The New York Tobacco Control Program may be cost-effective

Money, cigarettes and health

Reducing the prevalence of tobacco use is best achieved through a range of tobacco control measures designed to both prevent or reduce uptake, and encourage and support existing smokers to stop smoking. These measures can include education, taxation and cessation support, but while all of these have been shown to be effective, there remains uncertainty whether they are cost-effective. During a period of economic austerity, this can be an important consideration.

This study, by Simpson and colleagues, attempted to model the effectiveness, benefits and costs of three components of the New York Tobacco Control Program. The programme supports telephone quitlines, cessation support activities, and media campaigns to promote cessation. Data from the New York Adult Tobacco Survey (NY-ATS), a telephone survey designed to be representative of adults in New York State, were used in the study, together with published data.

Methods

Breaking the habit

Expected rates of quitting were extrapolated from the literature

Analysis of NY-ATS data from 2005-2008 provided rates for media awareness, as well as quitline and nicotine replacement therapy (NRT) utilisation.

Quit rates for adults receiving combinations of these interventions were taken from the published literature. Together, this generated six groups of adult smokers:

  1. Media-aware smokers who call the quitline and receive NRT;
  2. Media-aware smokers who call the quitline but do not receive NRT;
  3. Media-aware smokers who do not call the quitline;
  4. Non-media-aware smokers who call the quitline and receive NRT;
  5. Non-media-aware smokers who call the quitline but do not receive NRT;
  6. Smokers who are reached by none of the three interventions.

Because the programme only distributes NRT through the quitline, no groups received NRT without also receiving quitline counselling.

Costs associated with the programme were computed from expenditures and budgets, taking into account the ongoing nature of the programme. This meant that a proportion of the expenditures from years before and after the period studies were also included. The overall estimated cost of the programme was $204.1 million, comprising $46.7 million for media, $9.6 million for the quitline, and $20.9 million for NRT. The remaining $126.9 million represents all other expenditures (e.g., administration).

Benefits were quantified as the smoking costs avoided due to smokers quitting as a result of the programme. This was based on estimates of smoking-attributable mortality, years of potential life lost, and productivity losses. This provides a gross benefit of $2,108 per quit per year of life without smoking. Since improved outcomes do not materialise immediately when someone stops smoking, cumulative benefits over 20 years were calculated incorporating the time it takes for benefits to be realised, resulting in a gross benefit of $20,011.53 over 20 years for a former smoker.

Results

The analysis did not include cost savings for smokers

The analysis of cost benefits focused on the health system and did not include cost savings for smokers

Monte Carlo simulation was used to analyse these data, including expected probabilities for each of the six outcomes, 10-year relapse rate and background quit rate, and some uncertainty around these values. The simulations were used to estimate additional quits (after taking into account relapses) resulting from the program, and therefore the net benefit. These indicated a likely 47,658 additional quits (95% CI 19,878 to 87,561), and a likely resulting net benefit of $771,864,156 (95% CI $211,089,844 to $1,575,068,586).

Conclusions

The authors conclude that their results indicate the likely effectiveness of the New York Tobacco Control Program, and its cost-effectiveness, in the form of reduced health costs and improved productivity, in excess of expenditures. However, there are a number of important limitations to the study which should be considered.

First, these are simulation results; although they are based on well-established figures for quit rates, costs associated with smoking and so on, we can’t be certain that the values chosen correspond to the reality in this population at this point in time. This also means that we cannot draw causal conclusions about the reach of the programme and quit rates.

Second, the authors don’t take into account taxation revenue lost as a result of smokers quitting. Whether or not this would substantially change the conclusion that the programme is cost-effective will depend on how heavily tobacco products are taxed. And whether or not we think that this is an important consideration is another issue of course!

The public health benefits of people stopping smoking are well established, but there is less evidence regarding whether tobacco control efforts are cost-effective. This study shows that under some plausible assumptions they probably are, although differences across countries (e.g., in the costs associated with smoking) will mean that these results cannot be directly generalised to other tobacco control programmes.

Reference

Simpson, S.A. & Nonnemaker, J.M. (2013). New York Tobacco Control Program cessation assistance: Costs, benefits, and effectiveness. International Journal of Environmental Research and Public Health, 10, 1037-1047.  doi:10.3390/ijerph10031037