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.

A telephone helpline (quitline) for smoking cessation was examined. The helpline operated 51 hours per week through three to four telephone lines. When the service was closed, or all lines were busy, an answering machine and a 24-hour interactive voice response served as back-up and all calls were registered on computerised patient records.

Treatment.

Cost-effectiveness analysis.

The study population comprised smokers in the general population.

The setting was the community. The economic study was carried out in Sweden.

The effectiveness data and most resource use data were derived from studies published between 2001 and 2004. The price year was 2002.

The effectiveness evidence was derived from a synthesis of completed studies.

The outcomes estimated from the literature were the quit rates and life expectancy. Quit rates or abstinence were defined as not a single puff of smoke 7 days before follow-up by self-report.

The primary studies appear to have been identified selectively rather than through a systematic review of the literature. The quit rates came from a study carried out in Sweden from February 2000 to November 2001 that reported on the implementation of the quitline in a sample of 1,131 individuals followed for 2 years. The majority of these were women (about 80%). Life tables from official statistics were used to derive life expectancy for the general population in Sweden, while a large cohort study carried out in the UK was used to assess life expectancy for smokers and quitters.

Not stated.

Not stated.

Not reported.

Three primary studies provided the clinical data.

A narrative approach was used to combine the primary estimates.

Not stated.

The absolute quit rate (or abstinence) associated with the use of the helpline was 31%. In a more conservative approach, the quit rate was estimated at 24% when it was taken into account that 7% were spontaneous quitters. The 7% of spontaneous quitters can be considered as the quit rate associated with the comparator (no intervention).

Life expectancy was reported for groups of differing age and gender.

Gains in life expectancy for women were 7.7 years for quitting smoking at age 35 years, 7.2 years at age 45, 5.6 years at age 55 and 5.1 at age 65. The corresponding values for men were 8.5 years (age 35), 7.1 years (age 45), 4.8 years (age 55) and 2.0 years (age 65).

The summary benefit measures used were the quit rate and the estimated life-years gained (LYG) with the quitline in comparison with no intervention. Discounting was applied at two different rates, 3% and 5%.

The perspective adopted in the analysis of the costs was not stated. The cost categories considered in the study were the fixed costs associated with the telephone service and the costs of pharmacological treatments used to help quitting, such as bupropion. The costs were grouped as personnel, materials and services. Materials covered office supplies, library service, forms, stationary and miscellaneous costs. Services covered rent of office premises, equipment, information technology services, printing, advertising, telephone, fax, travel, cleaning, and the cost of consultants. Potential future cost-savings were not taken into consideration. The unit costs were not presented separately from the quantities of resources used for all items. Resource use was estimated on the basis of both data gathered in the Swedish study and authors' opinions. The costs came from the financial records of the Center for Tobacco Prevention, which was the operator of the quitline, and out-of-pocket expenses borne by the patients (i.e. pharmacological treatment). Discounting was appropriately applied, as the costs were incurred during a 22-month period, and both 3% and 5% rates were applied. The price year was 2002 and prices obtained from other years were converted to 2002 using the Consumer Price Index.

The costs were treated deterministically.

The indirect costs were not included in the economic evaluation.

The costs were estimated in Swedish kroner (SEK) and converted into US dollars ($) at an average exchange rate of 9.721 in 2002.

A sensitivity analysis was carried out to assess the robustness of the base-case cost-effectiveness ratios to variations in the quit rates and LYG. The sensitivity of the data on life-years saved was tested by calculating the effects of 2, 4 and 6 life-years saved, instead of an average of 8 life-years saved. The impact of using alternative quit rates ranging from 6 to 31% was also estimated. Finally, a 0% discount rate was assumed for health benefits.

The (discounted) accumulated number of LYG was 2,400. The quit rate was 31% when only the benefits of the quitline were considered and 24% when the effect of 7% of spontaneous quitters was taken into consideration.

The costs of the helpline during the 22-month study period were broken down into $475,095 for personnel, $23,766 for materials and $200,382 for services. Thus, the total costs of the quitline were $699,243.

Incremental cost-effectiveness ratios were calculated to combine the costs and benefits of the quitline in comparison with no intervention. It is worth noting that the costs associated with the implicit comparator were set at 0.

The incremental cost per quitter was $1,052 using the more conservative approach (7% of spontaneous smokers) and $1,360 when assuming a 31% quit rate. The incremental cost per LYG ranged from $311 (more conservative approach) to $401.

The sensitivity analysis showed that the cost per LYG did not change substantially when assumptions about LYG were changed. With an abstinence rate as low as 6%, the cost per LYG was $1,607 (worst-case scenario).

When no discount rate for the benefits was considered, the cost per LYG was $29 when using a 31% quit rate.

Quitlines were a cost-effective strategy for smoking cessation in Sweden and compared favourably with other smoking cessation policies.

The comparator was not explicitly stated or described. However, implicitly, it was 'no intervention', and its costs and effectiveness (in the base-case) were assumed to have been zero. You should decide whether this is a valid comparator in your own setting.

The effectiveness evidence came from a synthesis of studies that were presumably identified selectively. The authors described extensively the characteristics of the study sample included in the study that provided the quit rates. This study was specific to the Swedish setting and reported on the effectiveness evaluation of the Swedish quitline. The other sources of data were used to derive information on life expectancy, which was required to calculate the LYG. Data extracted from the three sources were combined in a narrative. The use of US survival data in a Swedish population might have introduced some bias, but the authors stated that the smoking panorama is quite comparable between the two countries. Sensitivity analyses were carried out to assess the impact of different quit rates on the cost-effectiveness results.

The use of LYG as the summary benefit measure was appropriate as it captures the impact of the intervention on the most relevant dimension of health (i.e. survival). Further, life expectancy can be compared with the benefits of other health care interventions. The impact of using different discount rates was investigated. An intervention-specific summary benefit was also used. The authors noted that the analysis focused on mortality, but smoking cessation leads also to reductions in morbidity and improvements in quality of life. Thus, the benefits of the quitline might have been underestimated.

It was unclear whether the cost categories included were appropriate since the perspective adopted in the study was not stated. Information on the unit costs and the quantities of resources used was not presented separately, which limits the possibility of replicating the results of the study in other settings. The source of the costs was clear. Statistical analyses were not carried out, and the impact of variations in the cost estimates was not investigated in the sensitivity analysis. The price year was reported, which means that reflation exercises in other time periods will be possible.

The authors reported the cost-effectiveness ratios of several smoking cessation strategies, and the current intervention was one of the most cost-effective. In effect, the authors noted that individuals participating in the quitline programme were likely to have been more motivated to quit smoking than the average smoker. Thus, a range for cost-effectiveness ratios was calculated in order to consider both hypothetical and more 'realistic' scenarios of intervention efficacy. The authors did not compare their results with those from other published studies because this was the first published study evaluating the cost-effectiveness of a quitline. The issue of the generalisability of the study results to other settings was not addressed and the costs were specific to the Swedish setting. In general, the external validity of the analysis appears to have been low, although the authors conducted a sensitivity analysis in which different quit rates were used.

The study results support the implementation of the quitline programme, which should form part of a comprehensive, publicly funded, national tobacco control policy.

Helgason A, Tomson T, Lund KE, et al. Factors related to abstinence in a telephone helpline for smoking cessation. Eur J Public Health 2004;14:306-10.

Warner KE. Cost effectiveness of smoking cessation therapies. Interpretation of the evidence and implications for coverage. Pharmacoeconomics 1997;11:538-49.

Zhu SH, Anderson CM, Tedeschi GJ, et al. Evidence of realworld effectiveness of a telephone quitline for smokers. N Engl J Med 2002;347:1087-93.