Reductions in smoking due to ratification of the Framework Convention for Tobacco Control in 171 countries

Prolonged tobacco smoking kills more than half of long-term smokers, shortens life by roughly a decade, and contributes to more than 7 million tobacco deaths annually worldwide. Although adult smoking prevalence has declined in many countries, the absolute number of tobacco deaths continues to grow, particularly in low- and middle-income countries due to population growth, earlier onset of smoking, and the emergence of long-term smoking effects at older ages. The WHO Framework Convention on Tobacco Control (FCTC), adopted in 2003 and entering into force in 2005, is the principal global strategy to reduce smoking, encompassing demand reduction measures such as excise taxation, smoke-free policies, health warnings, and restrictions on advertising, promotion and sponsorship. The World Bank’s evidence has identified substantial increases in excise taxation as the most effective single intervention. Despite widespread ratification and the MPOWER implementation framework introduced in 2008, the magnitude of FCTC’s impact on smoking initiation and cessation—especially in LMICs—remains debated due to methodological limitations in prior evaluations. This study aims to rigorously assess whether FCTC ratification accelerated declines in youth smoking numbers and prevalence and increased cessation among older adults, and to quantify the role of concurrent tobacco tax increases.

Prior evaluations of the FCTC’s impact have produced mixed conclusions, often relying on limiting assumptions or focusing on MPOWER policy coverage rather than outcomes. Cross-country regression analyses are challenged by heterogeneity in survey instruments, shifting income levels, and concurrent policy changes, limiting causal inference. Earlier ITSAs sometimes used a fixed intervention year (e.g., 2003) irrespective of each country’s ratification timing and relied on long historical per-capita cigarette consumption series, which may weakly capture later-period dynamics. Evidence from the World Bank and other studies consistently highlights large excise tax increases as the most effective single measure for reducing initiation and increasing cessation. The present study addresses earlier limitations by aligning analyses to each country’s actual ratification year, focusing on age-appropriate outcomes (youth smoking and older adult cessation), and explicitly examining the interaction with tax changes.

Design: A multi-country interrupted time series analysis (ITSA) compared 10 years before versus 10 years after FCTC ratification, aligned to each country’s ratification year (range 2004–2010; median 2005). Outcomes and populations: (1) Number and prevalence of current smokers aged 10–24 years (initiation-sensitive group); (2) quit ratio among adults aged 45–59 years, defined as former-to-ever smokers (cessation-sensitive age). China was analyzed separately and excluded from aggregated global and income-group analyses due to its unique market structure. Data sources: Annual country-level estimates from the Global Burden of Disease (GBD) study (1990–2020) for smoking prevalence by 5-year age groups and numbers of current and former smokers, derived from 3,625 nationally representative surveys and spatiotemporal modeling where survey data were absent. Numbers were computed by multiplying prevalence by population. Country groupings followed 2021 World Bank income classifications. Tax exposure: Countries (n=161 with tax data, including China) were classified by change in total tax share on the most-sold cigarette brand between 2008 and 2012 (first years with WHO tax data; 95% of countries ratified by 2012): high-tax change (≥10 percentage-point increase) versus low-tax change (<10 pp increase, unchanged, or decrease). Modeling: Outcomes were log-transformed to estimate relative changes and address skewness. Five ITSA specifications accommodated nonlinear pre/post trends: models included combinations of linear, quadratic, and cubic trend terms pre- and post-intervention. Parameters captured level changes at ratification and slope changes after ratification. Models were estimated using ordinary least squares with Newey–West standard errors to address autocorrelation and heteroskedasticity. Model selection for best fit used Akaike Information Criterion (AIC). Stata 17 was used; margins computed average pre-ratification trends, average post-ratification trend changes, and 10-year cumulative effects. Aggregation: For each country group (e.g., by income, tax-change category), time series were centered on each country’s ratification year to maintain consistent pre/post alignment across countries. Main analyses used 10-year pre and 10-year post windows; sensitivity analyses considered longer pre-periods (up to 13 years), alternative intervention timing (e.g., setting 2005 as start for early ratifiers), and a subset of countries with high-quality survey coverage (≥10 survey years). Statistical significance was assessed at P<0.05. Reporting: Results are presented as relative annual trend changes and 10-year cumulative effects versus counterfactual continuation of pre-FCTC trends, excluding China in the global aggregates.

- Study scope: 171 countries (89% of 2021 global population); main global and income-group analyses exclude China. In the ratification year across 170 countries (excluding China), there were 153 million current smokers under age 25 and 79 million former and 231 million ever smokers at ages 45–59 (quit ratio 0.34). - Global post-FCTC annual trend changes (vs pre-FCTC; excluding China): - Prevalence under age 25: −0.8% per year (95% CI −1.0 to −0.5). - Quit ratio ages 45–59: +0.1% per year (95% CI 0.1 to 0.2). - By income group (annual trend change, post vs pre): - Low-income countries: number of young smokers −6.0%/yr (95% CI −8.1 to −4.0); prevalence −4.0%/yr (95% CI −5.6 to −2.4); quit ratio −0.2%/yr (95% CI −0.2 to −0.1). - Lower-middle-income: number −0.8%/yr (95% CI −1.0 to −0.7); prevalence −0.7%/yr (95% CI −0.8 to −0.5); quit ratio +0.1%/yr (95% CI 0 to 0.3). - Upper-middle-income: prevalence −1.0%/yr (95% CI −1.5 to −0.6); quit ratio −0.2%/yr (95% CI −1.0 to 0.6). - High-income: prevalence slope increased +1.3%/yr (95% CI 0.6 to 1.9) relative to pre-FCTC trend, though absolute prevalence still fell over time; quit ratio +0.2%/yr (95% CI 0.1 to 0.2). - Tax interaction (annual trend change, post vs pre): - High-tax change (≥10 pp increase; 23 countries): prevalence −2.1%/yr (95% CI −2.8 to −1.4); number of young smokers −2.7%/yr (95% CI −3.6 to −1.8); quit ratio +0.5%/yr (95% CI 0.4 to 0.6). - Low-tax change (137 countries): prevalence −0.7%/yr (95% CI −1.3 to −0.2); number −1.2%/yr (95% CI −1.4 to −1.0); quit ratio −0.1%/yr (95% CI −0.2 to −0.1). - Ten-year cumulative effects after ratification (vs continuation of pre-FCTC trends; excluding China): - Youth prevalence: −7.5% (95% CI −10.6 to −4.5). - Quit ratio ages 45–59: +1.8% (95% CI 1.2 to 2.3). - By income: largest cumulative declines in LICs (e.g., youth prevalence −36.8% [95% CI −52.2 to −21.5]) and UMICs (−11.0% [95% CI −15.8 to −6.1]); HICs showed a deceleration in the decline in prevalence relative to pre-FCTC trend. - By tax change: high-tax change countries saw larger cumulative improvements (youth prevalence −21.8% [95% CI −29.6 to −13.9]; quit ratio +5.5% [95% CI 4.1 to 7.0]) than low-tax change countries (youth prevalence −7.1% [95% CI −12.2 to −2.1]; quit ratio −1.4% [95% CI −1.8 to −1.0]). - Absolute impacts over the first post-ratification decade (excluding China): - 24.0 million (95% CI 0.4 to 47.6) fewer smokers below age 25, equaling 15.5% of projected smokers without FCTC-accelerated declines. - 2.0 million (95% CI 1.4 to 2.6) additional quitters ages 45–59. - Counterfactual: if all countries had matched high-tax change increases, there would have been 44.0 million (95% CI 10.8 to 54.2) fewer young smokers and 5.2 million (95% CI 4.0 to 6.9) more quitters. - China (analyzed separately) showed paradoxical increases in smoking after FCTC ratification, likely reflecting state-owned industry dynamics and brand switching across diverse price tiers.

The findings indicate that FCTC ratification was associated with a statistically significant acceleration in declines of the number and prevalence of young smokers and an increase in cessation among older adults over the first decade post-ratification. Effects were substantially larger in countries that paired FCTC ratification with sizeable tobacco tax increases, consistent with extensive evidence that large excise tax hikes are the most effective single measure to reduce initiation and increase cessation. While post-FCTC declines in youth prevalence were strongest in LMICs, increases in cessation were more pronounced in HICs, aligning with prior literature. The exception of China underscores the influence of market structure and pricing tiers on policy effectiveness. Complementary non-price measures (e.g., smoke-free laws, warnings) likely reinforced price effects by raising the non-monetary costs of smoking. Overall, aligning intervention timing to each country’s ratification and focusing on age-appropriate outcomes allowed clearer attribution of changes beyond secular trends, supporting the conclusion that the FCTC contributed meaningfully to global reductions in tobacco use and increases in cessation.

This study demonstrates that during the decade following FCTC ratification, countries experienced accelerated declines in youth smoking numbers and prevalence and increased cessation among older adults, avoiding an estimated 24 million young smokers and adding about 2 million quitters (excluding China). Countries with substantial concurrent tax increases achieved at least twice the reductions and gains compared with countries with little or no tax change. The results support the effectiveness of the FCTC and highlight excise taxation as pivotal for future progress. To sustain and amplify gains, countries should ratify and fully implement the FCTC, prioritize large, regular excise tax increases, and continue complementary non-price policies. Future research should include sex-disaggregated analyses, country-level evaluations incorporating covariates such as income and affordability, and assessments of policy packages and enforcement to optimize impact.

- Analyses pooled males and females, precluding sex-specific assessments, particularly relevant in regions with low female smoking prevalence. - Country-level analyses (aside from China) were not performed; grouping countries limits accounting for country-specific covariates (e.g., per capita income, changes in affordability) and may allow populous countries (e.g., Brazil) to influence category results. - Reliance on GBD estimates involves modeled and smoothed data where surveys were sparse, which could affect quasi-experimental estimates; however, sensitivity analyses using countries with ≥10 survey years yielded similar results. - ITSA validity depends on accurately characterizing pre-intervention trends; model selection and sensitivity analyses were used to mitigate this concern. - China was excluded from the main aggregated results due to its unique market; findings may not generalize to such contexts. - Tax change classification (2008–2012) may not fully capture subsequent tax dynamics or other concurrent policies and enforcement differences.