HEALTHY LIVING

Health Evidence Bulletins - Wales (logo)

Literature searches completed on 27.7.99

Chapter 1. Smoking

This bulletin is a supplement to, not a substitute for, professional skills and experience. Users are advised to consult the supporting evidence for a consideration of all the implications of a recommendation.
Smoking is the single greatest cause of preventable illness and premature death in the UK, causing 120,000 deaths each yeari.
i. Callum C. The UK smoking epidemic: deaths in 1995. London: Health Education Authority, 1998 (Type IV evidence – statistics)
The Statements The Evidence
1.1 Background
1.1a. The British Doctors Study estimated the relative risk of dying from lung cancer as 14.9 in smokers compared to non-smokers, from chronic obstructive pulmonary disease 12.7, from ischaemic heart disease 1.6, cerebral thrombosis 1.3, cerebral haemorrhage 1.4 and from aortic aneurysm 4.1. For most of the causes of death, the relative risk rises with the number of cigarettes smoked and the duration of smokingi. i. Doll R, Peto, R, Wheatley, Gray R, Sutherland I. Mortality in relation to smoking: 40 years’ observations on male British doctors. British Medical Journal 1994; 309: 901-911
http://www.bmj.com/cgi/content/full/309/6959/901
(Type IV evidence – very large cohort study with long-term follow up)
1.1b. Excess death rates, in smokers compared to non-smokers, were observed for cancers of the mouth, oesophagus, pharynx, larynx, lung, pancreas, and bladder; from chronic obstructive pulmonary disease and other respiratory diseases; from ischaemic heart disease, stroke and other vascular diseases; and peptic ulcers. Most of these associations are thought to be causal in parti. i. Doll R, Peto, R, Wheatley, Gray R, Sutherland I. Mortality in relation to smoking: 40 years’ observations on male British doctors. British Medical Journal 1994; 309: 901-911 http://www.bmj.com/cgi/content/full/309/6959/901
(Type IV evidence – very large cohort study with long-term follow up)
1.1c. Environmental tobacco smoke (ETS, passive smoking) is a cause of lung cancer in never-smoking work colleagues, especially in women, and lung cancer and ischaemic heart disease in never-smoking spouses (especially women). ETS from parental smoking increases the risk of lower respiratory tract illness in infants. It worsens symptoms in children with asthma and increases the risk of pneumonia, bronchitis and bronchiolitis in children under 5. The risk increases with the number of smokers in the homei. i. Health Evidence Bulletins. Health Evidence Bulletins-Wales: Healthy Environments. Cardiff: Welsh Office,1998
http://hebw.cardiff.ac.uk/healthyenvironments/chapter5.html
(Summaries of evidence classified by evidence type)

 

 

Top

1.1d. In the UK cigarette consumption is 25% higher than the EU average, although smoking rates are similar to the EU averagei. The adult smoking rate is higher in Wales and Scotland than in Northern Ireland and England. In Wales the proportion of 18-64 year olds who reported smoking daily fell from 32.5% in 1985 to 26.9% in 1996ii. i. Smoking Kills - a White Paper on Tobacco. Cm 4177. London: The Stationery Office, 1999
http://www.official-documents.co.uk/document/
cm41/4177/4177.htm

(Type V evidence – expert opinion citing type IV evidence, statistics)
ii. Lifestyle Changes in Wales: Health in Wales Survey 1996. Technical Report no 27. Cardiff: HPW, 1998
(Type IV evidence – large population based cross-sectional surveys with random samples)
1.1e. More than a quarter of Welsh school girls smoke at least once a week – more than in most European Union countriesi. At all ages between 11 and 15, girls are more likely to smoke than boys, but boys who smoke consume more cigarettes than girlsii. i. World Health Organisation Europe. The Health of Youth. A Report of the 1993/1994 Survey Results of Health Behaviour in School-Aged Children: A WHO Cross National Study. Copenhagen:WHO, 1996
(Type IV evidence – survey of European countries)
ii. Office for National Statistics. Young Teenagers and Smoking in 1997. London: Office for National Statistics, 1998
(Type IV evidence – qualitative and quantitative observational study)
1.1f. Most smokers (82%) start during their teenage yearsi. The peak age of taking up smoking is 14 to 16 years in boys and girls with a change in attitude and behaviour generally occurring at 13 years. After age 20 very few people take up smokingii. i. Thomas M, Walker A, Wilmot A, Bennett N, Office for National Statistics. Living in Britain: Results from the 1996 General Household Survey. London: The Stationery Office, 1998
(Type IV evidence – large survey)
ii. Townsend J. The Burden of Smoking. In: Benzeval M, Judge K, Whitehead M (editors). Tackling Inequalities in Health – An Agenda for Action. London: Kings Fund, 1995
(Type V evidence – expert opinion citing type IV evidence, survey)
1.1g. High smoking prevalence occurs in the most disadvantaged groups. Both men and women in lower socio-economic groups have responded less to past health publicity of the dangers of smoking than those in higher socio-economic groupsi. i. Townsend J, Roderick P, Cooper J. Cigarette smoking by socioeconomic group, sex, and age: effects of price, income and health publicity. British Medical Journal 1994; 309: 923-7.
http://www.bmj.com/cgi/full/content/full/309/6959/901
(Type V evidence – expert opinion summarising type IV evidence, observational studies)
1.1h. Very high smoking prevalence occurs in young married couples below 24 years with children, and especially in lone mothers. Multiple disadvantage makes matters worse and smoking prevalence of 70% has been reported in lone parents with low incomes, no educational qualifications, and living in council housingi. i. Townsend J. The Burden of Smoking. In: Benzeval M, Judge K, Whitehead M (editors). Tackling Inequalities in Health – An Agenda for Action. London: Kings Fund, 1995
(Type V evidence – expert opinion citing type IV evidence, surveys)

Top

1.1i. Young people whose parents smoke are twice as likely to smoke as children of non-smoking parentsi,ii. Those who perceive no parental disapproval are seven times more likely to smoke than young people who perceive strong parental disapproval. Young people with a sibling who smokes are up to four times more likely to smoke than those whose siblings do not smoke. In addition the prevalence of smoking among young people is higher in those living with a single parent and is higher still if the lone parent smokesiii,iv. i. Office for National Statistics. Young Teenagers and Smoking in 1997. London: Office for National Statistics, 1998
(Type IV evidence – statistics)
ii. Report of the Scientific Committee on Tobacco and Health. London: The Stationery Office, 1998
(Type V evidence – expert opinion based on type IV evidence citing: Why children start smoking. An enquiry carried by Social Survey Division of OPCS on behalf of the Department of Health. London: The Stationery Office, 1990)
iii. Pierce JP, Gilpin EA, Emery SL, White MM, Rosbrook B, Berry CC. Has the Californian tobacco control program reduced smoking? Journal of the American Medical Association 1998; 280: 893-9.
(Type IV evidence – summary of observational study)
iv. Royal College of Physicians. Smoking and the Young. A Report of The Working Party of the Royal College of Physicians. London: Royal College of Physicians, 1992
(Type V evidence – expert opinion based on type IV evidence, observational studies)
1.1j. The effect of peer smoking is important and more pronounced with increasing agei,ii. Whilst self-image, social representation and social identity are important factors in the uptake and continuation of smoking, the use of tobacco is affected by the wider environment more than by such individual factorsiii. Qualitative research suggests that health-damaging behaviours can enable people to cope under difficult circumstancesiv. i. Pierce JP, Gilpin EA, Emery SL, White MM, Rosbrook B, Berry CC. Has the Californian tobacco control program reduced smoking? Journal of the American Medical Association 1998; 280: 893-9.
(Type IV evidence – summary of observational study)
ii. Office for National Statistics. Young Teenagers and Smoking in 1997. London: Office for National Statistics, 1998
(Type IV evidence – surveys and qualitative studies)
iii. World Health Organisation Europe. Social Determinants of Health: The Solid Facts. Copenhagen: WHO, 1998
(Type V evidence – expert opinion citing type IV evidence, cohort, cross-sectional and qualitative studies)
iv. Benzeval M, Webb S. Family Poverty and Poor Health. In: Benzeval M, Judge K, Whitehead M (editors). Tackling Inequalities in Health – An Agenda for Action. London: Kings Fund, 1995
(Type V evidence – expert opinion citing qualitative type IV evidence)

Top

1.2 Smoking prevention or cessation interventions at a societal level

1.2a. The real price of cigarettes influences consumption among existing adult smokersi,ii. Excise tax is a large component of cigarette price. For the population as a whole, cigarette consumption decreases by about 0.5% for a 1% increase in price, adjusted for inflation.

Several studies have demonstrated that the uptake of smoking by young people shows the highest price elasticity (‘sensitivity’ or ‘responsiveness’ to price)iii,iv, which is important in the long term given that most smokers start as young people. In New Zealand, a 17% increase in cigarette price as a result of taxation in July 1991 appeared to accelerate the smoking reduction effect of an advertising ban in December 1990v.
(Health gain notation – 3 "trade off between beneficial and adverse effects")
Caveat: It is unlikely that cigarette price rise will increase smoking cessation in the most deprived households in Wales, and the price rise may make cessation more difficult and be harmful to people in these groups iii.

Real cigarette price has an unequal effect on cigarette consumption among socioeconomic groups. In the most disadvantaged groups smoking rates have remained stable against a background of real price rises. Furthermore, real cigarette price has been shown to affect the living standards in poor households, where a higher proportion of disposable income is spent on tobacco. Low income households on income support where parents smoke lack more basic amenities than similar households where parents do not smoke. One study suggests that living in hardship is the main deterrent to quitting smokingiii.

i. Godfrey C, Maynard A. Economic aspects of tobacco use and taxation policy. British Medical Journal 1988; 297: 339-43.
(Type V evidence – expert opinion based on type IV evidence)
ii. Townsend J, Roderick P, Cooper J. Cigarette Smoking by socioeconomic group, sex, and age: effects of price, income, and health publicity . British Medical Journal 1994; 309: 923-7
http://www.bmj.com/cgi/full/content/full/309/6959/901
(Type IV evidence – ecological epidemiological evidence and economic data)
iii. Acheson D (Chair). Independent Inquiry into Inequalities in Health. London: The Stationery Office, 1998
(Type V evidence – expert opinion based on a systematic review of type IV and econometric evidence)
iv. Report of the Scientific Committee on Tobacco and Health. London: The Stationery Office, 1998
(Type V evidence – expert opinion based on type IV evidence citing Diamond A, Goddard E. Smoking among secondary school children in 1994. OPCS Social Survey Division. London: The Stationery Office, 1995)
v. Report of the Scientific Committee on Tobacco and Health. London: The Stationery Office, 1998
(Type V evidence – expert opinion based on type IV evidence citing New Zealand Ministry of Health. Tobacco Statistics 1996. Cancer Society of New Zealand)

 

 

 

 

Top

1.2b. There is evidence that tobacco advertising and promotion influences tobacco consumptioni. Tobacco advertising and promotion helps to recruit young smokersii. The cigarette brands smoked most by children are the most heavily advertisediii,iv. Boys whose favourite sport on television was motor racing had a 12.8% risk of becoming regular smokers compared to 7.0% for boys who did not follow itv. Past tobacco advertising campaigns targeted at young women were associated with a large rise in uptake of smoking by women younger than legal age for cigarette purchasevi.
(Health gain notation – 6 "likely to be harmful")

 

i. Godfrey C, Maynard A. Economic aspects of tobacco use and taxation policy. British Medical Journal 1988; 297: 339-43.
(Type V evidence – expert opinion based on type IV evidence)
ii. Department of Health, Department of Health and Social Security Northern Ireland, Scottish Office Department of Health, Welsh Office. Report of the Scientific Committee on Tobacco and Health. London: The Stationery Office, 1998 p.35
(Type V evidence - expert opinion based on some type IV qualitative, IV quantitative and other type V evidence)
iii. Foulds J, Godfrey C. Counting the cost of children’s smoking. British Medical Journal 1995; 311: 1152-4.
http://www.bmj.com/cgi/content/full/311/7013/1152
In: Report of the Scientific Committee in Tobacco and Health 1998.
(Type V evidence – expert opinion based on type IV evidence)
iv. Smoking Kills - a White Paper on Tobacco. Cm 4177. London: The Stationery Office, 1999
http://www.official-documents.co.uk/document/
cm41/4177/4177.htm [accessed 10.1.00]

(Type V evidence – expert opinion based on type IV evidence citing Barron J. Young teenagers and smoking in 1997. London: Office for National Statistics, 1998)
v. Report of the Scientific Committee on Tobacco and Health. London: The Stationery Office, 1998 p.39
(Type V evidence - expert opinion)
vi. Report of the Scientific Committee on Tobacco and Health. London: The Stationery Office, 1998 (Type V evidence – expert opinion based on type IV evidence citing Pierce JP, Lee L, Gilpin EA. Smoking initiation by adolescent girls, 1944 through 1988: an association with targeted advertising. Journal of the American Medical Association 1994; 271: 608-11)
1.2c. Long-term ecological evidence from many different countries consistently shows that tobacco advertising bans are associated with reduced tobacco consumption. Bans in Norway (in 1975) and Finland were associated with a 9% and 7% reduction respectively, although when bans were introduced in Canada, Australia and New Zealand the reduction was lessi. In particular, the Norwegian ban led to a substantial reduction in smoking in school students and adult menii. In New Zealand heavy advertising before the ban in December 1990, was closely followed by a peak in smoking among 15 to 19 year olds in 1991 despite several preceding years of sharp decline. The peak declined by the second half of 1991. It is thought that the effect of an advertising ban interacts with concurrent measures on legislation, taxation, health promotion programmes and publicity campaignsiii.
(Health gain notation – 1 "beneficial")
Report of the Scientific Committee on Tobacco and Health. London: The Stationery Office, 1998
(Type V evidence – expert opinion based on type IV evidence) citing the following.
  1. Department of Health. Effect of Tobacco Advertising on Tobacco Consumption: A Discussion Document Reviewing the Evidence. London: Department of Health, 1992
  2. Kersler DA, Barnett PS, Witt A, Zeller MR, Mande JR, Schultz WB. The legal and scientific basis for the FDA’s assertion of jurisdiction over cigarette and smokeless tobacco. Journal of the American Medical Association 1997; 277: 405-9
  3. New Zealand Ministry of Health. Tobacco Statistics 1996. Cancer Society of New Zealand

Top

1.2d. A Cochrane review is in progress on community action for reducing smoking among adultsi. i. Secker-Walker R. Community action for reducing smoking among adults (Protocol). Cochrane Database of Systematic Reviews. Cochrane Library 1999 Issue 4
(Type I evidence – systematic review in progress)
1.2e. Mass media campaigns can be effective in preventing the uptake of smoking in young peoplei. Campaigns were more effective when they had a theoretical basis, used formative research in designing messages, and were of sufficient intensity over an extended period.
(Health gain notation – 2 "likely to be beneficial")
Caveats: Studies were in different countries. Media campaigns tend to be culturally sensitive so the response in Wales may be different.
i. Sowden AJ, Arblaster L. Mass media interventions for preventing smoking among young people. Cochrane Library 1998 Issue 4
http://www.update-software.com/ccweb/cochrane/
revabstr/ab001291.htm
[accessed 3.3.00]
(Type I evidence – systematic review of 6 controlled trials of children and young adults under 25. Some randomised trials. At least 57 uncontrolled trials available but not reviewed)
1.2f. A widely advertised telephone helpline, with support material available through it, could be effective in promoting smoking cessation in smokers in the general populationi. In Scotland a before-and-after comparison of such a helpline suggested that 6% of the smoking adult population contacted the helpline (which was advertised on television and posters). In a 10% random sample followed up over a year, 23.6% claimed they were non-smokers, 19.6% were smoking less, and overall 42.4% had stopped at some time during the year. The population rate of decline in smoking was greater during and after the campaign than preceding it (2% compared to 0.8%). Confounding variables such as cigarette advertising, tobacco price and other smoking cessation adverts were considered. Unknown confounders could not be accounted for in this type of study.
(Health gain notation – 2 "likely to be beneficial")
i. Platt S, Tannahill A, Watson J, Fraser E. Effectiveness of antismoking telephone helpline: follow up survey. British Medical Journal 1997; 314: 1371-1375 http://www.bmj.com/cgi/content/full/314/7091/1371 [accessed 3.3.00]
(Type III evidence – follow-up study of a cohort of 848 adult smokers)

 

 

 

 

Top

1.2g. No Smoking Day is effective in achieving smoking cessation in the population. In Wales in 1996 a high level of awareness of the day was achieved and participation was high. Three month quit rates are estimated at 0.3 to 1.8%, which translates to a large number of smokers quitting at the population leveli. Smokers who had firm plans to quit are more likely to be motivated by the dayi.
(Health gain notation – 2 "likely to be beneficial")
i. Frith C, Roberts C, Kingdon A, Tudor-Smith C. An evaluation of the 1996 No Smoking Day in Wales. Health Education Journal 1997; 56: 287-95
(Type III evidence – before and after study)
1.3 Environmental tobacco smoke
1.3a. Cochrane reviews are under way to examine the effect of smoking prevention and control programmes for families and carers to reduce children’s exposure to environmental tobacco smokei and interventions for preventing tobacco use in public placesii.
See also Section 1.7 on ‘Smoking prevention and cessation in the workplace’.
i. Waters E, Campbell R, Webster P, Spencer N. Smoking prevention and control programmes for families and their carers to reduce children’s exposure to tobacco smoke (Protocol). Cochrane Database of Systematic Reviews. Cochrane Library 1999 Issue 4
(Type I evidence – systematic review in progress)
ii. Serra C, Cabezas C, Bonfil X, Pladevall-Vila M. Interventions for preventing tobacco use in public places (Protocol). Cochrane Database of Systematic Reviews. Cochrane Library 1999 Issue 4
(Type I evidence – systematic review in progress)
1.4 Prevention of smoking in children and young people
1.4a. Community action for preventing smoking in young people can be effective when multi-faceted approaches are used, targeting multiple sites such as schools, work places and churches. Multi-faceted approaches include age restriction on tobacco purchase, smoke-free public places, media campaigns and special programmes in schoolsi.
(Health gain notation – 2 "likely to be beneficial")
i. NHS Centre for Reviews and Dissemination. Preventing the Uptake of Smoking in Young People. Effective Health Care. 1999; 5(5)
http://www.york.ac.uk/inst/crd/ehc55.htm
[accessed 10.3.00]
(Type I evidence – systematic review citing Sowden A, Arblaster L. Community interventions for preventing smoking in young people. Cochrane Database of Systematic Reviews. Cochrane Library 1999 Issue 4)
1.4b. Interventions with retailers lead to decreased illegal sales of cigarettes to ‘under age’ children and young peoplei.
(Health gain notation – 2 "likely to be beneficial")
Active enforcement and mobilising community support are more effective than retailer educational initiatives alone. The infringement penalty to the retailer may be important – a graduated system of penalties may prevent a backlash against tobacco control measures. Warning letters threatening prosecution are effective. Voluntary compliance programmes do not appear to be effective.

Vending machines – a locking device policy was effective in reducing the number of places selling cigarettes to those underage.
(Health gain notation – 2 "likely to be beneficial")
An enforced ban on vending machines may be more effective. Most studies that measured it failed to demonstrate a significant difference in perceived ease of access to cigarettes by children, despite reduced illegal sales. However no studies measured actual access (other than purchases).
(Health gain notation – 4 "unknown")

Although difficult to assess in the community, some effect of reducing smoking, particularly amongst 12 year olds, was noted in two of the better quality studies. A study by Forster et al (1996) cited in the review, appeared to influence smoking through reduced sales from intervention retailers posting warning notices and storing cigarettes behind counters.

In summary, comprehensive measures which include sustained law enforcement and community mobilisation, with the banning of vending machines, along with a graduated system of retailer penalties is likely to lead to reduced cigarette sales. These measures would probably lead to reduced smoking in younger age groups. There is a lack of evidence that actual non-purchase access to cigarettes increases significantly as a results of these interventionsi.
(Health gain notation – 2 "likely to be beneficial")

i. Lancaster T, Stead LF. Interventions for preventing tobacco sales to minors. Cochrane Library 1999 Issue 4 http://www.update-software.com/ccweb/cochrane/
revabstr/ab001497.htm
[accessed 3.3.00]
(Type I evidence - systematic review of 27studies, 13 of which were controlled)

 

 

Top

1.4c. A Cochrane review is being conducted on school based programmes for preventing smokingi.

To date it appears that programmes based mainly in schools are limited in their effectiveness. Social reinforcement/social norms type programmes seem to be more effective than solely knowledge based programmes. Problems that have been identified include the training of teachers and programmes targeted at children after they have started smokingii. In an intention to treat analysis, a recent large cluster randomised trial of 3 computer sessions and 3 classroom sessions (based on a trans-theoretical ‘stages of change’ model) over a year showed no difference between intervention and control schools in the prevalence of smoking in 13-14 year oldsiii. A large randomised trial of peer-led schools programmes has yet to be conducted. However, a small evaluation study in the former Mid Glamorgan showed promising results, although bias or chance could have explained themiv.
(Health gain notation – 4 "unknown")

In a combined quantitative and qualitative study the most important factor that influenced smoking prevalence among school students at a given school was the culture of the school, and this was over and above any socioeconomic factorsv. The European Network of Health Promoting Schools may also be important in helping to prevent smoking in pupils and staff by encouraging healthy lifestyles through policy implementation and supportive environmentsii.
(Health gain notation – 2 "likely to be beneficial")

i. Thomas R, Busby K.School based programmes for preventing smoking (Protocol). Cochrane Database of Systematic Reviews. Cochrane Library 1999 Issue 4
(Type I evidence – systematic review in progress)
ii. NHS Centre for Reviews and Dissemination. Preventing the Uptake of Smoking in Young People. Effective Health Care. 1999; 5(5) http://www.york.ac.uk/inst/crd/ehc55.htm[accessed 3.3.00]
(Type I evidence – overview of seven systematic reviews)
iii. Aveyard P, Cheng KK, Almond J, et al. Cluster randomised controlled trial of expert system based on the transtheoretical (‘stages of change’) model for smoking prevention and cessation in schools. British Medical Journal 1999; 319: 948-953
http://www.bmj.com/cgi/content/full/319/7215/948
[accessed 3.3.00]
Type II evidence – large cluster randomised controlled trial with intention to treat analysis)

iv. Bloor M, Frankland J, Parry Langdon N, et al. A controlled evaluation of an intensive, peer-led, schools based, anti smoking programme. Health Education Journal 1999; 58: 17-25
(Type III evidence – small non-randomised study with control group)
v. Report of the Scientific Committee on Tobacco and Health. London: The Stationery Office, 1998
(Type V evidence – expert opinion based on type IV evidence citing Lloyd B, Lucas K. Why do young girls smoke? A quantitative/behavioural study. A report for the Department of Health 1996 [unpublished])

 

 

Top

1.5 Smoking cessation in children and young adults
1.5a. There is a lack of published research on effective interventions for smoking cessation in young people. Effective interventions for adults may or may not be effective in children and young adults, and there may be additional problems in making any effective intervention available to this group.
There is a need for evaluation of the effectiveness of smoking cessation programmes amongst children and young adults.
 
1.6 Smoking cessation in pregnancy
1.6a. Simple and more complex advice to pregnant women to quit smoking is effective in increasing smoking cessation (all intervention over no interventions, for continued smoking in late pregnancy, odds ratio 0.51 [95% CI: 0.45, 0.58], absolute difference 6.6%). Interventions did not include advice to cut down. For women who had already stopped smoking at first antenatal visit, there was a trend towards a greater reduction in relapse rate in those advised not to smoke. Additional group sessions for smoking cessation in pregnancy are poorly accepted by womeni.
(Health gain notation –1"beneficial")
Caveat: The results of these studies did not give differential results for these interventions for women according to socioeconomic group.

The percentage of pregnant women who recalled receiving anti-smoking advice from a health professional was 49% in the prospective Health Education Authority studyii but 85% in the retrospective Infant Feeding studyiii.
(Health gain notation –2"likely to be beneficial")
Caveat: One third of women receiving advice from a GP, and almost one half receiving advice from a midwife, recall being advised to cut down cigarette consumption rather than to quitii.

i. Lumley J, Oliver S, Waters E. Interventions for promoting smoking cessation during pregnancy. Cochrane Library 1999 Issue 4
http://www.update-software.com/ccweb/cochrane/
revabstr/ab001055.htm
[accessed 3.3.00]
(Type I evidence – systematic review of randomised controlled trials. Main outcome derived from pooled data of 30 trials. Subjects were healthy pregnant women, mainly in a hospital or community antenatal clinic setting)

ii. Report of the Scientific Committee on Tobacco and Health. London: The Stationery Office, 1998
(Type V evidence – expert opinion based on type IV evidence citing: Health Education Authority. Trends in smoking and pregnancy 1992-1997Report of the Scientific Committee on Tobacco and Health. London: The Stationery Office, 1998)
iii. Report of the Scientific Committee on Tobacco and Health. London: The Stationery Office, 1998
(Type V evidence – expert opinion based on type IV evidence citing: Social Services Division of the Office for National Statistics. Infant Feeding 1995. London: The Stationary Office, 1997)

 

Top

1.7 Smoking prevention and cessation in the workplace
1.7a. Tobacco policies at work were found to reduce cigarette consumption at work and worksite environmental tobacco smoke exposurei.
(Health gain notation – 2 "likely to be beneficial")
Smoking cessation group programmes were found to be more effective than minimal treatment programmesi.
(Health gain notation – 2 "likely to be beneficial")
Caveat: It is difficult to summarise the result of this review. It is probably systematic although this is not clear. Most studies were in US work settings. A numerical meta-analysis was not possible. The statements are based on consistent findings from 52 original studies.

A meta-analysis of worksite smoking cessation programmesii showed that quit rates at one year of 12 to 18% could be achieved, the weighted average quit rate being 13%. However the analysis was not on an intention to treat basis. Programmes with the following independent predictor variables had larger quit rates: had a cessation group component; heavier smokers; with more intensive intervention; that were not complicated; in smaller companies; and shared company and employee time.
(Health gain notation – 2 "likely to be beneficial")

i. Eriksen MP, Gottileb NH. A review of the health impact of smoking control in the workplace. American Journal of Health Promotion 1998; 13: 83-104
(Type IV evidence - a narrative systematic review of only reasonable quality including randomised, other interventional and observational studies)
ii. Fisher KJ, Glasgow RE, Tervorg JR. Worksite smoking cessation: A meta-analysis of long term quit rates from controlled studies. Journal of Occupational Medicine 1990; 32(5): 429-439
(Type III evidence - a systematic review and meta-analysis of controlled studies, not necessarily randomised)
1.8 Smoking cessation in adults
1.8a. Guidelines can be effective in changing practitioner behaviour if local circumstances are taken into consideration, they are disseminated by educational interventions and are used with patient-specific reminders. They are more likely to have an effect if used as part of more complex measures to improve and change servicesi. Evidence based guidelines exist for smoking cessation in adultsii. i. Getting evidence into practice. Effective Health Care 1999; 5(1)http://www.york.ac.uk/inst/crd/ehc51.htm [accessed 6.4.00]
(Type I evidence – overview of 44 systematic reviews)
ii. British Thoracic Society. Smoking Cessation Guidelines and their Cost Effectiveness. Thorax 1998; 53 (supplement): S1-S18
(Guidelines based on type I, II, III, VI and V evidence)

Top

1.8b. Group behavioural therapy is more effective than self-help programmes (odds ratio 2.10 [95% CI: 1.64, 2.70]), and more effective than no intervention or minimal contact (odds ratio 1.91 [95% CI: 1.20, 3.04]). It is probably of similar effectiveness to individual counselling of comparable intensityi.
(Health gain notation – 1 "beneficial")
i. Stead LF, Lancaster T. Group behaviour therapy programmes for smoking cessation. Cochrane Library 1998 Issue 4
http://www.update-software.com/ccweb/cochrane/
revabstr/ab001007.htm
[accessed 3.3.00]
(Type I evidence – meta-analysis of systematic review of randomised controlled trials. 19 studies compared a group programme with another method or control. Most recruited community volunteers; 2 from primary care settings: one in people with cardiovascular disease and one in those with diabetes)
1.8c. Brief simple advice, and more complex advice to quit smoking (not to cut down) is effective when delivered by any type of health care professional, compared to no intervention (for psychologist/ social worker/ counsellor: odds ratio 1.8 [95% CI 1.5, 2.2]; for physician odds ratio: 1.5 [95% CI 1.2, 1.9]; for dentist/ nurse/ pharmacist: odds ratio 1.4 [95% CI 1.1, 1.8])i. There is no clear advantage to any professional type but very high cessation rates occur when many types of professionals reinforce the advice to each client (odds ratio compared to no intervention 3.8 [95% CI 2.6, 5.6])i.
(Health gain notation – 1 "beneficial")
i. Fiore MC, Bailey WC, Cohen SJ, et al. Smoking Cessation. Clinical Practice Guideline No 18. Rockville, MD: US Department of Health and Human Services, Public Health Service, Agency for Health Care Policy and Research. AHCPR Publication No. 96-0692. April 1996
(Type I evidence – meta-analysis of systematic review of randomised controlled trials)

Now superceded by the Tobacco Cessation Guideline on
http://www.surgeongeneral.gov/tobacco/default.htm (accessed  2.8.00).
1.8d. Individual behavioural counselling from a healthcare worker is effective in smoking cessation. (Odds ratio for smoking cessation 1.55 [95% CI: 1.27, 1.90] ). Intensive counselling is not superior to brief counselling (Odds ratio 1.17 [95% CI 0.59, 2.34]i.
(Health gain notation – 1 "beneficial")
Caveats: Counselling was defined by a minimum time spent with the smoker by a counsellor trained in smoking cessation. This did not include counselling by doctors and nurses as part of clinical care, nor any specific behavioural technique.
i. Lancaster T, Stead LF. Individual behavioural counselling for smoking cessation. Cochrane Library 1999 Issue 2.
http://www.update-software.com/ccweb/cochrane/
revabstr/ab001292.htm
[accessed 3.3.00]
(Type I evidence – meta-analysis of systematic review of controlled trials. 11 trials identified, all except one randomised. 7 studies in hospitalised patients. Level of motivation to quit difficult to assess. Intervention in control groups varied.)

 

Top

1.8e. Systematically identifying and documenting smoking status results in much higher rates of smoking cessation interventions by clinicians (odds ratio 3.1 [95% CI 2.2, 4.2]) and a trend to higher rates of smoking cessation (odds ratio 2.0 [95% CI 0.8, 4.8])i.
(Health gain notation – 1 "beneficial")
Caveat: Based on small numbers in only three studies. Most of the studies did not set out to measure smoking cessation as the outcome.
i. Fiore MC, Bailey WC, Cohen SJ, et al. Smoking Cessation. Clinical Practice Guideline No 18. Rockville, MD: US Department of Health and Human Services, Public Health Service, Agency for Health Care Policy and Research. AHCPR Publication No. 96-0692. April 1996
(Type I evidence – meta-analysis of systematic review of randomised controlled trials)
1.8f. Training healthcare professionals in smoking cessation techniques is effective in increasing smoking cessation in their patientsi. (Odds of stopping smoking if attended trained professional versus control professional: odds ratio 1.48 [95% CI 1.20, 1.83]). The use of prompts and reminders in practice, as well as training increases the odds of quitting i. (odds ratio 2.37 [95% CI 1.43, 3.92])
(Health gain notation – 1 "beneficial")
Caveat: All the trials were in North America.
i. Lancaster T, Silagy C, Fowler G, Spiers I. Training health professionals in smoking cessation. Cochrane Library 1998 Issue 4
http://www.update-software.com/ccweb/cochrane/
revabstr/ab000214.htm
[accessed 3.3.00]
(Type I evidence – meta-analysis of systematic review of randomised controlled trials. 9 trials identified)
1.8g. Simple brief advice by a general practitioner to quit smoking is effectivei,ii (GP advice versus no advice odds ratio 1.32 [95% CI 1.18,1.48], NNT 35)ii. (Odds ratio for abstinence at least at 6 months, for brief advice over no advice 1.73 [95% CI: 1.47, 2.02])i, (odds ratio 1.5 [95% CI 1.2, 1.9]ii. Advice was defined as taking 3 minutes or less.
(Health gain notation – 1 "beneficial")
i. Ashenden R, Silagy C, Weller D. A systematic review of the effectiveness of promoting lifestyle changes in general practice. Family Practice 1997; 14(2): 160-175. In: University of York NHS CRD DARE Search Document 4
(Type I evidence – systematic review)
ii. Fiore MC, Bailey WC, Cohen SJ, et al. Smoking Cessation. Clinical Practice Guideline No 18. Rockville, MD: US Department of Health and Human Services, Public Health Service, Agency for Health Care Policy and Research. AHCPR Publication No. 96-0692. April 1996 http://www.ahcpr.gov/meta.htm [accessed 3.3.00]
(Type I evidence – meta-analysis of systematic review of randomised controlled trials)
1.8h. Simple brief advice to quit (not cut down) smoking from nurses and health visitors is effective in increasing smoking cessation compared to no intervention or usual care (odds ratio 1.43 [95% CI: 1.24 to 1.66])i. There is no evidence that more intensive interventions are more effective than less intensive ones. Advice to quit is effective for both hospitalised and non-hospitalised patients.
(Health gain notation – 1 "beneficial")
i. Rice VH, Stead LF. Nursing interventions for smoking cessation. Cochrane Library 1999 Issue 3.
http://www.update-software.com/ccweb/cochrane/
revabstr/ab001188.htm
[accessed 3.3.00]
(Type I evidence – meta-analysis of a systematic review)

 

Top

1.8i. Self-help materials have a small effect on smoking cessation, compared to no intervention (Odds ratio 1.23 [95% CI 1.01, 1.51])i. Personalised, tailored self-help material is more effective (odds ratio 1.51 [95% CI 1.13, 2.02]). Telephone follow up in addition to self-help material is more effective than self-help alone (odds ratio 1.62 [95% CI 1.33, 1.97]). The self-help material had to provide structured behavioural strategies (not just information) to assist an unaided attempt to give up smoking. There is no additional benefit over brief therapist contacti.
(Health gain notation – 1 "beneficial")
i. Lancaster T, Stead LF. Self-help interventions for smoking cessation. Cochrane Library 1998 Issue 4
http://update-software.com/ccweb/cochrane/
revabstr/ab0118.htm
[accessed 3.3.00]
(Type I evidence – meta-analysis of systematic review of mostly randomised controlled trials. 41 trials identified. Uncertain of settings and groups actually discovered by review although studies of any smokers other than in pregnancy searched for)
1.8j. Proactive telephone counselling of known smokers for smoking cessation is likely to have a small positive effect (odds ratio at 12 to 18 months follow up 1.20 [95% CI: 1.06, 1.37])i.
(Health gain notation – 1 "beneficial")
i. Liechtenstein E, Glasgow RE, Lando HA, Ossip-Klein DJ. Telephone counseling for smoking cessation: rationales and meta-analytic review of the evidence. Health Education Research 1996; 11: 243-57
(Type I evidence – meta-analysis of systematic review of 13 randomised controlled trials, nine included in the final pooled odds ratio after testing for homogeneity)
1.8k. In smokers motivated to quit, a ‘buddy system’ is very effective in the short term in a general practice nurse-led smoking cessation clinic. Smoking cessation at 4 weeks after quit date is doubled in smokers paired with each other for mutual support between clinic sessions (‘buddy’ system), compared to smokers attending alone (‘solo’ system). (Odds ratio 2.6, p<0.05, absolute risk difference 15%, NNT 6.7)i. All smokers were seen by the trained practice nurse and given brief simple structured advice.
(Health gain notation – 1 "beneficial")
Caveat: Inner urban general practice population in South East London. Need for caution about generalisability. Smokers in both groups were offered nicotine replacement therapy unless it was inappropriate.
i. West R, Edwards M, Hajek P. A randomized controlled study of a ‘buddy’ system to improve success at giving up smoking in general practice. Addiction 1998; 93(7): 1007-1011
(Type II evidence – un-blinded randomised controlled trial of 172 smokers in general practice setting)

 

 

 

Top

1.8l. A Cochrane review is underway to examine the effect of interventions for smoking cessation in hospitalised patientsi. i. Rigotti NA, Munafo M. Interventions for smoking cessation in hospitalised patients (Protocol) Cochrane Database of Systematic Review. Cochrane Library 1999 Issue 4
(Type I evidence – systematic review in progress)
1.9 Smoking cessation in adults - Pharmacological products
1.9a. Nicotine replacement therapy (NRT) is an effective component of cessation strategies in heavier smokers who are motivated to quit (Odds ratio for abstinence with NRT over control 1.73 [95% CI: 1.60, 1.86])i. 8 weeks of patch therapy is as effective as longer courses. There is no evidence that tapered therapy is better than abrupt withdrawal. Wearing the patch during waking hours is as effective as wearing it for 24 hours/day.
(Health gain notation – 1 "beneficial")
There is evidence for the ineffectiveness of nicotine replacement therapy for individuals smoking less than 10-15 cigarettes per dayi.
(Health gain notation – 5 "unlikely to be effective")

Further research is needed to compare different types of nicotine replacement therapy..

i. Silagy C, Mant D, Fowler G, Lancaster T. Nicotine replacement therapy for smoking cessation. Cochrane Library 1998 Issue 4
http://www.update-software.com/ccweb/cochrane/
revabstr/ab000146.htm
[accessed 3.3.00]
(Type I evidence – meta-analyses within systematic review of randomised controlled trials, including 25,600 smokers in total)
1.9b. Clonidine appears to have some beneficial effect in smoking cessation (Pooled odds ratio for abstinence at 12 weeks of clonidine over placebo 1.87 [95% CI: 1.27, 2.77]i.
(Health gain notation – 3 "trade off between beneficial and adverse effects")
Caveats: Because of concerns about the methods of some of the primary studies and of the frequency of adverse effects, it cannot be recommended as a first line treatment. It appears to be useful in
  • Extreme agitation during withdrawal
  • Multiple drug withdrawal
i. Gourlay SG, Stead LF, Benowitz NL. Clonidine for smoking cessation. Cochrane Library 1998 Issue 4
http://www.update-software.com/ccweb/cochrane/
revabstr/ab000058.htm
[accessed 3.3.00]
(Type I evidence – which included meta analysis of 5 randomised controlled trials, assessing smoking cessation at least 12 weeks after end of treatment. Total of 722 participants. Uncertain about the effects of bias)

 

Top

1.9c. There is insufficient evidence to recommend antidepressants as first line treatment in smoking cessation in preference to nicotine replacement therapyi.
(Health gain notation – 4 "unknown")
i. Hughes JR, Stead LF, Lancaster TR. Anxiolytics and antidepressants in smoking cessation. Cochrane Library 1998 Issue 4
http://www.update-software.com/ccweb/cochrane/
revabstr/ab000031.htm
[accessed 3.3.00]
(Type I evidence – systematic review)
1.9d. There is insufficient evidence to recommend the use of mecamylamine in smoking cessationi.
(Health gain notation – 4 "unknown")
i. Lancaster T, Stead L. Mecamylamine (a nicotine antagonist) for smoking cessation. Cochrane Library 1998 Issue 4
http://www.update-software.com/ccweb/cochrane/
revabstr/ab001009.htm
[accessed 3.3.00]
(Type I evidence – systematic review but the two small studies identified does not allow confident assessment of effect size)
1.9e. There is lack of evidence for the effectiveness of lobeline resulting in long term abstinence in smokingi.
(Health gain notation – 4 "unknown")
Caveat: There is evidence for the ineffectiveness of lobeline resulting in short term abstinence in smoking.
i. Stead LF, Hughes JR. Lobeline for smoking cessation. Cochrane Library 1998 Issue 4
http://www.update-software.com/ccweb/cochrane/
revabstr/ab000124.htm
[accessed 3.3.00]
(Type III evidence – systematic review but no randomised controlled trials of long term follow up available)
1.9f. Existing evidence suggests that silver acetate is ineffective in smoking cessation (Pooled odds ratio for quitting smoking for silver nitrate over placebo 1.05 [95% CI: 0.63, 1.73]i.
(Health gain notation – 6 "likely to be ineffective or harmful")

 

i. Lancaster T, Stead L. Silver acetate for smoking cessation. Cochrane Library 1998 Issue 4
http://www.update-software.com/ccweb/cochrane/
revabstr/ab000191.htm
[accessed 3.3.00]
(Type I evidence – systematic review including two randomised controlled trials, showing outcome as sustained smoking abstinence at 6-12 months)
1.9g. There is evidence that anxiolytics are ineffective in smoking cessationi.
(Health gain notation – 6 "likely to be ineffective or harmful")

 

i. Hughes JR, Stead LF, Lancaster TR. Anxiolytics and antidepressants in smoking cessation. Cochrane Library 1998 Issue 4
http://www.update-software.com/ccweb/cochrane/
revabstr/ab000031htm
[accessed 3.3.00]
(Type I evidence – systematic review)
1.10 Smoking cessation in adults - Alternative and complementary therapies
1.10a. Acupuncture is ineffective in achieving sustained smoking cessation (odds ratio at 12 months 1.02 [95% CI 0.72, 1.43])i. No one method of acupuncture is superior to control
(Health gain notation – 6 "likely to be ineffective")

 

i. White AR, Rampas H, Ernst E. Acupuncture for smoking cessation. Cochrane Library 1998 Issue 4
http://www.update-software.com/ccweb/cochrane/
revabstr/ab000009.htm
[accessed 3.3.00]
(Type I evidence – systematic review and meta-analysis of 14 trials)
1.10b. There is insufficient evidence at present to support the use of hypnotherapy in smoking cessationi.
(Health gain notation- 4 "unknown")
i. Abbot NC, Stead LF, White AR, Barnes J, Ernst E. Hypnotherapy for smoking cessation. Cochrane Library 1998 Issue 4
http://www.update-software.com/ccweb/cochrane/
revabstr/ab001008.htm
[accessed 3.3.00]
(Type I evidence – systematic review of 9 randomised controlled trials. Insufficient data to perform meta-analysis owing to small studies, heterogeneity of interventions and controls)
1.10c. There is insufficient evidence for supporting the use of aversive therapy methods in smoking cessationi.
(Health gain notation- 4 "unknown")
Caveat: Although the pooled odds ratio for repulsion through rapid smoking vs control is 2.08 [95% CI 1.39, 3.12], a funnel plot showed an absence of small studies with negative results. Other aversion methods did not appear to be effective. Most studies showed serious methodological problems likely to lead to spurious positive results.
i. Hajek P, Stead LF. Aversive smoking for smoking cessation. Cochrane Library 1998 Issue 4
http://www.update-software.com/ccweb/cochrane/
revabstr/ab000546.htm
[accessed 3.3.00]
(Type I evidence – systematic review of 24 randomised controlled trials. Insufficient data with methodoligical problems and probable publication bias)

Top


Contents Home

Health Evidence Bulletins: Wales, Duthie Library, UWCM, Cardiff CF14 4XN. e-mail: weightmanal@cardiff.ac.uk