Skip navigation

Stopping Smoking

Just recently there have been a number of superb articles on smoking from both sides of the Atlantic - the 'Human cost of tobacco' series in the New England Journal [1,2], and Doll, Peto and colleagues reporting of the 40-year follow up on British doctors [3], and associated commentaries [4,5]. Bandolier wanted to know the most effective way to stop current smokers from smoking, and this focus article centres on that issue.

The human cost of a cigarette

The human costs of using tobacco use are charted in admirable starkness by Robert Schrier and his colleagues from Colorado, prompted because the prevalence of smoking in the US has plateaued at 26% after two decades of decline. Their two reviews, containing 114 references mainly to the US literature, chart the health and economic costs.

Most people, especially health professionals, think they know the risks associated with tobacco smoking. Reading the first of these articles [1] and the report on mortality of British doctors [3], is likely to make all except the most expert think again.

These reports are not 'mindstretchers', but mindblowers. The way in which tobacco use constitutes the single largest threat to the health of the nation has been consistently understated - the more studies are performed, the greater are the number of diseases where tobacco use makes an impact.

If cigarettes cost more, then fewer people will smoke them, especially women and members of the lowest socio-economic groups where smoking prevalence is still as high as 50% [6]. Reduce prices, as happened in the UK between 1977 and 1979, and smoking goes up; increase prices, as has generally been the case since the early '80s, and smoking falls. The commitment of government to steadily increase the tax on cigarettes year-by-year should have an impact.

The economic cost of a cigarette

In the US it has been estimated that the average lifetime medical costs for a smoker are $6,000 greater than those for a non-smoker. The Congressional Office of Technology Assessment has estimated that the total financial cost of smoking to society in 1990 was $2.59 per pack of cigarettes (about £1.70 per packet at current exchange rates). There are numerous ways in which the economic costs of smoking and smoking-related diseases can be calculated, but whichever way it is calculated, all the numbers are big.

Other big numbers are seen in tobacco company profits - Philip Morris made more money in 1992 than any other US industrial corporation - $4,900,000,000. (Ironically some tobacco companies have diversified into insurance, which have higher charges for smokers than non-smokers.) Another large number is that of tobacco company spending on advertising - $2,000,000,000 a year by Philip Morris alone.

Tobacco advertising isn't Mickey Mouse!

When TV advertising of cigarettes was stopped in the US, the advertising revenue of magazines went up by an average of $5.5 million a year, and smoking actually increased.

Buying sponsorship for sport is also very effective. During the 1989 broadcast of the Marlboro Grand Prix, lasting 93 minutes, the Marlboro name was mentioned 11 times and the logo shown a staggering 5,922 times, for a total of 46 minutes of exposure of which 18 minutes was "clear, in-focus air time".

The most chilling evidence of effectiveness of tobacco advertising relates to children recognising tobacco symbols. An RJR Nabisco campaign in 1988 featured a character, possibly modelled on James Bond, called Old Joe Camel. Three years into the campaign, Camel was recognised much more by children than adults as the most frequently advertised brand. Over 30% of 3-year olds and over 80% of 6-year olds were able to associate a picture of Old Joe Camel with a packet of cigarettes - for 6-year olds about the same number who were able to associate the Disney logo with Mickey Mouse!

Campaigns to prevent tobacco use

The second of the NEJM articles [2] has a plethora of interesting facts on economics, advertising and issues relating to preventing tobacco use. Though these are directed at the US, most issues relating to reducing tobacco use are relevant, including:-
  • increased taxation.
  • comprehensive smoking bans.
  • advertising and sponsorship bans.
  • restricting sales to children.
  • financial support for counter-advertising.
  • community education programmes.

How to help the smoker

All of these issues are important parts of strategies to prevent smoking in populations. What about the individual smoker - what can be offered by healthcare professionals to help in giving up smoking, and what is effective?

Nicotine replacement therapy (NRT)

Nicotine replacement therapy is based on the idea that replacing nicotine in the body allows smoking behaviour to be stopped. A gradual weaning of the subject from nicotine follows without the pharmacological sequelae.

There have been three meta-analyses of nicotine replacement therapy published this year, looking mainly at nicotine chewing gum and nicotine patches [7-9]. Though slightly different in the range of studies included and methods used, they all had the same conclusion - that NRT is significant in helping smokers stop smoking.

Tang et al [7] analysed 28 randomised trials of 2 mg nicotine gum, six trials of 4 mg nicotine gum and six trials of transdermal patch. They used as a main outcome measure the difference between percent of control and NRT-treated patients who had stopped smoking at one year. The results showed that 2 mg nicotine chewing gum helped an extra 6% of smokers quit over controls, but this was as high as 11% in self-referred subjects and as low as 3% in invited subjects, suggesting that the desire to quit was essential.

When examined by dependency and whether patients were self referred or invited, the results showed that nicotine gum was helpful in highly dependent subjects.

The meta-analysis by Silagy et al [8] examined 53 trials, 42 with gum, nine with patch, one with intranasal spray and one with inhaler. The results were generally similar, though expressed differently. The odds ratios for abstinence were increased with use of NRT, but differently for different forms. This report also looked at the numbers needed to be treated (NNT) to obtain one extra non-smoker at 12 months beyond the number who would achieve that with the control intervention.

The nicotine patch was the particular focus of the third meta-analysis [9]. Here 17 studies were identified, with nearly 5100 subjects. At six months 22% treated with patch had stopped smoking compared with 9% for placebo. The patch type (24 vs. 16 hours), patch treatment duration (more or less than 8 weeks), weaning, nor counselling format or intensity made no difference to result. There was some evidence that intensive behavioural counselling had a modest effect on increasing rates of smoking cessation.

Unaided smoking cessation

What about all the smokers who give up on their own, without medical help; how do they do it? Lennox and Taylor [10] used postal questionnaires in Aberdeen to investigate this. The simple finding was that light and heavy smokers found it easier to give up than did moderate smokers.

Those who succeeded thought they had social support, and were more likely to have 'simply just stopped'. They were less likely to have used nicotine gum or believe that smoking was harmful. Failures experienced more withdrawal symptoms and were likely to be tempted by others smoking. Eleven percent had never tried to stop; these were older, but were more likely to stop for financial reasons.

What about other methods?

To find methods of smoking cessation which have used objective markers of smoking cessation, Bandolier performed a MEDLINE search from 1989-1994 using the words cotinine and urine. Cotinine is a metabolite of nicotine, and measurement of cotinine and hydroxyl metabolites of cotinine in urine (or blood or other body fluids) can be a useful objective test of nicotine intake. Obviously the use of gum or patch invalidates the test. While vegetables and other foods contain nicotine, huge amounts would need to be eaten to invalidate a cotinine test for active smoking, though that may not be so for passive smoking [11].

Measurement of urinary cotinine is likely to be helpful in determining the smoking behaviour in smoking cessation. Thus in a Japanese study of 49 patients who claimed to have stopped smoking, urinary cotinine concentrations one month after intensive instructions to stop smoking in a series of cardiac patients [12], only 30 had actually stopped smoking. There was a clear decrease of urinary cotinine in the quitters compared with no change in those who had not stopped.

Behaviour therapy

A German report [13] of an RCT of structured extensive behaviour therapy compared with a single unstructured antismoking advice session given by a physician in diabetic patients was disappointingly negative. Of 794 insulin-treated patients, only 89 consented to enter the study in which smoking cessation was measured objectively by urinary cotinine.

After six months, 2/44 patients randomised to the intensive behaviour therapy had stopped, compared with 7/45 who received the unstructured intervention.

Getting to mothers of infants

An attempt to use infants' cotinine levels was completely without effect [14] as a warning to mothers to reduce or stop smoking in a RCT where the physician telephoned the mother to report the urinary cotinine result and explain its meaning .

Smoking cessation in pregnancy

Because maternal smoking is associated with increased foetal risk and low birthweight, trying to prevent pregnant women smoking has top priority. One RCT [15] compared an immediate 20-minute intervention by a practice nurse with an evening class providing guidance on a self-help program for two hours on a group basis. Smoking cessation was confirmed by urinary cotinine measurement.

None of the women randomised to the intensive evening class attended, compared with 93% assigned to the immediate intervention. Rates of smoking cessation immediately after intervention, at 36 weeks gestation and postpartum were about 6% for the former and 14% for the latter.

However, there is one study which shows some success. Again, this was an RCT begun in early pregnancy, with randomisation between standard obstetric care and an intervention with self-help materials on smoking cessation in addition [16].

Self reported smoking behaviour was confirmed with a urinary cotinine test. This showed that 25% of women who said they were not smoking actually did smoke. The smoking cessation rates were not significantly different during pregnancy at about 24%, but at 8 weeks postpartum 29% of smokers had given up in the intervention group compared with under 10% in the non-intervention group. The cost of the intervention (self-help manual and audio-tapes) was $50-111 per patient.


Nicotine addiction through cigarette smoking is recognised as the largest single cause of poor health in the UK. Nicotine replacement therapy has clearly been demonstrated to help people give up smoking, and directed government policies on issues like tax are also of great value in deterring smokers.

Strategies for smoking cessation seem not to be well developed, and though NRT is unequivocally helpful, this seems to be true to a limited extent in certain smokers. It is disappointing that more effective smoking cessation strategies have not been developed in the face of the enormity of the problem.

Bandolier recognises that effective smoking cessation strategies may have been missed by its limited search. Readers who are aware of strategies of proven success are invited to send details for publication in a future issue.


  1. CE Bartecchi, TD MacKenzie, RE Schrier. The human costs of tobacco use 1. New England Journal of Medicine 1994 330: 907-912.
  2. TD MacKenzie, CE Bartecchi, RE Schrier. The human costs of tobacco use 2. New England Journal of Medicine 1994 330: 975-980.
  3. R Doll, R Peto, K Wheatley et al. Mortality in relation to smoking: 40 years' observation on male British doctors. British Medical Journal 1994 309:901-910.
  4. RM Davis. Slowing the march of the Marlboro man. British Medical Journal 1994 309: 889-890.
  5. R Peto. Smoking and death: the past 40 years and the next 40. British Medical Journal 1994 309: 933-939.
  6. J Townsend, P Roderick, J Cooper. Cigarette smoking by socioeconomic group, sex and age: effects of price, income and health publicity. British Medical Journal 1994 309: 923-927.
  7. JL Tang, M Law, N Wald. How effective is nicotine replacement therapy in helping people to stop smoking? British Medical Journal 1994 308: 21-26.
  8. C Silagy, D Mant, G Fowler, M Lodge. Meta-analysis on efficacy of nicotine replacement therapies in smoking cessation. Lancet 1994 343:139-142.
  9. MC Fiore, SS Smith, DE Jorenby, TB Baker. The effectiveness of the nicotine patch for smoking cessation. Journal of the American Medical Association 1994 271:1940-1947.
  10. AS Lennox, RJ Taylor. Factors associated with outcome in unaided smoking cessation, and a comparison of those who have never tried to stop smoking with those who have. British Journal of General Practice 1994 44: 245-250.
  11. RA Davis, MF Stiles, JD deBethizy, JH Reynolds. Dietary nicotine: a source of urinary cotinine. Food Chemistry & Toxicology 1991 29: 821-827.
  12. K Miwa, Y Miyagi, H Asanoi et al. Augmentation of smoking cessation education by urinary cotinine measurement. Japanese Circulation Journal 1993 57: 775-780.
  13. PT Sawicki, U Didjurgeit, I Mühlhauser, M Berger. Behaviour therapy versus doctor's anti-smoking advice in diabetic patients. Journal of Internal Medicine 1993 234: 407-409.
  14. BA Chilmonczyk, GE Palomaki, GJ Knight et al. An unsuccessful cotinine-assisted intervention strategy to reduce environmental tobacco smoke exposure during infancy. American Journal of Diseases of Children 1992 146: 357-360.
  15. AM O'Connor, BL Davies, CS Dulberg et al. Effectiveness of a pregnancy smoking cessation program. Journal of Obstetric, Gynecological and Neonatal Nursing 1992 21: 385-92.
  16. L Petersen, J Handel, J Kotch et al. Smoking reduction during pregnancy by a program of self-help and clinical support. Obstetrics & Gynecology 1992 79: 924-30.

next story in this issue