This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn.

The study examined the use of first-line pharmacotherapies to aid smoking cessation. The pharmacotherapies comprised four nicotine replacement therapies (NRTs) and an antidepressant (bupropion). The NRTs were nicotine gum, nicotine patch, nicotine spray and nicotine inhaler. Each pharmacotherapy was given as an adjunct to cessation counselling provided by a general practitioner (GP).

Treatment.

Cost-effectiveness analysis.

The study population consisted of patients who smoked, on average, 20 cigarettes per day.

The setting was primary care in one of six Western industrialised countries: Canada, France, Spain, Switzerland, the UK and the USA.

The effectiveness evidence referred to 2000 to 2002 and the resource use data to 1993 to 2003. The price data related to 2002 to 2003.

The effectiveness data were derived from published studies.

A Markov state-transition model was used to estimate the cost-effectiveness of pharmacotherapies compared with cessation counselling alone over a lifetime horizon. Further details of the Markov model can be found in the appendix "Markovmodel.pdf" located on the Tobacco Control website (http://www.tobaccocontrol.com). The model assumed that the effectiveness of treatment and the associated effects on mortality were the same across all countries.

The outcomes assessed in the review included:

the natural cessation rate among smokers;

the proportion of smokers prepared to quit and who were, therefore, candidates for smoking cessation therapies;

the effectiveness of treatment;

the mortality effects of smoking cessation;

the length of compliance with treatment; and

the relapse rate.

The authors based their estimates of treatment efficacy on published meta-analyses. Mortality effects of smoking cessation were derived from the American Cancer Society Prevention Study II (CPSII). No inclusion or exclusion criteria were specified for the studies considered in the review.

Not reported.

Not reported.

The validity of the primary studies does not appear to have been assessed.

The review included approximately 15 primary studies.

Not reported.

The authors do not appear to have investigated differences between the primary studies.

The results of the review indicated that the natural cessation rate among smokers was 2.5% (range: 1 to 4).

The proportion of smokers in the "preparation stage" of quitting was 25% (range: 10 to 40).

The odds ratio (OR) for cessation at one year compared with no treatment was estimated 1.73 (range: 1.46 to 2.03) for cessation counselling alone.

The OR for cessation at one year compared with cessation counselling alone was 1.66 (range: 1.52 to 1.82) for nicotine gum, 1.80 (range: 1.61 to 2.01) for nicotine patch, 2.35 (range: 1.63 to 3.38) for nicotine spray, 2.14 (range: 1.44 to 3.18) for nicotine inhaler and 2.51 (range: 1.5 to 3.0) for bupropion.

The percentage of smokers still under treatment was 50% (range: 40 to 60) after the first month and 20% (range: 15 to 25) after the second month.

The lifetime probability of relapse after one year of abstinence was 35% (range: 10 to 50).

The authors made several model assumptions when deriving the effectiveness estimates.

Quit ORs were assumed to be the same for men and women.

The mortality rate of former smokers was assumed to rejoin that of never smokers approximately 20 years after quitting.

Based on the findings of the CPSII, the authors extrapolated the mortality curves to age 90 and assumed a phase-in period of 25 years for former smokers' mortality risk to return to that of non smokers.

The measure of benefits used was the life-years gained.

The study included the direct costs to a third-party payer. These included the costs of NRT, bupropion and GP time. The quantities of resource use were described, but the unit costs were not reported. The direct cost data were derived from public and private price data, official government data and private medical reimbursement rates. The price data obtained from retail outlets may not match the price paid by hospitals or third-party payers. In addition, the data from public pricing lists are unlikely to correspond to private prices. Discounting was relevant given that a lifetime horizon was used, so a discount rate of 3% per annum was applied. The price data related to 2002 to 2003.

Since direct sampled data were not available, a stochastic analysis was not possible.

The indirect costs were not included in the analysis. This was appropriate given the third-party payer perspective.

US dollars ($). The 2002 to 2003 conversion rates were US $1 equals 1.42 Swiss francs, 0.64 UK pounds sterling, 1.01 Euros and 1.54 Canadian dollars.

The authors conducted one-way sensitivity analyses around the model parameters, including the discount rate. They explored variability in the data by starting age of the cohort and treatment efficacy, among other parameters. The authors stated that the source of the ranges tested for the treatment efficacy parameters were the reported 95% confidence intervals. The source of the other ranges tested was not reported.

The authors did not report the benefits used to calculate the incremental cost-effectiveness ratios (ICERs).

The authors did not report the total costs used in calculating the ICERs.

The costs and benefits were combined to calculate the cost per life-year gained.

The cost per life-year gained with cessation counselling only compared with no treatment was, for 45-year-old men and women respectively:

$190 and $288 in Spain,

$375 and $567 in Switzerland,

$389 and $588 in Canada,

$479 and $724 in France,

$623 and $941 in the USA, and

$773 and $1,168 in the UK.

The minimum and maximum ICERs for each pharmacotherapy in persons aged 45 year were:

$2,230 for men in Spain and $7,643 for women in the USA for nicotine gum;

$1,758 for men in Spain and $5,131 for women in the UK for nicotine patch;

$1,935 for men in Spain and $7,969 for women in the USA for nicotine spray;

$3,480 for men in Switzerland and $8,700 for women in France for nicotine inhaler; and

$792 for men in Canada and $2,922 for women in the USA for bupropion.

A discount rate of 3% per annum was applied to the costs and benefits.

"The cost-effectiveness of the first-line pharmacotherapies for smoking cessation varied significantly across Canada, France, Spain, Switzerland, the United Kingdom, and the United States. In each case, bupropion and secondly the patch were the most cost-effective treatments. In each country, pharmacotherapies for smoking cessation would be considered favourable in terms of cost-effectiveness as compared to several other common preventive drug treatments."

Although no explicit justification for the choice of the comparators was provided, the authors included no treatment and a range of smoking cessation strategies and did not appear to omit any important relevant comparators. You must assess whether the range of comparators considered is relevant to your own setting.

The authors did not claim that a systematic review of the literature had been undertaken. The estimates of treatment efficacy were obtained from meta-analyses, which themselves might have been based on a systematic review. It was unclear, however, if these meta-analyses represented all available data, or if they were identified in a systematic way. It was also unclear how data from the primary studies were combined, or if there was any combination of primary data. The source of the ranges tested in the sensitivity analyses was stated for the estimates of treatment efficacy, but not for other input parameters. The impact of differences between the primary studies on the effectiveness estimates was not considered. The authors stated that they had to assume that treatment efficacy was the same in each country and across individuals, owing to a lack of data on differential effects. Sensitivity analyses were conducted to improve both the internal validity of the study and the generalisability of the results.

The estimation of health benefits was modelled. The authors stated that the Markov model used to estimate the health benefits had been validated for this purpose. The authors did not report the health benefits separately, which inhibits interpretation of the study results. The authors acknowledged that they failed to incorporate health-related quality of life, but concluded that this is likely to be conservative with respect to estimating the cost-effectiveness of smoking cessation therapies.

The authors considered only the costs of providing treatment. They omitted the costs of treating smoking-related diseases, which would have been relevant given the third-party payer perspective. This is likely to have a conservative impact on estimating the cost-effectiveness of smoking cessation therapies, as smoking-related costs may be expected to be higher with no treatment. The unit costs were not reported, which reduces the ability of readers to assess the validity of the model inputs. A sensitivity analysis of the costs was conducted, although the source of the ranges used was not clear. The price year was not reported, although the majority of prices and the exchange rates used related to 2002 to 2003. The authors acknowledged that the lack of a standardised data source for all countries studied was a limitation of their analysis, as it allowed variation in the methods used to collect data. Since the costs were incurred during a long period (more than 2 years), discounting was relevant and was appropriately conducted.

The authors made appropriate comparisons of their findings with those from other studies, and found them to be in broad agreement. They concluded that the ICERs of pharmacotherapies for smoking cessation fell within the bounds of other preventive treatments considered cost-effective. The issue of generalisability between countries was a major focus of the study. The authors failed to report the unit costs, estimated total costs and total benefits separately, reporting only the ICERs. The authors' conclusions reflected the scope of the analysis.

The authors recommended that to increase the use of pharmacotherapies for smoking cessation, GPs must be offered encouragement to systematically recommend their use and pharmacotherapies must be made accessible to patients.

None stated.