RESEARCH

Smoking-related deaths averted due to three years of policy progress

Des décès liés au tabagisme évités grâce à trois ans d'avancée politique

Fallecimientos derivados del tabaco evitados gracias al progreso de las medidas de control a lo largo de tres años

David T Levy^I,*; Jennifer A Ellis^II; Darren Mays^I; An-Tsun Huang^I

^IDepartment of Oncology, Georgetown University, 3300 Whitehaven Street NW, suite 4100, Washington, DC 20007, United States of America (USA)
^IIBloomberg Philanthropies, New York, USA

ABSTRACT

OBJECTIVE: To evaluate the global impact of adopting highest-level MPOWER tobacco control policies in different countries and territories from 2007 to 2010.
METHODS: Policy effect sizes based on previously-validated SimSmoke models were applied to determine the reduction in the number of smokers as a result of policy adoption during this period. Based on previous research suggesting that half of all smokers die from smoking, we also derived the estimated smoking-attributable deaths (SADs) averted due to MPOWER policy implementation. The results from use of this simple yet powerful method are consistent with those predicted by using previously validated SimSmoke models.
FINDINGS: In total, 41 countries adopted at least one highest-level MPOWER policy between 2007 and 2010. As a result of all policies adopted during this period, the number of smokers is estimated to have dropped by 14.8 million, with a total of 7.4 million SADs averted. The largest number of SADs was averted as a result of increased cigarette taxes (3.5 million), smoke-free air laws (2.5 million), health warnings (700 000), cessation treatments (380 000), and bans on tobacco marketing (306 000).
CONCLUSION: From 2007 to 2010, 41 countries and territories took action that will collectively prevent nearly 7.5 million smoking-related deaths globally. These findings demonstrate the magnitude of the actions already taken by countries and underscore the potential for millions of additional lives to be saved with continued adoption of MPOWER policies.

RÉSUMÉ

OBJECTIF: Évaluer l'impact global de l'adoption de politiques de lutte antitabac MPOWER du plus haut niveau dans les différents pays et territoires de 2007 à 2010.
MÉTHODES: Les ampleurs des effets des politiques basées sur des modèles SimSmoke préalablement validés ont été appliquées pour déterminer la baisse du nombre de fumeurs à la suite de la mise en place de politiques antitabac au cours de cette période. Sur la base de recherches antérieures suggérant que la moitié des décès de fumeurs sont dus au tabagisme, le nombre de décès attribuables au tabagisme (DAT) évités grâce à l'adoption de politiques MPOWER a pu être évalué. Les résultats de cette méthode simple et performante sont conformes à ceux envisagés par les modèles SimSmoke préalablement validés.
RÉSULTATS: Au total, 41 pays ont adopté au moins une politique MPOWER du plus haut niveau entre 2007 et 2010. En conséquence de toutes les politiques adoptées au cours de cette période, il a été estimé que le nombre de fumeurs a baissé de 14,8 millions, pour un total de 7,4 millions de DAT évités. La plupart des DAT ont été évités grâce à l'augmentation des taxes sur les cigarettes (3,5 millions), aux lois antitabac (2,5 millions), aux campagnes d'informations sanitaires (700 000), aux traitements de sevrage (380 000) et à l'interdiction de la commercialisation du tabac (306 000).
CONCLUSION: De 2007 à 2010, 41 pays et territoires ont pris des mesures qui ont permis d'éviter près de 7,5 millions de décès liés au tabagisme dans le monde. Ces résultats démontrent l'ampleur des mesures déjà prises par tous les pays et soulignent la possibilité de sauver des millions d'autres vies en poursuivant l'adoption de mesures MPOWER.

RESUMEN

OBJETIVO: Evaluar la repercusión global de la adopción de las medidas de control del tabaco de alto nivel MPOWER en diferentes países y territorios desde el año 2007 al 2010.
MÉTODOS: A efectos de determinar la reducción del número de fumadores como consecuencia de la adopción de las medidas de control durante ese período, se aplicaron tamaños del efecto del plan de medidas basados en los modelos SimSmoke validados con anterioridad. Con arreglo al anterior estudio, que sugería que la mitad de todos los fumadores fallecían por motivos derivados del tabaco, también colegimos los fallecimientos relacionados con el tabaco evitados gracias a la implementación del plan de medidas MPOWER. Los resultados obtenidos del uso de este método sencillo, pero potente, concuerdan con los previstos utilizando los modelos SimSmoke validados anteriormente.
RESULTADOS: En total, 41 países adoptaron al menos una medida de alto nivel MPOWER entre 2007 y 2010. A consecuencia de todas las medidas adoptadas durante este período, se estima que el número de fumadores descendió en 14,8 millones, con un total de 7,4 millones de fallecimientos relacionados con el tabaco evitados. La mayoría de los fallecimientos se evitó gracias al incremento de los impuestos al tabaco (3,5 millones), la legislación antitabaco (2,5 millones), las advertencias sanitarias (700 000), los tratamientos para dejar de fumar (380 000) y las prohibiciones publicitarias (306 000).
CONCLUSIÓN: De 2007 a 2010, 41 países y territorios tomaron medidas que evitaron en conjunto unos 7,5 millones de fallecimientos relacionados con el tabaco. Estos resultados demuestran la magnitud de las acciones ya instauradas por los países y ponen de relieve el potencial de la adopción continuada de las medidas MPOWER para salvar millones de vidas más.

Introduction

The World Health Organization (WHO) Framework Convention on Tobacco Control (FCTC) has been one of the most rapidly embraced treaties in the history of the United Nations. To assist countries with implementing FCTC obligations, in 2008 WHO introduced the MPOWER package of evidence-based tobacco control measures. The MPOWER package includes: monitoring tobacco use and tobacco control policies; protecting people from the dangers of tobacco smoke; offering help to quit tobacco; warning the public about the dangers of tobacco; enforcing bans on tobacco advertising, promotion and sponsorship; and raising tobacco taxes. These measures are supported by substantial evidence of their impact for reducing smoking. ^1–4

Since their introduction, countries have steadily implemented recommended MPOWER policies. Early leaders in adopting rigorous MPOWER measures include Brazil,⁵ Ireland⁶ and Thailand. ⁷ As the movement to adopt MPOWER policies gains momentum, more countries have passed laws and many others have put advocacy efforts in place to support legislation. This progress has also been aided by significant support from private donors.

In 2008, WHO began systematically gathering information about MPOWER measures in every Member State to assess progress with the adoption of legislation, motivate further change and provide examples of excellence. In 2008, 2009 and 2011 WHO published reports that provide information about MPOWER implementation. ^8–10 Findings of the most recent report, which includes data from 2010, showed that 3.8 billion people were covered by at least one MPOWER category globally and demonstrated progress in every MPOWER category.

The goal of the MPOWER package is to reduce smoking-attributable deaths (SADs), which are projected to rise to 8.3 million annually by 2030. ¹¹ The purpose of this study is to determine the number of SADs averted as a result of the implementation of MPOWER policies. By capitalizing on the previously-validated SimSmoke model^5–7,12–20 to estimate the impact of MPOWER policies for reducing SADs, we employ a simple but powerful method to estimate the long-term impact of MPOWER policy implementation in the first three years.

Methods

Assessment of incremental policy change

The WHO reports^8–10 show the status of each MPOWER policy, by country or territory, for 2007, 2008 and 2010 using a five-level categorization. The highest level comprises the MPOWER measures that WHO placed in the highest category in terms of the completeness of the legislation. For example, WHO categorizes smoke-free laws at the highest level as those calling for "all public places smoke-free or > 90% of the population covered by subnational smoke-free legislation".

Using these criteria, we created a list of countries and territories that had adopted a highest-level policy after 2007, commensurate with the timing of WHO reports. We also used WHO corrigenda to clarify or correct policy status based on additional knowledge gained after publication of the reports. ²¹ The list of countries and territories adopting highest-level policies was confirmed by representatives of the WHO Tobacco Free Initiative.

We compared 2010 data for the POWER (monitoring is excluded from this report) categories to 2007 data to determine which countries not in the highest-level category in 2007 had progressed to the highest level by 2010. For each country that had adopted a highest-level policy by 2010, policy data from earlier reports were used to determine incremental policy changes because the effect of a policy change depends upon the initial policy level (e.g. the incremental effect of a complete ban is less if a partial ban was in place than if no policy existed).

Effect size parameters

Effect sizes for incremental policy change were derived from the SimSmoke tobacco control policy model. ^2,22,23 The SimSmoke model has been validated for four states of the United States of America and more than 20 countries. ^{5–7,12–20,23,24}Table 1 summarizes policies and effect sizes based on expert panels and published literature reviews. ²³ We provide upper and lower bounds for effect sizes based on variability estimates from our previous work. ¹⁴

To determine policy effect size, we used the per cent reductions (in relative terms) in smoking prevalence resulting from a particular policy. The policy effect size is classified as being either short-term or long-term: (i) a short-term effect size is defined as the percentage change in smoking prevalence in the first three years of policy implementation; (ii) a long-term effect size is the percentage change in smoking prevalence after 40 years of implementation. In SimSmoke models, a short-term effect size is based on the change in policy by 2010 from its 2007 level, and a long-term effect size is derived by multiplying the short-term effect by a long-term multiplier and by adjustment factors for health knowledge, urbanization and enforcement. In previously-developed versions of SimSmoke, the long-term multiplier is estimated for each policy (LT-MULT_p) as the ratio of the relative change in prevalence after 40 years to the relative change in short-term prevalence in the first year.

Except for tax policies, effect sizes are based primarily on policy evaluation studies from high-income countries or territories (HICs) due to the lack of studies for low- and middle-income countries or territories (LMICs). Accordingly, for LMICs, the effect size is adjusted by a health knowledge variable (AWARE_p > 1 in LMICs, and 1 in HICs), reflecting the ability to affect health awareness from policy p (excluding price policies) in LMICs (e.g. Blecher et al. ²⁶ found that advertising bans have larger effects in LMICs). Some policies are also subject to the percentage of urban population in country i, URBAN_i, measured as (1 minus the percentage employed in agriculture). ²⁷ The urban adjustment incorporates the ability to reach and influence the population and those not covered by smoke-free-air laws (i.e. those not working indoors).

For smoke-free air (SFA) effects and impact of smoke-free air laws, we posit that half of the effects will occur automatically through passage of the law (e.g. due to a change in norms), while the remaining half of the effects will depend on enforcement (ENF, 0 < ENF < 1, using the MPOWER index = 1,...10 divided by 10), and a publicity index based on tobacco control spending (PUB = 0.5 if low [< 0.05 United States dollars (US$) per capita], = 0.75 if medium [> US$ 0.05 and < US$ 0.50 per capita], and = 1 if high [> $US 0.50 per capita]). Letting ST-SFA be the short-term effect size for a HIC with high enforcement and publicity (PUB), the long-term SFA effect (LT-SFA_i) for country i is:

where ST-SFA_i is country i's short-term effect size from the change in policy if it were a HIC.

The marketing restrictions (MR) equation omits the URBAN and PUB effect (i.e. URBAN = PUB = 1), so the long-term MR effect for country i is:

For cessation treatment policies (CTP), the long-term effect (LT-CTP_i) is:

where ST-CTP_i is the short-term CTP effect size in a HIC from the change in CTP policy.

For health warnings (HWs), URBAN is set equal to one. The long-term HW effect is:

The effect for price policies (PRICE) depends on standard economic demand elasticities, which we assume are constant over price changes and are applied to the smoking prevalence. The formula for the prevalence elasticity, E, is defined in terms of the inflation-adjusted price, Pr, and prevalence, PREV. Using an arc elasticity formula, price change is measured as the price change between 2007 and 2010 relative to the average price, i.e. (P₂₀₁₀ – P₂₀₀₇)/[0.5(P₂₀₁₀ + P₂₀₀₇)]. The short-term prevalence elasticity is negative and written as:

Since the change in prevalence can be calculated as:

and assuming constant price elasticities,²⁵ the price effect size for country i is calculated as:

Based on previous reviews,²⁸ overall short-term elasticities were set at −0.15 for HICs, −0.2 for MICs and −0.2 for LICs, with a long-term multiplier of 2.

Reduction in smokers and smoking-attributable deaths

Adjusted smoking prevalence for those 15 years of age or older, by sex, for 2007 or a previous year were obtained from the 2011 WHO report for all countries except Colombia, whose adjusted prevalence was obtained from InfoBase. Prevalence reported is for "current smoking" and includes people who smoke cigarettes and other tobacco products, such as kreteks and bidis, every day and some days. The number of smokers of each sex was obtained by multiplying the respective smoking prevalence by population data for sex from the World population prospects, 2008 revision. ¹⁷ The number of smokers, by sex, for country i is obtained as:

where PREV is smoking prevalence and Pop is populations (ages 15+), both distinguished by male (m) and female (f).

Applying the relevant policy long-term effect size to the number of smokers, we calculated the reduction in smokers in country i as a result of policy p, ΔSMOKERS_i,p, as:

where LT-EFFECT SIZE_i,p is the long-term policy effect size of a specific policy p in country i.

The number of SADs averted was calculated by applying an algorithm based on Doll et al. ,^29,30 who concluded that "half of all regular cigarette smokers will eventually be killed by their habit". Similar results have been obtained in studies in the United States. ^31,32 We applied the 50% figure to smokers who quit as a result of the policy to estimate the deaths averted, ΔSMOKERS DEATHS, as:

Results

Study population

Table 2 presents national income status, initial smoking prevalence by sex, policy effect size, estimated reduction in the number of smokers between 2007 and 2010, and reduction in SADs by policy and country. In total, 41 countries implemented a highest-level MPOWER policy; 8 implemented more than one highest-level policy. Smoking prevalence estimates varied from < 1.0% for females in Egypt to > 50% for males in Malaysia and Turkey.

Smoke-free air policies

Of the 20 countries implementing complete SFA policies, three are classified as LICs, 12 as MICs, and five as HICs. SFA laws were already in use at worksites in Spain in 2007; restaurants and bars were added by 2010. For other countries (e.g. Turkey), SFA policies were implemented for most locations between 2007 and 2010. Countries such as Burkina Faso and Chad have large rural sectors leading to small effects, whereas Colombia, Nauru, Peru, Seychelles and Turkey have a smaller agricultural sector and hence show larger effects. Owing to their large populations, Pakistan, Thailand and Turkey had the greatest number of SADs averted. In total, 5 032 629 individuals who were smoking in 2007 quit smoking as a result of SFA policies. Consequently, an estimated 2 516 314 SADs are averted (range: 1 258 157–3 774 472).

Cessation treatment policies

Two HICs and two MICs adopted highest-level cessation policies. Effect sizes vary from −1.8% for Turkey to −5.4% for Israel depending on the country's income status and the policies in place in 2007. Collectively, changes in cessation policies are estimated to lead to 759 150 fewer smokers and 379 575 averted SADs (range 189 768 to 759 150).

Health warnings

Seven countries adopted highest-level pack warning policies. All were MICs except for Djibouti. Since most countries were middle-income and their health warnings changed from weak to strong, their long-term effect sizes were −3.0%. For other countries, changes in the warnings or a change to HIC status yielded an effect size of −1.5%. These policy changes reduced the number of smokers by 1 379 580 and averted 689 790 SADs (range:344 895–1 034 685).

Advertising policies

Highest-level bans on tobacco advertising and promotions were adopted by four countries, all of which are MIC except for Chad. Panama adopted a marketing ban without prior restrictions (effect size: −13%), whereas most other countries had several restrictions in place in 2007. Overall, advertising policy changes resulted in 306 323 SADs averted (range: 153 162–459 485).

Price and taxes

Taxes were raised to the MPOWER goal of 75% of the final retail price in 14 countries, including one LIC, five MICs and eight HICs (Table 2). Relative changes in price from 2007 levels varied. Finland had relatively high tax rates in 2007 with increases of less than 5%, as did the Czech Republic and Greece. Other countries had inflation-adjusted price changes of at least 10%: Argentina (21%), Estonia (30%), Israel (41%), Italy (10%), Latvia (102%), Lithuania (53%), Madagascar (28%), Romania (69%), Slovenia (15%) and Turkey (49%). The largest number of deaths was averted in Argentina, Italy, Romania and Turkey, a reflection of the magnitude of the price increase, of initial smoking prevalence and of the size of the total population. Price increases are estimated to lead to a reduction of 7 056 006 smokers and to the aversion of 3 528 003 SADs (range: 2 646 002–4 410 004).

Total effects

The total estimated impact of each of the five MPOWER policies on the number of current smokers and of deaths prevented is shown in Table 3. The estimated number of smokers will be reduced by 14 840 011 and an estimated 7 420 006 SADs will be averted (range: 4 592 004–10 437 796) among smokers alive in 2007.

Discussion

The highest-level MPOWER policies adopted from 2007 to 2010 will result in 15 million fewer smokers, and 7.4 premature deaths will consequently be averted by 2050. These findings underscore the urgent need for countries to focus on adopting, implementing and enforcing MPOWER policies. Although more than 40 countries adopted one or more MPOWER policies from 2007 to 2010, nearly half of the world population is still not covered by a single MPOWER policy. The policies having the greatest impact – smoke-free air laws and taxation – are alarmingly under-adopted. Only 11% of the world's population is protected by SFA laws and less than 8% resides in countries with the recommended minimum tobacco tax rates.

MPOWER policy implementation is likely to yield additional benefits beyond the nearly 7.5 million estimated SADs averted. The estimates do not include the beneficial effects of adopted policies that have strong components but do not qualify as being of the highest level in the WHO reports. Our estimates also exclude smokers who may have initiated smoking after 2007 in the absence of strong policies. However, smokers who were included in our estimates and who may have quit smoking later despite the absence of such policies potentially offset some of the benefits of policy implementation that were omitted from our analysis.

We used data from WHO reports and an extensively validated statistical model to estimate MPOWER policy impact. This supports the validity of our estimates. Despite these strengths, the findings should be interpreted in light of the limitations of this work. The estimation method does not incorporate dynamic aspects of changing demographics and smoking rates and the effects of policies over time. To explicitly consider the impact of incorporating these dynamic aspects, we applied the SimSmoke simulation model in a separate analysis to nine countries that are reaching MPOWER goals (data not shown). For the nine countries we compared the effects on smoking prevalence and SADs with the effects reported in the previous section and found that they were reasonably close to the reported findings. This suggests that the dynamic aspects of policy change that were not taken into account in our analysis did not substantially bias our findings.

The policy data used for our analysis were reported by WHO, which adheres to a specific set of policies described by the FCTC. To collect the data, WHO reviews actual legislation in its original language and catalogues policy status. The categorization of the policy for inclusion in the report is then reviewed at the regional and national levels by various entities, including each country's health ministry. Although these are the best available data that could be ascertained for our analyses, the characteristics of a policy as implemented in a specific location may not be fully captured by WHO reports. For example, the effects of smoking cessation policies may depend on financial reimbursement and the involvement of health-care workers, but these data are not collected. More specific national-level data may allow for more precise estimation of the impact of FCTC policy on smoking-related morbidity and mortality. Another limitation lies in the ability to model the combined effects of more than one policy simultaneously. Although there may be synergies, policies may also offset one another. The literature provides little guidance on the potential synergistic or offsetting effects of multiple concurrent policies.

Finally, our analysis depended on estimates of relative risks of death based on data for HICs. Studies indicate that LMICs have lower relative risks: e.g. 1.35 for China^33–35 and 1.6 for the Republic of Korea and Taiwan, China,^36–38 compared with > 2.0 for the United States. ^39–41 These differences may reflect initiation at older ages or a higher background risk of premature death from other causes. To consider these effects, we multiplied the SAD estimates for LMICs (using World Bank classifications) by 0.6, based on results from previous SimSmoke models that show 40% fewer deaths in MICs. ^7,42 The total estimated number of SADs averted was reduced to 5 130 076 (range: 3 139 571–7243, 484). Of these averted SADs, 2 298 131 resulted from taxes, 1 610 787 from SFA laws, 791 657 from health warnings, 245 707 from cessation treatment and 183 794 from marketing restrictions. However, smoking deaths in LMICs, especially those approaching high-income status (e.g. Mexico and Turkey), may increase as the intensity and duration of smoking increase and risks not related to smoking decline. ⁴³

Our findings may have implications beyond the outcomes examined in this study. The effect of newly implemented policies could be extended to reductions not only in SADs, but also to benefits in other smoking-related outcomes (e.g. reduced adverse birth outcomes related to maternal smoking, such as low birth weight,⁴⁴ better quality of life, lower health-care costs, less productivity loss). Moreover, evidence suggests that additional deaths may be averted among non-smokers because of reductions in exposure to second-hand smoke. ⁴⁵ Finally, we have not considered the possible effects of the MPOWER policies on reductions in smokeless tobacco prevalence, which would likely bring additional public health benefits. ^46,47

In conclusion, we found a substantial projected impact on SADs resulting from the adoption of MPOWER policies from 2007 to 2010. Our main finding – that nearly 7.5 million smoking-related deaths will be averted – shows that evidence-based MPOWER tobacco control measures have an enormous potential to reduce premature smoking-related mortality. The 41 countries that adopted such policies from 2007 to 2010 represent a cumulative population of nearly 1 billion people (one seventh of the 2008 world population of 6.9 billion). If the progress attained by these 41 countries were extended globally, tens of millions of smoking-related deaths could be averted. It is imperative that the public health community continue to advocate for MPOWER policies of the highest level.

Funding: David Levy has received funding from the International Union Against Tuberculosis and Lung Disease to conduct this study and has received funding from the Cancer Intervention and Surveillance and Modeling Network (CISNET of DCPS, NCI under grant U01-CA97450-02 for general development of the SimSmoke model).

Competing interests: None declared.

References

1. Hopkins DP, Briss PA, Ricard CJ, Husten CG, Carande-Kulis VG, Fielding JE et al. ; Task Force on Community Preventive Services. Reviews of evidence regarding interventions to reduce tobacco use and exposure to environmental tobacco smoke. Am J Prev Med 2001;20(Suppl):16–66. doi: http://dx.doi.org/10.1016/S0749-3797(00)00297-X PMID: 11173215        

2. Levy DT, Chaloupka F, Gitchell J. The effects of tobacco control policies on smoking rates: a tobacco control scorecard. J Public Health Manag Pract 2004;10:338–53. PMID: 15235381        

3. Centers for Disease Control and Prevention. Preventing tobacco use among young people: a report of the Surgeon General. Atlanta: National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 1994.         

4. Centers for Disease Control and Prevention. Reducing tobacco use: a report of the Surgeon General. Atlanta: National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2000.         

5. Levy D, de Almeida LM, Szklo A. The Brazil SimSmoke policy simulation model: the effect of strong tobacco control policies on smoking prevalence and smoking-attributable deaths in a middle income nation. PLoS Med 2012;9:e1001336. doi: http://dx.doi.org/10.1136/tc.2007.022319 PMID: 18218810        

6. Currie LM, Blackman K, Clancy L, Levy DT. The effect of tobacco control policies on smoking prevalence and smoking-attributable deaths in Ireland using the IrelandSS simulation model. Tob Control 2012. epub May 26. doi: http://dx.doi.org/10.1136/tobaccocontrol-2011-050248 PMID: 22634570        

7. Levy DT, Benjakul S, Ross H, Ritthiphakdee B. The role of tobacco control policies in reducing smoking and deaths in a middle income nation: results from the Thailand SimSmoke simulation model. Tob Control 2008;17:53–9. doi: http://dx.doi.org/10.1371/journal.pmed.1001336 PMID: 23139643        

8. WHO report on the global tobacco epidemic, 2008: the MPOWER package. Geneva: World Health Organization; 2008.         

9. WHO report on the global tobacco epidemic, 2011: the MPOWER package. Geneva: World Health Organization; 2012.         

10. WHO report on the global tobacco epidemic, 2009: implementing smoke-free environments. Geneva: World Health Organization; 2009.         

11. Mathers CD, Loncar D. Projections of global mortality and burden of disease from 2002 to 2030. PLoS Med 2006;3:e442. doi: http://dx.doi.org/10.1371/ journal.pmed.0030442 PMID: 17132052        

12. Levy DT, Currie L, Clancy L. Tobacco control policy in the UK: blueprint for the rest of Europe? Eur J Public Health 2013;23:201–6. doi: http://dx.doi.org/10.1093/eurpub/cks090 PMID: 22826505        

13. Levy D, Gallus S, Blackman K, Carreras G, La Vecchia C, Gorini G. Italy SimSmoke: the effect of tobacco control policies on smoking prevalence and smoking attributable deaths in Italy. BMC Public Health 2012;12:709. doi: http://dx.doi.org/10.1186/1471-2458-12-709 PMID: 22931428        

14. Levy D, Zaloshjna E, Blackman K, Chaloupka F, Fong GT, editors. The role of tobacco control policies in reducing smoking and deaths caused by smoking in the eighteen nations with the largest smoking burden. Rockville: National Cancer Institute. Forthcoming.         

15. Levy DT, Blackman K, Currie LM, Mons U. Germany SimSmoke: the effect of tobacco control policies on future smoking prevalence and smoking-attributable deaths in Germany. Nicotine Tob Res 2013;15:465–73. doi: http://dx.doi.org/10.1093/ntr/nts158 PMID: 22855886        

16. Levy DT, Boyle RG, Abrams DB. The role of public policies in reducing smoking: the Minnesota SimSmoke tobacco policy model. Am J Prev Med 2012;43(Suppl 3):S179–86. doi: http://dx.doi.org/10.1016/j.amepre.2012.07.037 PMID: 23079215        

17. United Nations. World population prospects: the 2008 revision. New York: UN Population Division, Department of Economic and Social Affairs; 2010.         

18. Levy DT, Hyland A, Higbee C, Remer L, Compton C. The role of public policies in reducing smoking prevalence in California: results from the California tobacco policy simulation model. Health Policy 2007;82:167–85. doi: http://dx.doi.org/10.1016/j.healthpol.2006.09.008 PMID: 17055104        

19. Levy DT, Ross H, Powell L, Bauer JE, Lee HR. The role of public policies in reducing smoking prevalence and deaths caused by smoking in Arizona: results from the Arizona tobacco policy simulation model. J Public Health Manag Pract 2007;13:59–67. PMID: 17149101        

20. Nagelhout GE, Levy DT, Blackman K, Currie L, Clancy L, Willemsen MC. The effect of tobacco control policies on smoking prevalence and smoking-attributable deaths: findings from the Netherlands SimSmoke Tobacco Control Policy Simulation Model. Addiction 2012;107:407–16. doi: http://dx.doi.org/10.1111/j.1360-0443.2011.03642.x PMID: 21906197        

21. Corrigenda WHO report on the global tobacco epidemic, 2011. Geneva: World Health Organization; 2012. Available from: http://apps.who.int/tobacco/global_report/2011/en_tfi_global_report_2011_corrigendum_1.pdf [accessed 4 April 2013]          .

22. Abrams DB, Graham AL, Levy DT, Mabry PL, Orleans CT. Boosting population quits through evidence-based cessation treatment and policy. Am J Prev Med 2010;38(Suppl):S351–63. doi: http://dx.doi.org/10.1016/j.amepre.2009.12.011 PMID: 20176308        

23. Levy DT, Bauer JE, Lee HR. Simulation modeling and tobacco control: creating more robust public health policies. Am J Public Health 2006;96:494–8. doi: http://dx.doi.org/10.2105/AJPH.2005.063974 PMID: 16449585        

24. Levy DT, Blackman K, Currie LM, Levy J, Clancy L. SimSmokeFinn: how far can tobacco control policies move Finland toward tobacco-free 2040 goals? Scand J Public Health 2012;40:544–52. doi: http://dx.doi.org/10.1177/1403494812456635 PMID: 22899560        

25. Levy DT, Cummings KM, Hyland A. Increasing taxes as a strategy to reduce cigarette use and deaths: results of a simulation model. Prev Med 2000;31:279–86. doi: http://dx.doi.org/10.1006/pmed.2000.0696 PMID: 10964642        

26. Blecher E. The impact of tobacco advertising bans on consumption in developing countries. J Health Econ 2008;27:930–42. doi: http://dx.doi.org/10.1016/j.jhealeco.2008.02.010 PMID: 18440661        

27. The world factbook. Washington: Central Intelligence Agency; 2012. Available from: http://www.theodora.com/wfb/abc_world_fact_book.html [accessed 4 April 2013]          .

28. Jha P, Chaloupka F, editors. Tobacco control in developing countries. New York: Oxford University Press; 2000.         

29. Doll R, Peto R, Wheatley K, Gray R, Sutherland I. Mortality in relation to smoking: 40 years' observations on male British doctors. BMJ 1994;309:901–11. doi: http://dx.doi.org/10.1136/bmj.309.6959.901 PMID: 7755693        

30. Doll R, Peto R, Boreham J, Sutherland I. Mortality in relation to smoking: 50 years' observations on male British doctors. BMJ 2004;328:1519. doi: http://dx.doi.org/10.1136/bmj.38142.554479.AE PMID: 15213107        

31. Taylor DH Jr, Hasselblad V, Henley SJ, Thun MJ, Sloan FA. Benefits of smoking cessation for longevity. Am J Public Health 2002;92:990–6. doi: http://dx.doi.org/10.2105/AJPH.92.6.990 PMID: 12036794        

32. Kenfield SA, Stampfer MJ, Rosner BA, Colditz GA. Smoking and smoking cessation in relation to mortality in women. JAMA 2008;299:2037–47. doi: http://dx.doi.org/10.1001/jama.299.17.2037 PMID: 18460664        

33. Liu BQ, Peto R, Chen ZM, Boreham J, Wu YP, Li JY et al. Emerging tobacco hazards in China: 1. Retrospective proportional mortality study of one million deaths. BMJ 1998;317:1411–22. doi: http://dx.doi.org/10.1136/bmj.317.7170.1411 PMID: 9822393        

34. Chen ZM, Xu Z, Collins R, Li WX, Peto R. Early health effects of the emerging tobacco epidemic in China. A 16-year prospective study. JAMA 1997;278:1500–4. doi: http://dx.doi.org/10.1001/jama.278.18.1500 PMID: 9363969        

35. Niu SR, Yang GH, Chen ZM, Wang JL, Wang GH, He XZ et al. Emerging tobacco hazards in China: 2. Early mortality results from a prospective study. BMJ 1998;317:1423–4. doi: http://dx.doi.org/10.1136/bmj.317.7170.1423 PMID: 9822394        

36. Jee SH, Lee JK, Kim IS. Smoking-attributable mortality among Korean adults: 1981-2003. Korean J Epidemiol. 2006;28:92–9.         

37. Khang YH, Lynch JW, Jung-Choi K, Cho HJ. Explaining age-specific inequalities in mortality from all causes, cardiovascular disease and ischaemic heart disease among South Korean male public servants: relative and absolute perspectives. Heart 2008;94:75–82. doi: http://dx.doi.org/10.1136/hrt.2007.117747 PMID: 17591645        

38. Wen C-P, Tsai S-P, Chen C-J, Cheng T-Y, editors. The mortality risks of smokers in Taiwan: collection of tesearch papers presented at Tobacco or Health in Taiwan. Taipei: Division of Health Policy Research, National Health Research Institutes; 2002.         

39. Thun MJ, Myers DG, Day-Lally C, Namboodiri NM, Calle EE, Flanders WD et al. Age and the exposure-response relationships between cigarette smoking and premature death in Cancer Prevention Study II. In: US Department of Health and Human Services.Changes in cigarette related disease risks and their implication for prevention and control. Bethesda: National Cancer Institute; 1997 (Monograph 9). pp. 383-475.         

40. Burns D, Garfinkel L, Samet J, editors. Changes in cigarette-related disease risks and their implication for prevention and control. Bethesda: National Cancer Institute; 1997.         

41. Centers for Disease Control and Prevention. Reducing the health consequences of smoking: 25 years of progress: a report of the Surgeon General. Atlanta: National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 1989.         

42. Gallus S, Schiaffino A, La Vecchia C, Townsend J, Fernandez E. Price and cigarette consumption in Europe. Tob Control 2006;15:114–9. doi: http://dx.doi.org/10.1136/tc.2005.012468 PMID: 16565459        

43. Flanders WD, Lally CA, Zhu BP, Henley SJ, Thun MJ. Lung cancer mortality in relation to age, duration of smoking, and daily cigarette consumption: results from Cancer Prevention Study II. Cancer Res 2003;63:6556–62. PMID: 14559851        

44. Centers for Disease Control and Prevention. The Health consequences of smoking. Atlanta: National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2004.         

45. US Department of Health and Human Services. Women and smoking: a report of the Surgeon General. Executive summary. MMWR Recomm Rep 2002;51(RR-12):i–iv, 1-13. PMID: 12222832        

46. Wen CP, Tsai SP, Cheng TY, Chen CJ, Levy DT, Yang HJ et al. Uncovering the relation between betel quid chewing and cigarette smoking in Taiwan. Tob Control 2005;14(Suppl 1):i16–22. doi: http://dx.doi.org/10.1136/tc.2004.008003 PMID: 15923442        

47. Jha P, Jacob B, Gajalakshmi V, Gupta PC, Dhingra N, Kumar R et al. ; RGI-CGHR Investigators. A nationally representative case-control study of smoking and death in India. N Engl J Med 2008;358:1137–47. doi: http://dx.doi.org/10.1056/NEJMsa0707719 PMID: 18272886        

(Submitted: 5 October 2012 – Revised version received: 25 March 2013 – Accepted: 27 March 2013)

* Correspondence to David T Levy (e-mail: dl777@georgetown.edu).