Article Text

Smoking trends in Mexico, 2002–2016: before and after the ratification of the WHO’s Framework Convention on Tobacco Control
  1. Luis Zavala-Arciniega1,
  2. Luz Myriam Reynales-Shigematsu1,2,
  3. David T Levy3,
  4. Yan Kwan Lau4,
  5. Rafael Meza4,
  6. Daniela Sarahí Gutiérrez-Torres1,
  7. Edna Arillo-Santillán1,
  8. Nancy L Fleischer4,
  9. James Thrasher5
  1. 1 Departamento de Investigación sobre Tabaco, Centro de Investigación en Salud Poblacional, Instituto Nacional de Salud Pública, Cuernavaca, Mexico
  2. 2 Tobacco Research Department, Instituto Nacional de Salud Pública, Cuernavaca, Mexico
  3. 3 Oncology, Lombardi Comprehensive Cancer Center, Georgetown University, Washington, District of Columbia, USA
  4. 4 Epidemiology, University of Michigan, Ann Arbor, Michigan, USA
  5. 5 Health Promotion, Education, & Behavior, School of Public Health, University of South Carolina, Columbia, South Carolina, USA
  1. Correspondence to Dr Luz Myriam Reynales-Shigematsu, Tobacco Research Department, Instituto Nacional de Salud Pública, Cuernavaca 62100, Mexico; cisp11{at}


Introduction Mexico was the first Latin American country to ratify the Framework Convention on Tobacco Control (FCTC) in 2004, after which it implemented some key FCTC policies (e.g., taxes, smoke-free, pictorial warnings and ad bans). This study assessed trends in the prevalence of current, daily and non-daily smoking in Mexico before and after the implementation of key FCTC policies.

Methods Data were analysed from two comparable, nationally representative surveys (i.e., the National Survey on Addictions 2002, 2011 and 2016, and the Global Adult Tobacco Survey 2009 and 2015). The pooled sample comprised 100 302 persons aged 15–65 years. Changes in the prevalence of current, daily and non-daily smoking were assessed.

Results From 2002 to 2016, the prevalence of current smoking fell 11% in relative terms (from 21.5% to 19.0%). The decrease was registered between 2002 and 2009, and after that, a slight increase was observed (from 16.5% in 2009 to 19% in 2016). The prevalence of daily smoking decreased by about 50% between 2002 and 2016 (from 13.5% to 7.0%) with most of the decrease occurring by 2009. Conversely, the prevalence of non-daily smoking increased by 35% between 2009 and 2016 (from 8.8% to 11.9%).

Conclusions Full implementation of the FCTC is necessary to further reduce smoking. Specific interventions may be needed to target non-daily smokers, who now comprise more than half of current smokers in Mexico.

  • Low/Middle-income country
  • public policy
  • surveillance and monitoring

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


In 2004, Mexico became the first country in Latin America to ratify the WHO’s Framework Convention on Tobacco Control (FCTC).1 After signing the FCTC in 2004, the tobacco excise tax increased from 58.9% of the final price in 2007 to 68.8% in 2011,2–5 leading to an increase in the real price per pack of 20 cigarettes of 44% ($MX29.9 in 2007 to $MX 42.7 in 2011), after which taxes and prices have remained stable.4 6 7 The 2008 General Law on Tobacco Control (GLTC) included other key FCTC-recommended policies.8 9 Cigarette advertising was banned through all channels except point of sale, adult-only venues (e.g., bars) and adult-oriented magazines.8 Since 2010, pictorial health warnings were required on at least 30% of the front of cigarette packages, with textual warnings on 100% of the back and one side of the pack.10 The GLTC only allows smoking in restrictive-designated smoking areas (e.g., with extensive exhaust systems and physically separated from paths traversed by non-smokers).8 11 However, the GLTC was passed the same day as the comprehensive smoke-free law in Mexico City,12 which caused some confusion around implementation in Mexico City and the rest of the country. Concerns about pre-emption remained until the Supreme Court ruled in favour of the Mexico City legislation in September 2009.12 This likely contributed to the slow, incomplete adoption and implementation of smoke-free regulations in other states.13–16 Since 2011, no other national-level policies have been adopted.

Meanwhile, the tobacco industry has continued to lobby the legislature, primarily to avoid further tobacco tax increases.16 At the same time, the industry introduced innovative cigarettes with flavour capsules in the filter, whose market share has rapidly increased, as in many other Latin American countries.17–20 Furthermore, e-cigarette marketing and sales in Mexico are banned, similar to Uruguay and Brazil,21 although consumption is relatively high among adult smokers (4.7%) and youth (i.e., 12% of eighth graders).22–24 Mexico is a key country in Latin American because of its size and economic importance in the region. Therefore, the case study of Mexico’s FCTC experience in the context of the rapidly changing landscape of nicotine products may help to inform policy development in other settings, particularly other low-income and middle-income countries.

Longitudinal studies of Mexican smokers indicate the effectiveness of FCTC policies,7 25–28 yet the population-level effects of these policies on smoking prevalence remain understudied. This study assesses changes in the prevalence of current, daily and non-daily smoking in Mexico from 2002 to 2016 using nationally representative surveys over the implementation period for key FCTC policies.



Data come from two comparable, nationally representative cross-sectional surveys: 2002, 2011 and 2016 administrations of the National Survey on Addictions (NSA), and 2009 and 2015 administrations of the Global Adult Tobacco Survey (GATS). These household surveys used multistage sampling schemes to provide national-level estimates of tobacco use. Methodological details of each survey have been published previously.29–33 Pooling data across surveys resulted in a sample of 100 302 aged 15–65 years (NSA 2002 n=10 418; GATS 2009 n=12 294; NSA 2011 n=14 298; GATS 2015 n=12 777 and NSA 2016 n=50 515).


Current, daily and non-daily smoking prevalence definition

Current smokers reported smoking daily (daily smokers) or less often (non-daily smokers). Questions were slightly different across the surveys: Do you currently smoke tobacco on a daily basis, less than daily, or not at all? (GATS 2009, GATS 2015 and NSA 2016); How often do you currently smoke? (NSA 2011) and In the past 30 days, have you smoked tobacco? (NSA 2002), with those who indicated yes being asked Approximately, how many cigarettes were smoked daily in the past 30 days? (I don’t smoke daily; 1–5; 6–10; 11–20 and 20+ cigarettes). Responses to each question were recoded to indicate daily or non-daily smoking frequency. GATS did not ask whether participants had smoked 100 lifetime cigarettes, so this criterion was not used to define smoking status.

Sociodemographic variables

Sex (female/male), age (15–17, 18–24, 25–34, 35–44, 45–54 and 55–65 years), highest educational attainment (primary or less, secondary, high school and university or more) and place of residence (urban/rural) were all assessed.


Prevalence and 95% CIs were estimated for current, daily and non-daily smoking, adjusting for the sampling designs. Two-sample independent t-tests were used to determine differences in these estimates across years for key pre-milestones/post-milestones in policy implementation (2002 vs 2009; 2009 vs 2016 and 2002 vs 2016), both overall and within sociodemographic strata. Trends in prevalence are measured relative to the initial level. Statistical analyses were conducted using Stata V.15.


Overall, the prevalence of current smoking decreased in relative terms by 11% (from 21.4% to 19.0%) (see table 1) between 2002 and 2016, about 13% in men (from 33.9% to 29.5%) and 19% in women (from 11.5% to 9.3%). Over this period, reductions were statistically significant for higher education groups (i.e., university or more=29.6% to 20.7% and high school=26.6% to 21.2%) and among those aged 35–44 (25.4% to 18.8%) and 45–54 (23.1% to 18.1%) years. Overall, the relative decrease was limited to the period from 2002 to 2009 (−22.9%), after which prevalence increased in relative terms by 11.5% (2009=16.5% and 2016=19%). From 2002 to 2016, daily smoking decreased by approximately 50% (2002=13.5% and 2016=7.0%). The relative change in the percentage of daily smokers was generally consistent across sex, age, level of education and residence. By contrast, from 2009 to 2016, the prevalence of non-daily smoking increased by about a third in both men (2009=13.8% and 2016=18.4%) and women (2002=4.4% and 2016=6.0%). Increases were generally consistent across sociodemographic groups.

Table 1

Prevalence and changes in daily and non-daily smoking among Mexicans, 2002−2016


Our results indicate that the prevalence of current and daily smoking in Mexico decreased from 2002 to 2016, which covers the period from before to after FCTC ratification and policy implementation. However, most of the declines in the prevalence of current (2002=21.4%; 2009=16.5% and 2016=19.0%) and daily (2002=13.5%; 2009=7.7% and 2016=7.0%) smoking had occurred by 2009. This decrease was associated with a series of tobacco tax increases between 2000 and 2009 (i.e., the ad-valorem excise tax increased from 100% of the final retail price in 2000 to 160% in 2009).4 There was one further increase in 2011 due to the application of the specific tax of $MX 0.35 per cigarette, but our results indicate no meaningful change in daily smoking after 2009 and only a slight decrease in the average of cigarettes smoked per day (CPD) among daily smokers (9.4 CPD in 2009 to 7.7 CPD in 2015).32 By contrast, the prevalence of non-daily smoking increased from 2002 to 2016 (2002=7.9%; 2009=8.8% and 2016=11.9%). Historically, Mexican smokers have a relatively light smoking pattern, with a high proportion of non-daily smokers.34 35 Our results indicate that this pattern has increased over time, offsetting the declines in daily smoking that were found before 2009. It is important to highlight that non-daily smoking is still dangerous: low cigarette consumption is associated with higher all-cause and cause-specific mortality compared with not smoking.36 37

Tobacco control policy implementation in other jurisdictions has been accompanied by downshifts in smoking frequency,38–42 and policies likely account for the similar downshift found in Mexico. In other jurisdictions, however, these shifts have been accompanied by an overall decline in smoking prevalence, which, in Mexico, appears relatively stable since 2009.

Several reasons may explain the stability in the prevalence of current smoking in Mexico, despite the downshift in smoking frequency in recent years. Tobacco industry interference has disrupted the implementation of smoke-free policies.16 In particular, the GLTC allows smoking in designated areas, which may reduce compliance even in the 11 states that have adopted comprehensive smoke-free laws.12 Furthermore, the industry’s 2011 introduction and subsequent rapid growth of cigarettes with flavour capsules17 19 appear to appeal to and add perceived value (e.g., smoother, more stylish and lesser perceived harm) to smoking for smokers of all ages.17–19 While this innovation may have added value to smoking that helped to offset the cost of the last tax increase in 2011, further periodic increases in taxes may be necessary, especially given the evidence for the effectiveness of earlier tax increases. Also, the relatively small health warning on the front of the pack (30%) should be increased to align with FCTC guidelines.28 43 44

E-cigarettes could also explain partly the flattening of the trends. Although e-cigarettes are banned, the prevalence of e-cigarette use is not-negligible among some subgroups of current smokers (adolescents (9%), young adults (9%), female smokers (6%) and smokers with higher socioeconomic status (7%)).22–24 As in other countries, the strongest correlate of e-cigarette use is being a current smoker (4.7% current smokers vs 0.4% non-smokers).22 Nevertheless, it is not clear whether e-cigarette use has kept some people from quitting cigarettes,45 since e-cigarette use is relatively recent and is relatively low among current smokers (4.7%) and former smokers (1.1%). Indeed, the introduction of e-cigarettes has been associated with a decline in youth and young adult smoking prevalence in other countries where e-cigarette use is more prevalent.46 To evaluate how e-cigarettes influence smoking patterns in Mexico, further research is needed, particularly on how lighter smokers use e-cigarettes. Indeed, psychological dependence47 and potential genetic factors may help to explain low levels of consumption among populations of Mexican heritage,48 whose ancestors used tobacco for centuries before European colonisation of the Americas. Our study has several limitations. The surveys we analysed had different focus, with GATS being an international survey that only asks about tobacco, whereas the NSA gathers information not only about tobacco, but also about drug and alcohol use, and violence. Nevertheless, the NSA survey asks about tobacco before other topics. Furthermore, both surveys are nationally representative and used similar sampling designs, helping to minimise issues around their comparability. Similar conclusions are reached if we had just compared estimates from the NSA alone (i.e., 2002, 2011 and 2016). Another potential limitation was our inability to analyse results by income level because income was not available in all surveys. Instead, we used education as a proxy for socioeconomic status, which is likely to be correlated with income. The differential changes we found in daily smoking over time by education (i.e., greater decreases with higher educational attainment, although the equivalent prevalence in recent years) are, thus, likely to also occur across income groups; however, further confirmation is warranted.32 49


To accelerate tobacco control, it is necessary to expand current policies and enhance compliance with existing tobacco control policies and laws in Mexico. Full implementation of FCTC policies should give priority to continued tax increases, comprehensive smoke-free laws, increasing the size of health warnings, enhancing access to cessation therapies for non-daily and daily smokers, and banning all tobacco advertising and promotion. Further strategies are also needed to address the complexity of the rapidly changing landscape of nicotine products, including prohibition of flavours, especially flavour capsules.

What this paper adds

  • This study evaluates the changes in smoking prevalence in Mexico before and after the implementation of key policies from the Framework Convention on Tobacco Control (FCTC).

  • Daily smoking as well as overall current smoking prevalence decreased between 2002 and 2016. However, most of the decline had occurred between 2002 and 2009. In contrast, non-daily smoking increased between 2009 and 2016.

  • More than half of current smokers in Mexico are now non-daily smokers. Interventions may need to be tailored to target this group. Trends observed in Mexico may be relevant to other low middle-income countries.

Abstract translation

This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Contributors JT, LZ-A and LMR-S conceived the research. LZ-A wrote the draft. YKL, DSG-T and LZ-A conducted the statistical analyses. DTL, RM, NLF, EA-S, JT, LMR-S, DSG-T and YKL provided a critical revision to the manuscript. All the authors approved the final version of this manuscript.

  • Funding Research reported in this study was supported by of the Fogarty International Center of the National Institutes of Health under award number R01 TW010652. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Provenance and peer review Not commissioned; externally peer reviewed.