; 95% CI, 1.84–2.94) and former users (odds ratio, 2.27; 95% CI, 1.64–3.15) of smokeless tobacco products were significantly more likely than never users to initiate smoking. Current or former smokeless tobacco use was a much stronger predictor of initiation of smoking than a range of other behaviours, including rebelliousness, use of a seat belt, alcoholic beverage consumption, binge drinking, level of physical activity and fruit and vegetable intake.

A recent nationally representative cohort study of adolescent boys and young adult men in the USA examined the longitudinal relationship between use of smokeless tobacco and initiation of smoking (Tomar, 2003b). Data were from the 1989 Teenage Attitudes and Practices Survey and its 1993 follow-up that comprised 7960 people aged 11–19 years at baseline. Analyses were limited to 3996 boys and men with complete data on smoking and smokeless tobacco use at both interviews. Young men who were not smokers in 1989 but regularly used smokeless tobacco were more than three times more likely than never users to be current smokers 4 years later (23.9% versus 7.6%; adjusted odds ratio, 3.45; 95% CI, 1.84–6.47). In contrast, 2.4% of current smokers and 1.5% of never smokers at baseline had become current regular smokeless tobacco users by follow-up. More than 80% of baseline current smokers were still smokers 4 years later and less than 1% had switched to smokeless tobacco; in contrast, 40% of baseline current regular smokeless tobacco users became smokers either in addition to or in place of smokeless tobacco use. The results suggest that smokeless tobacco may be a starter product for subsequent smoking among young men and boys in the USA, but may have little effect on quitting smoking in that age group.

Another analysis (O'Connor et al., 2003) of the same cohort as that analysed by Tomar (2003b) suggested that smokeless tobacco was no longer a statistically significant predictor of initiation of smoking when psychosocial risk factors were included in multiple logistic regression modelling. In modelling of predictors of current smoking among men and boys who had never experimented with cigarettes at baseline, an adjusted odds ratio of 1.97 (95% CI, 0.69–5.65) was found for those who reported regular use of smokeless tobacco. The study also suggested an association between smokeless tobacco use and established risk factors for initiation of smoking, such as having a smoker in the household (odds ratio, 1.52; 95% CI, 1.10–2.11). Another analysis also suggested a positive association between regular smokeless tobacco use and initiation of smoking in a model that included experimentation with cigarettes, school performance, depressive symptoms, having a smoker in the household and several markers of risk-taking behaviour (odds ratio, 1.68; 95% CI, 0.83–3.41).

A repeat of the analytic approach of O'Connor et al. (2003) that limited the analysis to boys under 16 years of age at baseline found that boys who had used smokeless tobacco were significantly more likely than non-users to be current smokers at follow-up (odds ratio, 1.67; 95% CI, 1.03–2.70) in multivariable modelling that included race or ethnicity, geographical region of residence, experimentation with cigarettes, school performance, having a smoker in the house, depressive symptoms and two markers for risk-taking behaviour (Tomar, 2003c).

This series of analyses showed that smokeless tobacco use was an independent predictor of cigarette smoking among adolescent boys with a strength of association that was comparable with that of other established risk factors. However, regular use of smokeless tobacco was relatively uncommon at baseline in this cohort study and therefore the parameter estimates were fairly imprecise.

In an analysis of cross-sectional data from the 1987 National Health Interview Survey (NHIS), Kozlowski et al. (2003) found a significant association between ever use of smokeless tobacco and current smoking (odds ratio, 1.35; 95% CI, 1.05–1.74), but no association when men who had used cigarettes before smokeless tobacco were excluded from the analysis (odds ratio, 0.79; 95% CI, 0.56–1.11). On this basis, the authors con-cluded that the order of product use must be considered and that use of smokeless tobacco was unlikely to predict smoking. [The Working Group noted that the analysis did not exclude the many persons at any given age who had already become smokers; the large majority of men in the USA who initiate smoking do so without ever using smokeless tobacco, but that does not rule out the use of smokeless tobacco as a risk factor for nicotine addiction and initiation of smoking. The analytic approach of Kozlowski et al. (2003) was analogous to conducting a case–control study in which a very large proportion of the control group actually had the disease; such misclassification generally biases results toward the null (Rothman & Greenland, 1998).]

Tomar and Loree (2004) subsequently modelled smokeless tobacco use as a possible predictor of smoking by excluding from the analysis those who were already smokers at a particular age, and then examined whether smokeless tobacco use predicted subsequent smoking. In contrast to the conclusion of Kozlowski et al. (2003), Tomar and Loree (2004) found that white boys who used smokeless tobacco at age 15 years but had never smoked were significantly more likely than non-users of smokeless tobacco to become smokers subsequently, after controlling for age, geographical region and educational attainment (odds ratio, 1.80; 95% CI, 1.15–2.82). Similar results were found when the analysis was repeated for age 16 years (odds ratio, 1.53; 95% CI, 1.03–2.30) or age 17 years (odds ratio, 1.87; 95% CI, 1.17–2.98).

Only one study in the USA has explicitly examined the effectiveness of snuff use as a method for smoking cessation (Tilashalski et al., 1998). This pilot study found that 16 of the 63 subjects (25%) in the study had quit smoking at the 1-year follow-up by using snuff and six subjects (10%) had quit smoking by using some other method. [The study did not include a control group.] In a 7-year follow-up of 62 of the original 63 subjects, 28 (45%) had quit smoking, although fewer than half of subjects (n = 12) had reportedly done so by using snuff (Tilashalski et al., 2003).

Fiore et al. (1990) reported findings from the 1986 Adult Use of Tobacco Survey on the methods that smokers used to quit. In the mid-1980s, 6.8% of former smokers who had successfully quit smoking for at least 1 year had substituted cigarettes with other tobacco products (including snuff, chewing tobacco, pipes or cigars) during any attempt to quit and 4.0% during their last attempt to quit. However, the proportions were very similar among those who relapsed: 6.8% of smokers who had made a serious attempt to quit in the past year but were not successful had tried substituting other tobacco products at any attempt and 2.1% had tried that strategy at their last attempt to quit.

The most recent direct measurement of the extent of smokeless tobacco use in the USA as a method for quitting smoking derives from the 2000 NHIS. Tomar and Loree (2004) examined changes in tobacco use within the male birth cohorts that were included in the cross-sectional analysis of Kozlowski et al. (2003) of data from the 1987 NHIS. A compa-rison of the prevalence of tobacco use among men aged 23–34 years in 1987 with that of 36–47-year-olds in 2000 revealed a very small decline in the prevalence of current smoking among this birth cohort, from 34.1% (95% CI, 31.9–36.3%) in 1987 to 31.0% (95% CI, 29.1–32.9%) in 2000; the prevalence of current snuff use declined during the same period from 5.8% (95% CI, 4.6–7.0%) to 2.5% (95% CI, 1.9–3.1%). Former smokers in the 2000 NHIS were asked what method they had used to quit smoking completely. Only 1.2% (95% CI, 0.1–2.3%) of male former smokers aged 36–47 years in 2000 reported switching to snuff or chewing tobacco to quit smoking. Of male current smokers in that age group who had unsuccessfully tried to quit, 0.3% (95% CI, 0.0–0.7%) reported switching to smokeless tobacco on their last attempt to quit. In a birth cohort in which 15.5% of men, who included 19.0% of ever smokers, had used smokeless tobacco by the age of 34 years, this practice accounted for a very small proportion of smoking cessation. The authors calculated that the number of men in this birth cohort who used smokeless tobacco, apparently for reasons other than smoking cessation, was up to 68 times greater than the number who used it to quit smoking. The number of men in this birth cohort for whom smokeless tobacco was a probable starter product for smoking was estimated to be about 17 times that of men who reported quitting smoking by using smokeless tobacco.

A recent cross-sectional study examined the associations between snuff use and smoking in a representative sample of men in the USA (Tomar, 2002). The 13 865 subjects were men aged 18 years and older in the 1998 NHIS. Multiple logistic regression modelling was used to examine the association between the use of snuff and cessation of smoking. The study reported that, in 1998, 26.4% of men in the USA smoked, 3.6% used snuff and 1.1% used both products. After adjusting for age and race or ethnicity, the prevalence of current smoking was higher among former snuff users (39.4%) and occasional users (38.9%) than among daily users (19.2%) or never users (25.4%). Daily snuff users were significantly more likely than never users to have quit smoking in the preceding 12 months (odds ratio, 4.2; 95% CI, 2.2–8.3). Occasional snuff users were more likely than never users to have tried to quit smoking in the preceding year (odds ratio, 1.7; 95% CI, 1.0–2.8) but tended to be less likely to succeed (odds ratio, 0.5; 95% CI, 0.2–1.3). After adjustment for age and race or ethnicity, smokers who used snuff daily smoked significantly fewer cigarettes per day on average than those who never used snuff (11.4 versus 18.4 cigarettes; p = 0.0001). Men were nearly three times more likely to be former snuff users who currently smoked (2.5%) than to be former smokers who currently used snuff (0.9%). The author concluded that although some men may use snuff to quit smoking, men in the USA more commonly switched from snuff use to smoking.

Wetter et al. (2002) examined the characteristics, tobacco use patterns over time and predictors of tobacco cessation among 220 male concomitant users of cigarettes and smokeless tobacco in a large, randomized, worksite-based, matched-pair cancer prevention trial (n = 4886). High levels of dual use were found: 20% of smokeless tobacco users were also smokers (4% of the total study population). Compared with exclusively smokeless tobacco users, dual users were significantly more likely to be unmarried, to drink more alcoholic beverages, to live with a smoker and to use less smokeless tobacco per day, but had higher estimated exposure to nicotine. Dual users appeared to be less ready to change their use of smokeless tobacco than exclusively smokeless tobacco users. At the 4-year follow-up, exclusively smokeless tobacco users were the most likely (20.1%) and dual users were the least likely (11.3%) to have quit all tobacco use; 15.7% of exclusively smokers had quit. Among men who were dual users at baseline, 44.3% were still dual users at the 4-year follow-up, 27.0% were exclusively smokers and 17.4% were exclusively smokeless tobacco users. Men who were exclusively smokers or smokeless tobacco users at baseline showed little inclination to switch products completely, and comparable proportions added use of the other product: 4.6% of baseline smokers began using smokeless tobacco exclusively or in combination with cigarettes and 3.4% of baseline smokeless tobacco users began to smoke either exclusively or in combination with smokeless tobacco. Traditional measures of nicotine dependence (e.g. number of cigarettes or smokeless tobacco uses per day) that predicted cessation among exclusive smokers or smokeless tobacco users were not related to smoking cessation among dual users. Whether due to subject characteristics or the nature of dual product use, dual users in this study had the lowest tobacco cessation rates and tended to shift product use in both directions.

Dual tobacco use has been found to be fairly prevalent in specific subpopulations in the USA, such as in certain Native American populations, among whom 18% of current smokers also used smokeless tobacco and 26% of smokeless tobacco users also smoked (Spangler et al., 2001a).

At the population level, a possible effect of the substitution of smokeless tobacco for cigarettes could be manifested by a trend of increasing prevalence of smokeless tobacco use and declining prevalence of smoking. The possibility of such a pattern was explored by examining survey data collected among senior high school pupils as part of the Moni-toring the Future Project, which has been conducted since 1975 by the University of Michigan under contract with the National Institute on Drug Abuse (Johnston et al., 2003). Data on cigarette smoking have been collected since the inception of the study and those on smokeless tobacco use since 1986. Trends in daily tobacco use among male senior high school pupils in the USA do not support a substitution effect of one product for another (Figure 4). The prevalence of daily smokeless tobacco use remained relatively constant from 1992 to 1996, and was 6–7% for young men. Following a slight increase in 1997 to 8.6%, the prevalence has declined gradually and was 4.3% in 2002. The prevalence of daily cigarette smoking increased from 17.2% in 1992 to 24.8% in 1997, after which it began to decline and returned to 17.2% in 2002. At the population level, therefore, it appears that daily use of either cigarettes or snuff has been declining since 1997.

Figure 4. Trends in prevalence of daily use of cigarettes or smokeless tobacco among male high school seniors.

Figure 4

Trends in prevalence of daily use of cigarettes or smokeless tobacco among male high school seniors. Monitoring the Future Project, 1992–2003

Another approach to the association between smokeless tobacco use and cigarette smoking in populations is to examine their prevalence by state. This was investigated by using data from the September 1998 and January and May 2000 Tobacco Use Supplements to the Current Population Survey. The Current Population Survey was conducted for the Bureau of Labor Statistics by the US Census Bureau and the Tobacco Use Supplements were deve-loped and sponsored by the National Cancer Institute. Linear regression analysis revealed a statistically significantly positive association (β = 0.456; p < 0.0001; R2 = 0.2984) between state-level prevalence of smokeless tobacco use and cigarette smoking among men aged 18 years and older (Figure 5). Similarly, there was a significantly positive asso-ciation (β = 1.291; p < 0.0001; R2 = 0.2841) between the prevalence of daily use of snuff and the prevalence of daily cigarette smoking among men in the states. The association between state prevalence of smokeless tobacco use and cigarette smoking was nearly iden-tical when analyses were limited to white men. Although cultural and economic factors may affect the use of either tobacco product within states, the ecological patterns of use do not support the existence of widespread product substitution or a ‘preventive’ effect, in which higher prevalence of smokeless tobacco use is associated with lower prevalence of cigarette smoking (Tomar & Loree, 2004; Tomar, 2007).

Figure 5. Linear regression model of prevalencea of currentb smokeless tobacco use on prevalence of current cigarette smoking among men aged 18 years and older in 50 states, USA, 1998 and 2000.

Figure 5

Linear regression model of prevalencea of currentb smokeless tobacco use on prevalence of current cigarette smoking among men aged 18 years and older in 50 states, USA, 1998 and 2000. From unpublished data from the September 1998 and January and May 2000 (more...)

(c) Summary

In many ways, the recent histories of snuff use in Sweden and the USA are very similar. In both countries, the products were heading towards extinction in the late 1960s, when the development of new products, new images and aggressive marketing led to a new surge in sales. In both countries, these products were adopted largely by young men.

The primary difference between the countries is that the prevalence of daily use of snuff grew to a much larger extent in Sweden, perhaps due to a long history of snuff use and greater cultural acceptance of snuff dipping. The difference may also be attributable to the dominance of a single domestic tobacco company, Swedish Match, in both the cigarette and snuff markets. Swedish Match may have been willing to expand one market (moist snuff) by fostering a transfer of customers from the cigarette market; the company even sold its cigarette business to an Austrian tobacco company in 2000 (Henningfield & Fagerström, 2001). The exact role that snuff has played in reducing the prevalence of smoking in Sweden is unclear, but it has probably been overstated (Tomar et al., 2003). The decline in smoking in Sweden during the past two decades occurred in an environment of increased taxation on cigarettes, increased availability of treatment, expansion of clean indoor air policies and increased communication about the dangers of smoking in Sweden (Henningfield & Fagerström, 2001). Evidence from ecological studies that the increasing prevalence of moist snuff use in Sweden has led to a decline in smoking is inconclusive because of the methodological limitations of ecological studies, which do not directly measure changes in behaviours by individuals. Data from the few available Swedish cohort studies do not support a conclusion that moist snuff was a major factor in the decline in smoking, and in even in areas of Sweden that have a relatively high use of moist snuff, adult smokers who have no previous history of snuff use rarely adopted these products. In the USA, cohort studies of young men suggest that a high proportion of young smokeless tobacco users subsequently initiate smoking, but very few smokers switch to using smokeless tobacco. Consistent with cohort studies, cross-sectional studies in the USA suggest that smokeless tobacco use is rarely used to quit smoking, even among birth cohorts with a substantial history of using those products. It is less clear what the effects might be if moist snuff is aggressively marketed in societies that have little previous experience with these products. Recent history suggests that snuff use will probably gain much more popularity among young men who have never used tobacco or are in the early stages of initiation of tobacco use than among middle-aged smokers who are looking for a cessation strategy.

1.4.6. Occupational exposure to unburnt tobacco

The manufacture of bidis is one of the largest cottage industries in India and provides employment to more than 3 million people (Govekar & Bhisey, 1992). On average, a bidi roller makes 500–1000 bidis per day and handles 225–450 g of tobacco flakes, and is thus exposed by dermal contact. In addition, the workers also receive airborne exposure to tobacco dust and volatile components.

(a) Exposure to tobacco dust

Several studies conducted in various countries suggest that tobacco workers are exposed to tobacco dust and particulate matter (Mengesha & Bekele, 1998; Uitti et al., 1998; Mustajbegovic et al., 2003; Yanev & Kostianev, 2004; Zuskin et al., 2004) (Table 65). In a study that assessed the extent of exposure to tobacco dust among workers in bidi tobacco processing plants (Bhisey et al., 1999a), the mean concentration of inspirable dust particles was 150 times higher than that in the control environment.

Table 65. Dust levels in tobacco factories.

Table 65

Dust levels in tobacco factories.

Yanev and Kostianev (2004) determined that the majority of tobacco dust particles had a size of 0.3 µm (range, 0.05–16 µm), and some anisometric forms ranged in size from 0.1 to 2.0 µm.

(b) Biomonitoring of bidi industry workers

Exposure to tobacco-specific compounds and to electrophilic moities through the occupational use of tobacco can be determined among bidi rollers by measuring urinary cotinine and thioethers, respectively. A series of studies have measured occupational exposure of bidi workers to nicotine and carcinogens through biomonitoring.

Ghosh et al. (1985) conducted a study of tobacco processing workers in India. Among non-tobacco users, none of the control subjects had detectable levels of urinary nicotine or cotinine; levels in exposed workers were 3.13 µg/mL and 3.4 µg/mL, respectively. The mean urinary nicotine and cotinine levels were higher among workers than among controls.

Urinary cotinine and thioethers were determined in samples from two groups of bidi rollers and controls from the same community (Bhisey & Govekar, 1991). None of the subjects used tobacco in any form. One group of bidi rollers lived in the most densely populated part of Mumbai and worked in a poorly ventilated room, while the other lived in an area with open spaces and worked singly in open courtyards. Urinary cotinine was not detected in control samples while it was present in most samples from bidi rollers. In both groups of bidi rollers, workers who rolled up to 1000 bidis per day showed higher urinary thioether excretion than those who made up to 500 bidis per day.

The same authors conducted a larger study that included a greater number of subjects (Govekar & Bhisey, 1992). Among those who had no personal use of tobacco, cotinine was not detected in the urine samples of workers who did not roll bidis but was present in samples of workers who did. Among tobacco users, the levels of urinary cotinine were similar in bidi rollers and non-bidi rollers. Mean urinary thioether levels were significantly elevated among bidi rollers with or without personal use of tobacco compared with samples from the respective workers who did not roll bidis.

In another study (Bagwe & Bhisey, 1993), occupational exposure to tobacco was evident from the higher mean salivary cotinine levels that were observed in samples from bidi rollers and tobacco processing plant workers who did not report any personal tobacco use compared with their respective non-occupationally exposed counterparts.

A more recent study confirmed the findings for cotinine in saliva and urine and for thioethers (Bhisey et al., 1999a).

Nicotine and cotinine levels were measured in blood and urine samples from 10 healthy nonsmoking tobacco harvesters and five healthy nonsmoking controls at six time-points during a regular working shift (D'Alessandro et al., 2001). Maximum values of plasma and urinary nicotine were 3.45 ± 0.84 and 158.3 ± 42.5 ng/mL, respectively. The maximum values for cotinine were 20.54 ± 9.55 and 108.84 ± 47.02 ng/mL, respectively. The levels of plasma and urinary nicotine and those of urinary cotinine were significantly higher in samples from tobacco harvesters than in those from unexposed controls.

1.5. Regulations

1.5.1. Framework Convention on Tobacco Control

The first international tobacco control treaty, the Framework Convention on Tobacco Control (FCTC), was adopted unanimously by the 192 Member States of the World Health Organization in May 2003 and was opened for signature for a 1-year period. At closure, on 29 June 2004, 168 countries had signed the treaty. The Convention entered into force on 27 February 2005, 90 days after it had been acceded to, ratified, accepted and approved by 40 States. The FCTC provides a comprehensive regulatory structure for all forms of tobacco use, including smokeless tobacco (Part 1, Article 1F). Throughout the FCTC, the term ‘tobacco products’ is used to include specifically smokeless tobacco together with combusted tobacco products. The treaty will lay the legal framework in each country that ratifies the Convention for regulation to restrict or eliminate the use of any form of tobacco and to promote healthy tobacco-free lifestyles (WHO, 2003a).

1.5.2. Australia and New Zealand

In 1986, the South Australian Government became the first government in the world to ban smokeless tobacco. The ban subsequently became national in 1991 (Chapman & Wakefield, 2001).

New Zealand has also banned smokeless tobacco (WHO, 1997).

1.5.3. Europe

(a) European Union

Since 2001, smokeless tobacco has been regulated in the European Union under Directive 2001/37/EC, which supercedes Council Directive 89/622/EEC of 13 November 1989 and Directive 92/41/EEC of 15 May 1992 (European Parliament and Council, 2001). Article 2.4 of the 2001 directive defines ‘tobacco for oral use’ as “… all products for oral use, except those intended to be smoked or chewed, made wholly or partly of tobacco, in powder or in particulate form or in any combination of those forms, particularly those presented in sachet portions or porous sachets, or in a form resembling a food product.” Article 8 of Directive 2001/37/EC requires that Member States prohibit the marketing of tobacco for oral use (as defined above), but explicitly exempts Sweden and the EFTA (European Free Trade Association) country Norway. Previously, all snuff packages had to carry the health warning “causes cancer” (Directive 92/41/EEC). This was changed in the 2001 Directive, which requires that smokeless tobacco products carry the following warning: “This tobacco product can damage your health and is addictive”. The warning must cover at least 30% of the package.

Manufacturers and importers of tobacco products are required to submit to the Member States, on a yearly basis, a list of all ingredients and quantities thereof used in the manufacture of tobacco products, together with toxicological data on their effects on health and any addictive effects. This list must be accompanied by a statement that sets out the reasons for their inclusion. It must also be made public and be submitted to the Commission on a yearly basis (Article 6).

Texts, names, trade marks and figurative or other signs that suggest that a particular tobacco product is less harmful than others is prohibited on the packaging of tobacco products (Article 7).

(b) Sweden

Most regulations that govern the marketing and contents of smokeless tobacco in Sweden stem from provisions of the Swedish Tobacco Act. English language text of the provisions of the Swedish Tobacco Act is available through the website of the WHO Regional Office for Europe (WHO EURO, 2004). The Swedish Tobacco Act bans the advertisement of all tobacco products on national television, cable and radio, in local maga-zines and newspapers and in cinemas. Advertising on billboards, outdoor walls and at the point of sale are not permitted to “be invasive, enticing or encourage use of tobacco”. Businesses may not market such products as shoes and clothing if they include a tobacco trademark (brand stretching).

The Swedish Tobacco Act also regulates the contents and packaging of all tobacco products: it requires the manufacturers to list the general ingredients on each package and, in accordance with EU Directive 2001/37/EC, requires a health warning label stating “This tobacco product can damage your health and is addictive”. Sales of all tobacco products are restricted to persons aged 18 years and older and merchants are required to request purchasers to provide proof of age. Guideline No. 7 of the National Board for Consumer Policies prohibits sponsorship of events by tobacco brands.

(c) Norway

The Norwegian Tobacco Act and regulation on the prohibition of tobacco advertising contains provisions on the marketing of smokeless tobacco. A translation of this legislation is available through the website of the WHO Regional Office for Europe (WHO EURO, 2004). The Norwegian Tobacco Act and above-mentioned regulation bans all forms of advertisement of tobacco products. Tobacco products must not be included in the adver-tising of other goods and services, and all free distribution of tobacco products is prohibited. Indirect advertising of tobacco products was also forbidden as of 1 January 1996. It is prohibited to produce in or import into Norway new types of product that contain tobacco or nicotine.

The Norwegian Tobacco Act also regulates the contents and packaging of all tobacco products. The provisions require the manufacturer to provide information of the general ingredients on each package. A health warning is also required on smokeless tobacco: “This tobacco product can damage your health and is addictive”.

Tobacco products cannot to be sold to persons under 18 years of age.

1.5.4. North America

(a) Canada

The most recent regulations in Canada on information on tobacco products were enacted in June 2000 (Health Canada, 2000, 2001).

These regulations require that every manufacturer of chewing tobacco or oral snuff include one of the following bilingual warnings on every package: (a) “THIS PRODUCTIS HIGHLY ADDICTIVE” and “CE PRODUIT CRÉE UNE FORTE DÉPENDANCE”; (b) “THIS PRODUCT CAUSES MOUTH DISEASE” and “CE PRODUIT CAUSE DES MALADIES DE LA BOUCHE”; (c) “THIS PRODUCT IS NOT A SAFE ALTERNATIVETO CIGARETTES” and “CE PRODUIT N'EST PAS UN SUBSTITUT SÉCURITAIRE À LA CIGARETTE”; or (d ) “USE OF THIS PRODUCT CAN CAUSE CANCER” and “L'USAGE DE CE PRODUIT PEUT CAUSER LE CANCER”.

Every manufacturer of nasal snuff is required to display one of the following bilingual health warnings on every package: (a) “THIS PRODUCT IS NOT A SAFE ALTERNATIVE TO CIGARETTES” and “CE PRODUIT N'EST PAS UN SUBSTITUT SÉCURITAIRE À LA CIGARETTE”; (b) “THIS PRODUCT CONTAINS CANCER CAUSINGAGENTS” and “CE PRODUIT CONTIENT DES AGENTS CANCÉRIGÈNES”; (c) “THIS PRODUCT MAY BE ADDICTIVE” and “CE PRODUIT PEUT CRÉER UNEDÉPENDANCE”; or (d ) “THIS PRODUCT MAY BE HARMFUL” and “CE PRODUIT PEUT ÊTRE NOCIF”.

Every manufacturer of chewing tobacco or snuff is also required to display on every package of chewing tobacco or snuff that they manufacture the mean amount of toxic constituents (nitrosamines, lead and nicotine) contained in the product, expressed in milligrams, micrograms or nanograms per gram of chewing tobacco or snuff and determined in accordance with the official method set out for that toxic constituent.

(b) USA

Most of the current federal regulations on the marketing of smokeless tobacco products were adopted as part of the Federal Comprehensive Smokeless Tobacco Health Education Act of 1986 (Public Law 99-252), which was signed into law in February 1986 (DHHS, 1989). The Act requires that one of three warnings be displayed on all packages and advertisements (except billboards) of smokeless tobacco. The three package warnings are: “WARNING: This product may cause mouth cancer; WARNING: This product may cause gum disease and tooth loss; and WARNING: This product is not a safe alternative to cigarettes.” It requires that the three package warnings “be randomly displayed…in each 12-month period in as equal a number of times as is possible on each brand of the product and be randomly distributed in all parts of the USA in which such product is marketed.” On advertisements, the law requires rotation of each warning every 4 months for each brand. The warnings on advertisements are required to appear in a circle-and-arrow format recommended earlier by the Federal Trade Commission for cigarette warnings. The Act prohibits Federal agencies or State or local jurisdictions from requiring any other health warnings on packages and advertisements (except billboards) of smokeless tobacco. No other Federal, State or local actions were pre-empted by the Act. The Federal Trade Com-mission issued regulations implementing the law on 4 November 1986.

The Comprehensive Smokeless Tobacco Health Education Act of 1986 also required that the manufacturers, packagers and importers of smokeless tobacco products provide annually a list of additives used in the manufacture of these products to the Secretary of Health and Human Services. The Secretary is required to treat the lists as “trade secret or confidential information”, but may report to Congress on research activities concerning the health risks of these additives. However, the Secretary is granted no specific authority to regulate any of the additives. It also required that manufacturers provide to the Secretary of Health and Human Services a specification of the nicotine content of smokeless tobacco products, but it does not require that nicotine content be listed on packages or in advertisements. The list is an amalgamation of all additives used by any manufacturer in any type of smokeless tobacco product and is not brand-specific. It also contains no information on quantity or concentration of these 500 ‘ingredients' in any product. More recently, manufacturers of smokeless tobacco were required to use a standardized protocol to determine the nicotine concentration, pH and moisture content in all of their smokeless tobacco products and to provide that information annually to the CDC (1999b). Similarly to the information on product additives, however, CDC is prohibited from releasing that information to the public.

The Comprehensive Smokeless Tobacco Health Education Act of 1986 also prohibited the advertisement of smokeless tobacco products on television or radio.

The legal age at which persons can purchase smokeless tobacco in the USA is currently set at the state level. As of 1998, all states and the District of Columbia prohibit the sale of smokeless tobacco products to persons under the age of 18 years (Fishman et al., 1999). In 1992, Congress passed a provision of the 1992 Alcohol, Drug Abuse, and Mental Health Administration Reorganization Act (the ‘Synar Amendment') that addressed the access of minors to tobacco products. The final Synar regulation, issued in 1996, requires states to conduct annually random, unannounced inspections on a representative sample of retail tobacco outlets to assess the extent of sales to minors, and to show they have significantly reduced them to specified target levels (Fishman et al., 1999). On 23 August 1996, the US Food and Drug Administration issued a regulation to restrict the sale and promotion of cigarettes and smokeless tobacco products to children and adolescents (Kessler et al., 1996). The first two provisions of the regulation made it illegal for retailers to sell cigarettes or smokeless tobacco to anyone under the age of 18 years and required that they check the photographic identification of anyone under the age of 27 years. These two provisions went into effect on 28 February 1997 and remained in effect until 21 March 2000, when the US Supreme Court ruled that the Food and Drug Administration lacked the statutory authority to regulate cigarettes and smokeless tobacco (Natanblut et al., 2001). While the provision was in effect, compliance checks conducted in 110 000 establishments in 36 states and the District of Columbia found that the rate of sales to minors was higher for smokeless tobacco (38%) than for cigarettes (24%) (Clark et al., 2000).

In November 1998, the US Smokeless Tobacco Company, the largest manufacturer of smokeless tobacco products in the USA, reached a legal settlement with attorneys general for 46 states, the District of Columbia and several US territories (National Association of Attorneys General, 1998). This settlement, known as the Smokeless Tobacco Master Settlement Agreement, included a number of provisions that were intended to reduce the promotion and accessibility of smokeless tobacco products to minors. These provisions include: (a) the prohibition of the targeting of youths by advertising and promotion; (b) a ban on the use of cartoon characters in tobacco advertisements or packaging; (c) limi-tations on tobacco brand name sponsorships, including prohibition of the sponsorship of certain athletic events and concerts; (d) the elimination of outdoor advertising and transit advertisements; (e) the prohibition of payments related to tobacco products and media, including product placement in motion pictures and television; ( f ) a ban on tobacco brand name merchandise, including clothing; (g) a ban on the access of youths to free samples of smokeless tobacco; (h) a ban on gifts to under age persons based on proofs of purchase, including coupons; (i) limitations of third-party use of smokeless tobacco brand names; (j) a ban on the use of nationally recognized or established non-tobacco brand names as the brand name for a tobacco product; (k) the prohibition of the provision of tobacco products to sports teams; (l) the promulgation or reaffirmation of corporate cultural commitments related to access and consumption of youths, including the identification of an executive level manager to be responsible for identifying methods to reduce the use of tobacco by youths; (m) limitations on lobbying, including a prohibition of opposition by the US Smokeless Tobacco company to the passage of state or local legislative proposals or administrative rules that are intended to reduce access to and use of tobacco products by youths; (n) the regulation and oversight of new tobacco-related trade associations; (o) the prohibition of agreements to suppress research; and ( p) the prohibition of material misrepresentations of fact regarding the health consequences of using any tobacco product.

1.5.5. Asia

(a) Overview of regulations on tobacco in Asia

The status of regulations on tobacco products in Asia in 2003 is given in Table 66 (Shafey et al., 2003). Some countries have regulations that are related to tobacco advertisement. In 11 countries, the contents or designs of tobacco advertisements are restricted. While six countries have banned the sponsorship of events by tobacco trans-nationals, no restrictions exist in eight. Sales of tobacco to minors are not regulated in nearly one-third of the countries, and verification of age at the point of sale is not enforced in any Asian country. Other provisions that are not regulated in some countries in the region include sale by minors in 11 countries, free products in 14 countries, misleading information on packaging in 15 countries and brand-stretching in 16 countries. [Brand-stretching is defined as the use of tobacco brand names on non-tobacco merchandise or services.]

Table 66. Status of regulations on tobacco products in Asia, 2003.

Table 66

Status of regulations on tobacco products in Asia, 2003.

Few countries in Asia have comprehensive anti-tobacco laws that are strengthened by key principles such as taxation, advertising bans, smoking restrictions and effective cessation and education programmes. Egypt, Pakistan and Qatar in the WHO Eastern Mediterra-nean Region (EMRO) adopted tobacco control laws in 2002. In the WHO South-East Asian Region (SEARO) in 2003, only Thailand had a comprehensive tobacco control policy that included smokeless tobacco products (Shafey et al., 2003); India, Myanmar and Sri Lanka have since followed (WHO Tobacco Free Initiative SEARO website). The Bangladesh Act does not cover smokeless tobacco products.

A number of countries in Asia have taken initiatives specifically to control the use of smokeless tobacco (Table 67). The manufacture of all types of smokeless tobacco product is prohibited in Israel, Taiwan (China) and Thailand, while the manufacture of nasal snuff is allowed in Hong Kong (Special Administrative Region) and Singapore. The promotion of smokeless tobacco products is not permitted in Hong Kong, Singapore, Taiwan (China) or Thailand. In addition to these four states, the sale of smokeless tobacco is not allowed in Bahrain, Bhutan, Israel or Turkey. The import of smokeless tobacco products is prohi-bited in Hong Kong, Iran, Israel, Japan, Kuwait, Saudi Arabia, Singapore, Taiwan (China), Thailand and the United Arab Emirates. Regulations in India, Thailand and Turkey are detailed below.

Table 67. Available information on legislative action to control the use of smokeless tobacco in Asian countries.

Table 67

Available information on legislative action to control the use of smokeless tobacco in Asian countries.

Bans on spitting are one of the measures that may influence the prevalence of smokeless tobacco use. In Singapore and in Goa, Tamil Nadu and West Bengal in India, spitting is prohibited in public places and in Maharashtra, India, in police stations only (Table 68). However, implementation is poor in India.

Table 68. Regulation of smokeless tobacco products in selected states in India.

Table 68

Regulation of smokeless tobacco products in selected states in India.

(b) India

Legislation in India began with the promulgation of the Cigarette Act, 1975 (Regulation of Production, Supply and Distribution Act). Following the example of the state of Maharashtra in 1987, some other states (Goa, Delhi) took initiatives to prevent smoking and spitting on government premises and have conducted educational campaigns against tobacco use. In June 1999, Indian railways, which operated under the Government of India, banned the sale of tobacco on railway platforms. In September 2000, the Government amended the Cable Network Rules and banned television advertisements for tobacco. Tobacco chewing is prohibited in schools that are run by the Union Government of India.

The Cigarettes and Other Tobacco Products Act, 2003 (Government of India, 2003) prohibits direct advertising in all media and sports sponsorship by tobacco companies. It also prohibits smoking in public places. It disallows the sale of tobacco in any form to persons under 18 years of age and within 100 yards of educational institutions. It also disallows the sale of tobacco in any form by persons under 18 years of age. Clear health warnings in local languages and in English are mandatory on all packages.

Recently, beginning with Tamil Nadu in 2001, banning orders have been issued in several states against the sale, manufacture and storage of gutka and, in some states, other forms of chewing tobacco and pan masala for a certain period of time (Gupta, 2001; Gupta & Ray, 2002). The production, sale, storage, distribution and use of smokeless tobacco pro-ducts have been banned in Bihar, Andhra Pradesh, Goa, West Bengal, Tamil Nadu, Kerala, Maharashtra and Rajasthan (Table 68), but opposition by industry through the courts has forced these states to modify the ban or postpone its implementation until the Supreme Court reaches a decision.

Unmanufactured tobacco that does not bear any brand name and is used mainly for chewing is exempt from excise duty. Chewing tobacco and snuff that have a brand name are subject to 50% ad-valorem excise duty. Until 1994–95, chewing tobacco with a brand name was taxed (basic and additional excise duty tax) at 40% (Government of India, 2001; Reddy & Gupta, 2004).

(c) Thailand

Thailand has been a leader in formulating comprehensive control of tobacco, including smokeless tobacco. In 1992, the Tobacco Products Control Act B.E. 2535 was enacted with provisions to: prohibit the sale of tobacco products to persons under 18 years of age; prohibit sale promotions, e.g. exchanges, additions, offers to attend games or shows free of charge, or services to buyers or persons returning tobacco products for exchange or redemption; prohibit free samples; prohibit advertisement in all media except live broad-casts from abroad and foreign publications; prohibit the manufacture, import and advertisement of goods that imitate tobacco products and their packages. In Section 11, the composition of tobacco products must be in accordance with Ministerial Rules; and in Section 12, the packages of tobacco products must exhibit labels in accordance with the Ministerial Announcement. The Ministerial Rule pursuant to Section 11 was passed and became effec-tive on 1 February 1997. This rule mandates manufacturers to disclose the ingredients of every brand of their products to the Ministry of Public Health. The Ministerial Announcements persuant to Section 12 were passed and became effective on 25 September 1993, and another announcement became effective on 16 October 1997 (WHO SEARO, 2004).

(d) Turkey

A strong anti-tobacco law (No. 4207) was enacted in Turkey in 1996. Sales of smokeless tobacco are banned, as is the advertisement of tobacco on radio and television and in government buildings. However, advertising is permitted in print media. Indirect advertising (using tobacco or tobacco products and their brand names) and any tobacco campaign that will promote and motivate the use of tobacco or tobacco products are banned. Restrictions on the access of minors to tobacco products were strengthened by increasing the minimum age at which tobacco or tobacco products may be bought to 18 years. Turkish radio and television and private television channels have to broadcast on the harmful effects of the use of tobacco and its products for at least 90 min per month (World Bank, 2000).

1.5.6. Africa

(a) Comprehensive anti-tobacco laws

Botswana, Mali, Mauritius and South Africa have comprehensive anti-tobacco laws that are based on key principles such as taxation, advertising bans, smoking restrictions, and effective cessation and education programmes (Shafey et al., 2003).

(b) Tobacco advertisements in certain media

Only a few countries, namely Algeria, Cape Verde, Libya, Morocco, Mozambique, Niger, South Africa, Sudan and Tunisia, have banned tobacco advertising in certain media. This represents 16.6% of the 54 African countries (Shafey et al., 2003). In Algeria, advertising of tobacco has been banned since 1985. In Egypt, a complete ban on television and radio advertisements for tobacco has been in force since 1977 (WHO, 1997).

In 27 (58.7%) of the 46 countries in the WHO African Region, the contents or designs of tobacco advertisements are not regulated. While three countries (6.5%) have banned the sponsorship of events by tobacco trans-nationals, no restrictions have been imposed in 29 (63%) (Shafey et al., 2003).

(c) Other provisions

Other provisions that are not regulated in a majority of countries in Africa include sale by minors in 32 countries (70%), sales of tobacco to minors in 29 countries (63%) (verification of age at the point of sale is not enforced in any African country), free products in 31 countries (67.3%), brand-stretching in 27 countries (58.7%), misleading information on packaging in 32 countries (70%), place of sale in 31 countries (67%), health warnings and messages in 25 countries (54.3%) and the indication of the amount of contents or constituents other than tar and nicotine on packaging in 32 countries (69%) (Shafey et al., 2003).

In Uganda, excise tax on tobacco use was increased by 45% in 1993 (WHO, 1997).

None of the African countries is known to have constituted a National Tobacco Control Committee, none requires constituent disclosures for public or confidential use and none has provisions to enable litigation or measures to reduce the smuggling of tobacco.

©International Agency for Research on Cancer, 2007.
Bookshelf ID: NBK326503

Views

  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (2.9M)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...