Content area
Full Text
Peto estimates that current cigarette smoking will cause about 450 million deaths worldwide in the next 50 years. Reducing current smoking by 50% would avoid 20â[euro]"30 million premature deaths in the first quarter of the century and about 150 million in the second quarter. 1 Preventing young people from starting smoking would cut the number of deaths related to tobacco, but not until after 2050. Quitting by current smokers is therefore the only way in which tobacco related mortality can be reduced in the medium term. There is evidence that some form of treatment aids an increasing number of successful attempts to quit. 2 This review aims to summarise evidence for the effectiveness of the available interventions.
Summary points
Advice from doctors, structured interventions from nurses, and individual and group counselling are effective interventions
Generic self help materials are no better than brief advice but more effective than doing nothing; personalised materials are more effective than standard materials
All forms of nicotine replacement therapy are effective
The antidepressants bupropion and nortriptyline increased quit rates in a small number of trials; the usefulness of the antihypertensive drug clonidine is limited by side effects
Anxiolytics and lobeline are ineffective
The effectiveness of aversion therapy, mecamylamine, acupuncture, hypnotherapy, and exercise is uncertain
Methods
The Cochrane Tobacco Addiction Review group identifies and summarises the evidence for interventions to reduce and prevent tobacco use; it produces and maintains systematic reviews to inform policymakers, clinicians, and individuals wishing to stop smoking. Twenty systematic reviews are available in the Cochrane Library and have contributed to the evidence base for smoking cessation guidelines. 3
Details of the methods and results of each review are available in the Cochrane Library (abstracts at www.update-software.com/ccweb/cochrane/revabstr/g160index.htm ). The reviews summarise results from randomised controlled trials with at least six months' follow up. Sustained abstinence is the preferred outcome, but point prevalence rates are used when these are not available. Where possible, the reviews include estimates of treatment effect based on meta-analysis, expressed as Peto odds ratios 4 with 95% confidence intervals. An odds ratio greater than 1 indicates more quitters in the intervention group. The odds ratio assumes that the relative effects of treatment are constant despite the use of different outcome measures. The absolute quit...