Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Nicotine replacement therapy for stopping smoking

Cost effectiveness

Bandolier 54 examined a Cochrane review of nicotine replacement therapy. Since then the review has been updated, nicotine replacement therapy (NRT) is much higher on the political agenda, and the way we look at data has changed a bit, so we thought it worth another look, together with some cost-effectiveness.


The Cochrane review [1] is typically thorough because the Cochrane Tobacco Addiction Group has its own ongoing register of trials that is being constantly updated. This review was of trials available up to April 2000. Included were randomised trials in which NRT was compared to placebo or no treatment, or where different doses of NRT were compared. Excluded were trials not reporting cessation rates or with follow-up of less than six months.

The main outcome measure was abstinence from smoking after at least six months of follow-up. The most rigorous definition of abstinence for each trial was used, with biochemically validated rates if available. The meta-analysis used fixed effects odds ratios, but since those lack intuitive meaning, Bandolier reports NNTs together with absolute quit rates.


The main results are in Table 1, for all trials, for large trials (those with combined placebo plus NRT patients of at least 250), and for low baseline quit rates of less than 10%. These sensitivity analyses seemed additionally useful to test the validity of the results, because many of the NRT trials included in the review were small (fewer than 100 patients) and the range of quit rates without NRT was wide (2-46% with gum, for instance).

Table: Results from nicotine replacement therapy meta-analysis with sensitivity analysis

    Patients stopped smoking at 6-12 months  
    NRT Placebo  
Type of NRT Number of trials Number/total Percent Number/total Percent NNT (95% CI)
All trials            
Gum 48 1453/7387 20 1084/9319 12 12 (11 to 14)
Patch 31 1384/9708 14 495/5969 8 17 (14 to 20)
Intranasal spray 4 107/448 24 52/439 12 8 (6 to 14)
Inhaler 4 84/490 14 44/486 8 12 (8 to 26)
Sublingual tablet 2 49/243 20 31/245 13 13 (7 to 103)
Large trials            
Gum 18 792/5126 15 710/7308 10 17 (14 to 22)
Patch 14 1115/8333 13 352/4615 8 17 (15 to 21)
Cessation rate with control<10%        
Gum 15 299/3370 9 315/5192 6 36 (25 to 61)
Patch 17 482/4219 11 193/3440 6 17 (14 to 22)
Large trials were those with more than 250 participants in NRT and placebo groups combined

Numbers needed to treat for nicotine replacement versus placebo or no treatment controls were of the order of 8 to 17 for different NRT preparations using all trials. There were many trials and patients for gum and patch, but limited numbers for intranasal sprays, inhalers and sublingual tablets (Table 1). Between 8% and 13% of patients stopped smoking at 6-12 months without NRT, and this increased to 14% to 24% with NRT.

Only gum and patch had trials of over 250 participants. The NNT for patch was 17 for all trials and 17 for large trials. For gum, the NNT of 12 for all trials rose to 17 for large trials.

Only gum and patch had sufficient trials with controls cessation rates of less than 10% for analysis. The NNT for patch was 17 for all trials and 17 for trials with lower cessation rates. For gum, the NNT of 12 for all trials rose to 36 for control cessation rates below 10%.


The overall result for the updated Cochrane review is similar to that obtained previously. We can be sure that nicotine patches will almost double the number of people stopping smoking at six to 12 months. For every 17 people using nicotine patch for about eight weeks to help smoking cessation, one more will stop smoking who would not have done with no patch or with placebo patch.

The background rate of cessation of smokers in general is estimated at about 1.5% a year. In the control groups in these trials, the rate of cessation is often much higher than this, showing that people wanted to stop smoking, and that personal motivation with some professional intervention can achieve a certain amount on its own. Nicotine patches can almost double the rate, which must be good. The evidence for patches is robust to sensitivity analysis, and based on a large number of trials and patients.

The evidence for gum is a bit flakey, because the NNTs increase substantially with larger trials and in those with lower control cessation rates. Inhalers and sprays look effective but based on relatively small numbers of trials and patients.

Cost effectiveness

So if nicotine patches are prescribed in general practice, is this a good buy for health services? The argument would be that the known effects of smoking are so awful, that by stopping people smoking we buy further years of life. So we should be able to compute how much the intervention(s) cost, how many life years we obtain, and therefore the cost per life year.

This has been done [2] off the back of a trial in which 4.5% of people stopped smoking with GP counselling, and 9.6% stopped with counselling plus nicotine patches (NNT 20). In the trial only one of 476 smokers who were still smoking at the end of the first week of treatment were abstinent at one year. Calculations were therefore based on a model that only allowed nicotine patches for a week in those who were still smoking (based on breath carbon monoxide measurement).

The cost per life year saved was £344 to £785 depending on the age of the patient. Other studies [3] come up with similar estimates. When examined against a library of life-saving interventions [4], nicotine replacement therapy is pretty cost-effective.


  1. C Silagy et al. Nicotine replacement therapy for smoking cessation (Cochrane Review). In: The Cochrane Library, Issue 1, 2001. Oxford: Update Software.
  2. JA Stapleton et al. Prescription of transdermal nicotine patches for smoking cessation in general practice: evaluation of cost effectiveness. Lancet 1999 354: 210-215.
  3. S Parrots et al. Guidance for commissioners on the cost-effectiveness of smoking cessation interventions. Thorax 1999 53 (Suppl 5): S1-S38.
  4. TO Tengs et al. Five-hundred life-saving interventions and their cost-effectiveness. Risk Analysis 1995 15:369-390.
previous or next story in this issue