Publications : Books : Nicotine Addiction in Britain :

1 Tobacco smoking in Britain: an overview

• 1.1 History of tobacco use in Britain
• 1.2 Trends in smoking prevalence in Britain
• 1.3 Risk factors and determinants of cigarette smoking
• 1.4 Smoking in pregnancy
• 1.5 Morbidity and mortality caused by smoking
• 1.6 The costs of smoking in Britain
• References

1.1 History of tobacco use in Britain

Earliest records

Tobacco is a native plant of the American continent. Historians believe tobacco began growing in the Americas around 6000 BC, and that American Indians started to use tobacco as early as the 1st century BC for medicinal and ceremonial purposes. The first pictorial record of tobacco being smoked was found on Guatemalan pottery dating from between the 7th and 11th centuries AD. By the time Europeans arrived on the American continent in the 15th century, smoking of tobacco among indigenous American people was widespread,1 and native tribes were not only growing and consuming tobacco but also trading tobacco leaves.2 Christopher Columbus was given tobacco, among other gifts, by American Indians in October 1492. Tobacco seeds and leaves were then brought back to Europe for the first time.

16th and 17th centuries: the age of the pipe

Tobacco was first introduced into English society in 1565 by Captain Sir John Hawkins, though Sir Walter Raleigh is more widely credited with making smoking fashionable in Britain some 20 years later.1 In the 16th and 17th centuries, tobacco was commonly smoked in pipes, and although initially a prerogative of the rich, the habit gradually spread to all sections of society. Tobacco smoking remained almost exclusively a male habit, at least in public, until the 19th century.

The early growth in popularity of smoking tobacco was largely due to its supposed healing properties.1 In 1571, a Spanish doctor, Nicholas Monardes, wrote a book on the history of medicinal plants of the New World in which he claimed that tobacco could cure 36 health problems, including toothache, worms, lockjaw and cancer. However, by the early 1600s the reported benefits of smoking were beginning to be questioned, most notably by King James I (James VI of Scotland) who, in 1604, produced a damning report entitled 'Counterblaste to Tobacco', in which he said that smoking is a

custome loathsome to the eye, hateful to the nose, harmful to the brain, [and] dangerous to the lungs.1

To discourage the smoking habit, King James increased the import tax on tobacco by 4,000%, and consumption fell dramatically as a result - but, after realising the effect of such a punitive level of duty on revenue from the tobacco trade, he subsequently reduced this tax and consumption rose again.

The first suggestion that tobacco smoking might be addictive was reported in 1610 when Sir Francis Bacon noted that trying to quit the habit was very difficult.1 Some 70 years later, a lawyer, John Selden, observed that, just as some dislike sermons but might learn to enjoy them, the same applied to

that which is the great pleasure of some men, tobacco; at first they could not abide it, and now they cannot be without it.2

Commercial cultivation of tobacco in America began in the early 17th century, and the first successful shipload of the new Virginian tobacco was sent to England in 1616.1 In 1619, London clay pipe makers formed a chartered body. In the following year, a trade agreement between the Crown and the Virginia Company banned commercial growing of tobacco in England in return for a duty on imported Virginia tobacco of 1 shilling per lb weight.1 By the late 1620s, an estimated 500,000 lb of tobacco were being brought into London every year, with smaller amounts going into the west coast ports,2 and by the mid-1660s trade in tobacco between America and Europe had became a major business.2 Tobacco use was now common among all sectors of society, and tobacco was sold in taverns, apothecaries' and tobacconists' shops. It was mostly in taverns that poorer people learnt to smoke, usually by sharing a communal pipe.

The 18th century: the age of snuff

In 1730, the first American tobacco factories were constructed, and in 1760 Pierre Lorillard established a factory in New York producing pipe tobacco, cigars and snuff.1 During the 18th century, the trade in tobacco between Britain and America escalated, such that by 1770 approximately 96,000 hogsheads of tobacco (each representing a gross weight of around 600 kg) were imported into Britain each year, of which Britons consumed around 14,000 hogsheads. The rest was re-exported to Continental Europe, thus establishing England's role as a major tobacco trade centre. Following the restoration of the monarchy in 1660, the courtiers of Charles II returned to London from exile in Paris, and brought the French court's practice of snuff taking with them. For a period thereafter, snuff gradually replaced pipe smoking as the aristocratic form of tobacco use.1

While the demand for tobacco grew inexorably, some physicians began to warn of the potential dangers. In 1701, Nicholas Andryde Boisregard warned that young people taking too much tobacco have trembling, unsteady hands, staggering feet and suffer a withering of 'their noble parts'. In 1761, Dr John Hill performed possibly the first clinical study of tobacco's effects, and noted that snuff users were vulnerable to cancers of the nose. Thirty years later, he again reported cases in which snuff use had caused nasal cancer. In 1795, Samuel Thomas von Soemmering of Maine reported on cancers of the lip in pipe smokers.1

The 19th century: the age of the cigar

By the end of the 18th century, snuff-taking was in decline and a revival in smoking had begun, this time in the form of cigars rather than pipes. The fashion for cigars had spread from Spain, and with it came a smoking etiquette amongst the upper classes, whereby smoking tended to be confined to certain parts of the house, and gentlemen generally did not smoke in female company. Smoking by ladies was still frowned upon, though working-class women were known to use snuff or smoke clay pipes.2 In 1826, the pure form of nicotine was discovered. By the 1860s, as smoking grew in popularity, it became more acceptable to smoke in public places. English railways introduced smoking carriages on trains, and public houses opened 'smoking saloons'.2 Between the 1830s and 1870s, annual tobacco consumption almost doubled from 14 oz to 24 oz (ca 400-700 g) per head.2 Although popular among the merchant and upper classes, smoking cigars remained an expensive habit, and poorer people still tended to smoke pipes or use snuff.2

It was the invention of the manufactured cigarette that transformed tobacco smoking into a truly mass habit. The origins of the cigarette lay in the Crimean war (1853-1856) when British soldiers copied the habit of hand-rolling tobacco from their Turkish allies. In 1854, London tobacconist Philip Morris began making hand-rolled cigarettes. The first cigarette factory opened in England in 1856, followed in 1871 by the establishment of Wills factory in Bristol and in 1888 by Player's in Nottingham. At the same time in the US, James 'Buck' Duke, the founder of American Tobacco, entered the cigarette manufacturing business.1

The 20th century: the age of the cigarette

James Duke was the first tobacco entrepreneur to use cigarette-making machines, and thus to initiate a revolution in the scale of cigarette production. In 1890, Duke formed The American Tobacco Company and set about taking over rival tobacco companies. He also set his sights on Britain, where a number of successful businesses such as Wills of Bristol had begun to use cigarette machines, producing around 85,000 cigarettes a day.3 In 1901, in response to the threat of an American take over, 13 British tobacco companies joined together to form the Imperial Tobacco Company. A year later, The American Tobacco Company and Imperial came to an agreement to stay in their own countries and unite to form the British American Tobacco company to sell both companies' brands abroad.3 Sales of manufactured cigarettes increased so rapidly from 1895 that by 1919 they accounted for more sales by weight than all other forms of tobacco combined.4 Cigarette smoking among men increased further during World War I when cigarettes were included in soldiers' rations. As a result, many of the soldiers who returned home had become regular cigarette smokers.

It remained socially unacceptable for women to smoke until the Suffragette movement in the 1920s, when significant numbers of women started smoking.5 The tobacco industry then began to market brands to women, using imagery associated with power and liberation; as a result, cigarette smoking escalated rapidly among both sexes during the 1930s and 1940s. By this stage, the British tobacco smoking epidemic was in full force.

1.2 Trends in smoking prevalence in Britain

The prevalence of tobacco smoking in Great Britain has changed dramatically during the 20th century. No direct measures of smoking prevalence are available for the first half of the century, but tobacco industry sales data demonstrate a greater than twofold increase in consumption of tobacco products during this period, from 4.1 g per adult per day in 1905 to a peak of 8.8 g in 1945 and 1946.4 These peak levels of consumption have since been equalled (in 1959) but never exceeded.4 The first available measure of smoking prevalence from tobacco industry sources is for 1948, when consumption was at 7.1 g per adult per day. At that time, an estimated 65% of men and 40% of women in Britain were regular smokers of manufactured cigarettes, and a further 16% of men smoked other tobacco products.4 Industry figures suggest that male smoking declined progressively over the next two decades, whilst female smoking rose to a peak of 44% in 1966 and again in 1969 (Fig 1.1).

Since 1972, the prevalence of smoking in Britain has been measured regularly and independently from the tobacco industry in nationally representative population samples as part of the General Household Survey (GHS).6 In the first year of the GHS, 52% of men and 41% of women in Britain were regular cigarette smokers, and a further 13% of men were smoking other tobacco products.6 The prevalence of smoking then declined progressively in both sexes until 1994, when 28% of men and 26% of women were regular cigarette smokers. However, in 1996 (at the time of writing, the latest year for which GHS data have been published), the estimate of smoking prevalence had increased again in both men and women to 29% and 28%, respectively - the first increase in estimated smoking prevalence observed in Britain in either sex for 30 years (Fig 1.1).6 Smoking prevalence estimates for late 1997 are available from the Omnibus Survey,7 which in previous years has produced figures similar to those from the GHS. These data suggest that by 1997 smoking prevalence had fallen again in both men and women to 26% and 27%, respectively. Preliminary unpublished data for 1998, released by the Office for National Statistics shortly before going to press, give prevalence estimates of 28% for men and 26% for women.

Overall, these estimates suggest that the prevalence of smoking in Great Britain may have been beginning to stabilise in recent years at a level of about one in four adults. However, inspection of age-specific rates in adults reveals that smoking prevalence has in fact been stable or increasing in recent years amongst most of the younger age groups, particularly in women (Fig 1.2). Similarly, inspection of trends in smoking amongst 15 year olds in England also shows a progressive increase during most of the 1990s,8 again more marked in females (Fig 1.3). The rate at which new smokers are joining the prevalent smoking population has therefore been increasing for some years, making it likely that unless cessation rates also begin to increase, the overall prevalence of smoking in Great Britain will soon again begin to rise. Trends in smoking cessation rates can be inferred from trends in the proportion of ex-smokers amongst UK adults. These showed a progressive increase between 1970 and 1990, but have since remained remarkably stable (Fig 1.4). This suggests that cessation rates are no longer increasing, and that the prevalence of cigarette smoking in Britain is indeed in danger of increasing again over the next few years.

Fig 1.2. Age-specific smoking prevalence in Great Britain, 1986-1996: (a) males; (b)
females (Source: General Household Survey6)

Fig 1.3. Prevalence of regular smoking (at least one cigarette per week) amongst 15 year
old boys and girls in England, 1982-1996 (Source: Office for National Statistics
smoking among secondary schoolchildren surveys, 1982-1996 ⁸)

1.3 Risk factors and determinants of cigarette smoking

Gender

The effect of gender on the likelihood of being a smoker is changing. As described above, smoking in Britain has been more common in men for most of this century, but the difference in prevalence between men and women has been decreasing for many years and in 1997 the point estimate of cigarette smoking prevalence was actually slightly higher in women than in men.7 This estimate did not allow for cigar or pipe smoking, so overall in Britain men are probably more likely than women to be smokers, but the difference between the sexes is now very small.

A trend towards female smoking has been evident for several years amongst schoolchildren, and the gap between the sexes has been increasing (Fig 1.5).8 For some time therefore, females have accounted for the majority of young smokers entering the smoking population. Although the future relative prevalence of smoking in young male and female adults will depend on uptake and cessation rates in both sexes during the later teenage years, these data indicate that the proportion of females in the smoking population may be set to increase further.

Age

Age is a major determinant of smoking behaviour. Smoking is very uncommon in children up to and including the age of 11 years, but increases substantially at 12-15 years old, to the extent that in 1996 28% of boys and 33% of girls were regular smokers by age 15 (Fig 1.6).8 Amongst adults, smoking prevalence is greatest in the 20-24 age group (Fig 1.2), thereafter decreasing progressively with age.

Fig 1.5. Prevalence of regular smoking (at least one cigarette a week) amongst boys
and girls aged 11-15 in England, 1982-1996 (Source: Office for National Statistics
smoking amongst secondary schoolchildren surveys, 1982-1996 ⁸).

Socio-economic status

Smoking behaviour is strongly related to socio-economic status. In relation to occupation in 1996, smoking prevalence was lowest in the professional (12%) and highest in the semi-skilled manual occupational groups (39%) (Fig 1.7).8 Data on the trend in smoking prevalence within non-manual and manual occupational groups suggest that the difference between these groups has, if anything, widened in recent years, more so in women (Fig 1.8). However, other measures of relative poverty or deprivation, including housing tenure, crowding, living in rented accommodation, being divorced or separated, unemployment, low educational achievement, and in women, single parent status, are also independently associated with an increased risk of smoking amongst adults.9 Analysis of trends in smoking based on a composite index of some of these measures indicates that over the period 1973-1996 smoking prevalence fell by more than 50% in the most advantaged sector of British society, but has remained unchanged in the most deprived group.9 Similar findings apply to smoking cessation rates, which also show a strong inverse relation with deprivation. Cessation rates have doubled in the most advantaged groups, but have remained almost unchanged over the past two decades in the most disadvantaged sectors of society.10

Fig 1.7. Prevalence of regular smoking by occupational group in males and females aged
16 or over in 1996 (Source: General Household Survey 1996)

Fig 1.8. Prevalence of regular smoking by adults aged 16 and over by occupational
group and gender in England, 1982-1996 (Source: General Household Survey 1986-96)

Region of residence

Smoking prevalence varies in the regions. Data for NHS Regional Office areas of England reveal that the highest prevalence is in the North West Region (30%) and the lowest in the South and West Region (25%) (Fig 1.9).8

Risk factors for smoking in children

The factors associated with smoking in children broadly reflect those established for adults. Recent survey data11 from children aged 11-15 years in England identify several factors associated with the likelihood of smoking in children, including:

• low educational achievement: children who are planning to take GCSE examinations, but with an expectation of passing in fewer than five subjects, are more than twice as likely to be smokers (26%) than those with higher expectations (10%);
  
• living with parents who smoke: children living with two parents who both smoke are nearly three times as likely to be smokers than those whose parents do not smoke, an effect particularly marked in girls (Fig 1.10);
  
• having siblings who smoke: children who have at least one sibling who smokes are four times more likely to smoke (26%) than those with no siblings who smoke (6%).

The following have been identified as additional potential risk factors:

• low socio-economic status
  
• having friends who smoke
  
• having teachers who smoke.12

1.4 Smoking in pregnancy

Smoking during pregnancy is a problem of particular importance because of the harm that maternal smoking causes to the unborn child. The adverse effects of smoking during pregnancy have been reviewed in detail elsewhere,12,13 but include spontaneous abortion, preterm birth, low birth weight and stillbirth. The children of mothers who smoke during pregnancy are at increased risk of neonatal mortality or sudden infant death syndrome, of asthma and/or wheezing illness in the first years of life, and they subsequently experience impaired physical growth and academic attainment compared to children of non-smoking mothers. These adverse effects are all imposed involuntarily and are, in principle, entirely avoidable.

Trends in smoking in pregnancy

In contrast to the overall trends in smoking in the general population, surveys of pregnant women conducted since 1992 by the Health Education Authority (HEA) in England suggest that smoking levels in this population have remained virtually unchanged. Published data from these surveys from 1992 to 1997 demonstrate little overall change in prevalence,14 but subsequent data for 1998 and 1999 (HEA, data previously unpublished) indicate that the prevalence of smoking in pregnancy may be rising (Fig 1.11). In 1999, 30% of pregnant women were smokers.

Determinants of smoking in pregnancy

Smoking during pregnancy is associated with many factors, particularly age, social class, education, marital status, presence of other smokers in the home, high parity, employment and ethnicity. An analysis of combined data from the HEA surveys described above confirms that smoking is twice as common amongst 15-24 year old pregnant women (42% smokers) than in those aged 35 and over (21% smokers), and that women in unskilled manual or unemployed groups were nearly six times more likely to smoke than those in professional and non-manual groups (45% and 8%, respectively) (Fig 1.12). Women who left full-time education at an early age were also much more likely to continue to smoke during pregnancy than other women (Table 1.1).

Relative to pregnant women as a whole, single/separated/divorced pregnant women and pregnant women who cohabit were also more likely to smoke (27%, 51% and 42%, respectively). In contrast, married women were less likely to smoke during pregnancy (17%) than pregnant women as a whole. Importantly, pregnant women with partners who smoke were four times more likely to be smokers themselves than were those with non-smoking partners (49% and 11%, respectively; 1999 survey). Family size was also associated with smoking during pregnancy: 23% of pregnant women with no children or one child were smokers, compared with 31% of pregnant women with two children, 37% with three children, and 45% with four or more children. Housing tenure was also related to smoking status, with 51% of pregnant women in rented council accommodation smoking compared with 18% of those who own their housing.

Smoking cessation in pregnancy

Although a significant number of women continue to smoke during pregnancy, many do make changes to their smoking behaviour around this time. The 1999 HEA survey found that 10% of women who were smoking immediately before pregnancy stopped and 4% cut down immediately beforehand. During pregnancy, however, more cut down than quit (33% and 20%, respectively). Even amongst those who managed to stop smoking either immediately before or during pregnancy, 19% had relapsed whilst still pregnant. These percentages have changed little over the last seven years.

The above findings show that smoking during pregnancy is a problem, particularly for women who are young, unemployed or from manual occupational groups, who left full-time education at a young age or are living in rented accommodation. The pattern of smoking in pregnancy therefore broadly reflects that in the general population, and demonstrates that interventions intended to reduce the prevalence of smoking in pregnant women, as in the general population, need to be directed especially towards socially disadvantaged groups. Recent trends in prevalence suggest that without new and effective interventions on smoking in pregnancy the target set in the recent White Paper, Smoking kills, is unlikely to be met:

to reduce the percentage of women who smoke during pregnancy from 23% to 15% by the year 2010; with a fall to 18% by the year 2005.15

On a more positive note, however, many pregnant women do attempt to change their smoking behaviour, particularly in the very early stages of pregnancy. This highlights the potential for supporting pregnant women to stop smoking - and stay stopped - provided that adequate resources are made available.

1.5 Morbidity and mortality caused by smoking

Cigarette smoking causes substantial mortality and morbidity, the extent of which has been reviewed extensively elsewhere,16-18 most recently for mortality in the UK in 1995 by Callum.19 To provide more recent estimates of mortality and morbidity for the purposes of this report, these analyses have been repeated using updated 1997 UK mortality data and 1996 smoking prevalence data obtained from the sources previously described.19 A similar approach has also been applied to estimate UK hospital admissions and general practitioner (GP) patients consulting in 1997-98 (April 1997-March 1998) from data provided to us by the Department of Health, the Office for National Statistics and the General Practitioner Research Database, and from equivalent sources in Wales, Scotland and Northern Ireland.

Deaths from smoking

In 1997, cigarette smoking accounted for an estimated 117,400 of the total of 628,000 deaths in the UK. Cigarette smoking is thus responsible for approximately one in every five deaths in Britain. This annual mortality translates into an average of 2,300 people killed by smoking every week, 320 every day and 13 every hour. The proportional impact of smoking is greater in younger age groups and in men, accounting for one in three male deaths in the 35-64 age group (Fig 1.13).

Years of life lost

Numbers of deaths do not convey a full sense of the loss to the community caused by smoking, since this depends not only on how many people die, but also how 'premature' their deaths are. Based on the distribution of deaths from smoking by age and mortality risks in never-smokers, we estimate that in 1997 cigarette smoking accounted for the loss of 205,000 years of life under age 65 and 551,000 years of life under age 75.

Deaths from smoking by disease

The number of deaths from individual causes, and the percentage of deaths attributable to smoking by cause for 1997 are summarised in Table 1.2. More than half of all smoking related deaths were due to respiratory disease (Fig 1.14). Smoking caused the majority of deaths from lung cancer, accounting for 89% of deaths from this disease among men and 74% among women. Similarly, 86% of deaths from chronic obstructive pulmonary disease (COPD) for men and 80% in women were attributable to smoking, as were approximately 17% of deaths from pneumonia. Cigarette smoking caused an estimated 17% of deaths from ischaemic heart disease, and 10% from stroke.

The estimated total of 117,400 deaths is a net figure which takes account of the small number of deaths from diseases prevented by smoking, principally from Parkinson's disease and endometrial cancer. Altogether, an estimated 118,800 deaths were caused by smoking and 1,400 deaths prevented. In 1997, therefore, smoking caused 85 times more deaths than it prevented.

Mortality and the individual smoker

Life-tables by smoking status, based on UK 1997 death rates and attributable percentages, can be constructed to portray the burden for the individual smoker. By this approach, a 35 year old man who smokes cigarettes can expect to die, on average, more than seven years earlier than a man who has never smoked (Fig 1.15). Cigarette smoking shortens life expectancy at age 35 years for women by six years compared with a woman who has never smoked cigarettes. Even for those who survive to age 65, cigarette smoking curtails their expected lifespan by more than six years among men and 5.5 years among women. The figures for ex-smokers lie between the two, closer to never-smokers than to current smokers. More than one in four men aged 35 who continue to smoke cigarettes can expect to die before age 65 compared to one in nine of never-smokers. The equivalent estimates for women are one in seven and one in 12, respectively. Overall, approximately one in every two smokers (51% of males and 45% of females) will die prematurely as a result of their smoking.

Admissions to hospital

In 1997-8, an estimated 364,000 hospital admissions in England were attributable to the diseases caused by smoking listed in Table 1.3. This translates into 7,000 hospital admissions per week, or 1,000 per day. The admissions are spread fairly evenly across the main groups of diseases caused by smoking (Fig 1.16):

• 109,000 from cancer
  
• 112,000 from respiratory disease other than lung cancer
  
• 134,000 from circulatory disease.

The major fatal diseases caused by smoking (listed in Table 1.2) were responsible for 6% of hospital admissions among those aged 35 years or over (9% for men and 4% for women). This understates the overall impact of smoking on hospital admissions to the extent that non-fatal diseases caused by smoking are excluded from these estimates.

Table 1.3. Estimated number and percentage of UK hospital admissions attributable to smoking by cause, 1997-98.

General practitioner consultations

In primary care in 1997/8, cigarette smoking caused an estimated 480,000 patients to consult their GP for ischaemic heart disease, 20,000 for stroke and nearly 600,000 for COPD.

Conclusions

The burden of premature mortality and morbidity caused by smoking in Britain is massive. No other single avoidable cause of disease accounts for such a high proportion of deaths, hospital admissions or GP consultations. Cigarette smoking is the single most important public health problem in Britain.

1.6 The costs of smoking in Britain

The harmful effects of smoking tobacco can also be considered from an economic perspective and expressed in money terms or as costs. Such costs can include the effects on the individual as well as those borne by the wider community.

The health burden of smoking

One of the main costs of tobacco smoking is the health burden it creates. For most purposes, for example in the evaluation of new medical therapies, the loss of quality and quantity of life is not directly valued, rather measures such as quality adjusted life years are adopted. However, there are monetary equivalents with, for example, the Department of Transport putting the value of the human costs of a loss of life in 1997 at £680,590.20 If this value per life is applied to the total number of deaths attributable to smoking in Table 1.2, the cost of smoking related mortality in the UK in 1997 prices is just under £80 billion. This estimate does not take account of the value to the individual of smoking related illness, or of related loss of quality of life. Clearly, while smokers know of the risks to health and may set aside some of these individual costs against the 'benefits' they receive from smoking, the majority of smokers wish to quit. Many smokers will spend considerable sums of money on a variety of smoking programmes before successfully stopping. The human costs of smoking expressed in money terms are thus considerable.

The costs of smoking to society

The costs of smoking to the rest of society can be divided into:

• the costs of the harmful effects of passive smoke exposure in non-smokers
  
• the costs imposed by smokers on the wider community through, for example, the use of scarce health service resources or by lower productivity in the workplace.

Passive smoking. The health risks arising from passive smoking have a direct impact on family members, especially on children and the unborn, and in the workplace. As with the direct health effects on smokers, a high value could be placed on any premature loss of life or passive smoking related illness. There can be considerable financial consequences for the health and other welfare agencies from a premature birth. Models of the cost-effectiveness of smoking cessation interventions with pregnant women indicate that these are potentially cost saving because of these high costs. Children living with smokers also have poorer health and higher health care expenditures than those in non-smoking households. Stoddard and Gray21 estimated that passive smoking was responsible for 19% of all expenditures for childhood respiratory conditions in the US. In Hong Kong, it has been estimated that the cost per child of GP consultations was 14% higher for children living with one smoker at home and 25% higher where there were two or more smokers.22 Godfrey et al23 estimated that the cost to the NHS of the effects of passive smoking on children was £410 million in England and Wales.

Fires caused by smoking are another effect which may affect both smokers and non-smokers. Buck and Godfrey24 estimated the costs of fires at £150 million for England and Wales in 1991. This figure excludes any value on the loss of life from such fires.

Health service resources. Clearly, there are effects on the health services from smoking related illness. Two general approaches can be used to estimate these costs to the NHS:

• by attributing the costs of different diseases to smoking
  
• by estimating the different health care costs of smokers compared to non-smokers.

Parrott et al25 used both methods to estimate the annual smoking related costs in England in 1996/1997 prices. In the first approach, estimates of smoking related, attributable fractions were applied to NHS costs by disease code. The total estimated cost was £1.5 billion for England. The breakdown by disease and type of NHS expenditure is shown in Table 1.4. The alternative method combined data on health care utilisation by current smokers and never-smokers, taken from the GHS with unit costs for different types of health care use from a variety of sources (see Ref 25 for details), and reached a similar total of £1.4 billion. The breakdown for different types of health service use estimated by this method are summarised in Table 1.5. The two methods yielded similar totals, but the second shows slightly higher primary care costs.

Table 1.4. Estimated smoking related costs to the NHS by disease group, England 1996-1997.25

Table 1.5. Health care costs estimated from observed differences
between smokers and non-smokers, England 1996-1997.25

These types of figures for smoking related health care expenditure are based on annual estimates. It is more difficult to estimate how such figures may change over time if smoking prevalence declines. However, there is clearly a large potential for NHS resources currently being used to treat smoking related problems to be diverted to treating other problems. It is sometimes argued that, although smokers consume more health care throughout their lifetime, their life expectancy is shorter. Authors have attempted to calculate whether smokers or non-smokers have the higher total lifetime health care expenditures. Results from studies have varied (see, for example, Refs 26 and 27), and we have no estimate for the UK. The implications from such studies are not clear. There may be some fairness issues involved, for example, do smokers pay their way? Similar arguments surround the balance of tax paid by smokers and benefits received.28

There are, however, more conflicting views as to whether the health care consequences of 'unrelated' conditions should be included in any analysis of the worth of interventions such as smoking cessation programmes. Extending these arguments to all types of health care conditions would suggest that there are 'costs' both to any life prolonging intervention and to most preventive measures. This would especially disadvantage interventions which save the lives of younger people, and it seems unlikely that such arguments would be upheld by society. Such an analysis obviously fails to acknowledge the benefits to society of people living longer. Also, if differential timing is taken into account by discounting or putting a lower weight on future events, the result would seem to indicate that smokers do have lifetime excess health care costs.27 This suggests that such empirical estimates are a diversion from the main point that smoking cessation interventions can yield high health gains at low cost.

Another area where smokers can have a major impact is the workplace. Smokers, in general, have more sickness absences, and in workplaces where smoking breaks are allowed there can also be a considerable loss in productivity. Parrott et al29 estimated that total productivity losses in Scotland due to smoking in the workplace are almost £400 million per year. Many smokers have to give up work because of smoking related illness. Buck and Godfrey24 estimated that in England and Wales in the year to March 1991, 34 million working days were lost from smoking related illness, at an estimated cost of £328 million.

Money figures are one way of trying to summarise the wide-ranging effects of smoking on society. It is difficult to give a total UK annual burden of smoking. Figures quoted above are taken from different studies at different time periods and covering different parts of the UK. There are also some missing figures, for example, of the cost to the unborn. Also, opinions vary as to the items which should be included or excluded in such a total figure. However, irrespective of the effect of these considerations on the total figure, it is clear that cigarette smoking has major health and economic impacts in Britain.

References

1. Borio G. Tobacco timeline, 1993-1998. www.tobacco.org/History/Tobacco_History.html.
   
2. Kiernan VG. Tobacco: a history. London: Hutchinson Radius, 1991.
   
3. Taylor P. The smoke ring: tobacco, money and multinational politics. London: Sphere Books, 1985.
   
4. Nicolaides-Bouman A, Wald N, Forey B, Lee P. International smoking statistics. Oxford: Oxford University Press, 1993.
   
5. Vierola H. Tobacco and women's health. Helsinki: Art House Oy, 1998.
   
6. Office for National Statistics. Living in Britain. London: The Stationery Office, 1997.
   
7. Freeth S. Smoking-related behaviour and attitudes, 1997. A report on research using the Omnibus Survey produced on behalf of the Department of Health. London: Office for National Statistics, 1998.
   
8. Statistics on smoking: England, 1976 to 1996. Department of Health Bulletin 1998/25. London: Department of Health, 1998.
   
9. Jarvis MJ, Wardle J. Social patterning of individual health behaviours: the case of cigarette smoking. In: Marmot M, Wilkinson R (eds). Social determinants of health. Oxford: Oxford University Press, 1999: 240-55.
   
10. Jarvis MJ. Patterns and predictors of smoking cessation in the general population. In: Bolliger CT, Fagerström KO (eds). The tobacco epidemic. Basel: Karger, 1997: 151-64.
    
11. Barton J. Young teenagers and smoking in 1997. London: Health Education Authority, 1998.
    
12. Smoking and the young. A report of a working party of the Royal College of Physicians. London: RCP, 1992.
    
13. Charlton A. Children and smoking: the family circle. Br Med Bull 1996; 52: 90-107.
    
14. Owen L, McNeill A, Callum C. Trends in smoking during pregnancy in England, 1992-7: quota sampling surveys. Br Med J 1998; 317: 728.
    
15. Smoking kills. A White Paper on tobacco. London: The Stationery Office, 1998.
    
16. Doll R, Peto R, Wheatley K, Gray R, Sutherland I. Mortality in relation to smoking: 40 years' observations on male British doctors. Br Med J 1994; 309: 901-11.
    
17. Peto R. Smoking and death: the past 40 years and the next 40. Br Med J 1994; 309: 937-9.
    
18. Peto R, Lopez AD, Boreham J, Thun M, et al. Mortality from smoking worldwide. Br Med Bull 1996; 52: 12-21.
    
19. Callum C. The UK smoking epidemic: deaths in 1995. London: Health Education Authority, 1998.
    
20. Department of Transport. Valuation of the benefits of prevention of road accidents and casualties. Highways Economic Note No 1, 1997. London: Department of the Environment, Transport and the Regions, 1998.
    
21. Stoddard JJ, Gray B. Maternal smoking and medical expenditures for childhood respiratory illness. Am J Pub Health 1997; 87: 205-9.
    
22. Peters J, McCabe CJ, Hedley AJ, Lam TH, Wong CM. Economic burden of environmental tobacco smoke on Hong Kong families: scale and impact. J Epidemiol Community Health 1998; 52: 53-8.
    
23. Godfrey C, Edwards H, Raw M, Sutton M. The smoking epidemic: a prescription for change. Health Education Authority: London, 1993.
    
24. Buck D, Godfrey C. Helping smokers give up: guidance for purchasers on cost-effectiveness. Health Education Authority: London, 1994.
    
25. Parrott S, Godfrey C, Raw M, West R, McNeill A. Guidance for commissioners on the cost-effectiveness of smoking cessation interventions. Thorax 1998; 53(Suppl 5, Part 2): S1-38.
    
26. Hodgson TA. Cigarette smoking and lifetime expenditures. Millbank Q 1992; 70: 81-125.
    
27. Barendregt JJ, Bonneux L, Van De Maas PJ. The health care costs of smoking. N Engl J Med 1997; 337: 1052-7.
    
28. Cohen D, Barton G. The cost to society of smoking cessation. Thorax 1998; 53(Suppl 2): S38-42.
    
29. Parrott S, Godfrey C, Raw M. Cost and benefit analysis of smoking cessation in the workplace. Final report to the Health Education Board for Scotland. University of York: Centre for Health Economics, 1996.

Contributors, Foreword and Key Points
1  Tobacco smoking in Britain: an overview
2 Physical and pharmacological effects of nicotine
3 Psychological effects of nicotine and smoking in man
4 Is nicotine a drug of addiction?
5 The natural history of smoking: the smoker's career
6 Regulation of nicotine intake for smokers, and implications for health
7 The management of nicotine addiction
8 Regulatory approaches to tobacco products in Britain
9 Summary and recommendations

This page last updated on August 1, 2007