Structured evaluation of gender integration in tobacco control research

Abstract

Background

Gender norms are known to influence smoking behaviours, but studies evaluating tobacco control policies frequently overlook these factors by relying on gender-blind methodologies. The degree to which gender-sensitive methodological approaches are used in tobacco control research since the adoption of the Framework Convention on Tobacco Control has not yet been evaluated.

Methods

We applied the European Institute for Gender Equality’s Gender Impact Assessment (EIGE-GIA) toolkit to 43 peer-reviewed studies to assess the integration of gender-sensitive approaches in tobacco control research. Original tobacco control research studies published after 2005 based on nationally representative data were identified from PubMed using targeted search terms and a reverse snowball strategy. Each study was coded against the EIGE-GIA’s core criteria of specifying the gendered target group and assessing the gendered impact of policy interventions.

Results

Among the 43 studies analysed, 40 identified specific target groups and outlined key challenges related to tobacco use and MPOWER policies, meeting the first EIGE-GIA criterion. However, only 16 studies assessed specific tobacco control policies, and only 5 evaluated the gender-specific impacts of these policies. Many studies failed to meet the second criterion, often relying on binary comparisons that ignore complex gender dynamics.

Conclusions

Our findings reveal persistent gaps in tobacco control research methodologies. Reliance on binary sex-disaggregated data that fail to explore gender-based determinants of health limits our understanding of the effectiveness of tobacco control policy interventions and fails to address gendered smoking behaviours. Researchers should use comprehensive frameworks to guide the assessment of gendered impacts of tobacco control interventions.

WHAT IS ALREADY KNOWN ON THIS SUBJECT

• Tobacco control policies are frequently designed without meaningful consideration of gender, and research evaluating these policies rarely goes beyond the presentation of simple sex-disaggregated data. The extent to which tobacco control impact evaluations account for gendered factors that shape smoking behaviours remains unknown.

WHAT IMPORTANT GAPS IN KNOWLEDGE EXIST ON THIS TOPIC

• Our study identifies significant gaps in the use of gender-sensitive approaches in tobacco control research. Only 5 out of the 43 studies evaluating tobacco control policies included in this review employed comprehensive gender-sensitive approaches, as recommended by the European Institute for Gender Equality’s Gender Impact Assessment toolkit.

• These findings suggest that contemporary research is not conducted in a way that would capture gendered factors influencing smoking behaviour, likely underestimating the significant impact of these social determinants of tobacco use.

WHAT THIS STUDY ADDS

• To improve tobacco control interventions, policy-makers and researchers should systematically embed gender considerations into policy evaluations and recommendations. Without these insights, tobacco control interventions risk overlooking the complex ways gender norms, roles and social influences shape smoking behaviours.

Introduction

Since the adoption of the WHO Framework Convention on Tobacco Control (WHO FCTC) in 2003, some Parties to the Convention and researchers have advocated for better research on the ways gender impacts tobacco control policies around the world.^1–4 Yet, health research informing the design of these policy interventions remains mostly gender-blind despite widespread knowledge of the gendered dimensions of tobacco use and related health harms.^{4 5}

Gender, a social construct often confused with sex, can be defined by the norms, roles, and expectations shaped by the distribution of power and resources in society.^{6 7} Sex, operationalised in this study as a category based on a set of biological attributes assigned at birth, interacts with gender to influence health behaviours and responses to interventions.^8–10 Academic research often relies on binary sex-disaggregated approaches, which fail to account for the complex set of sex-related factors that underlie sex categories and a spectrum of gender norms that intersect with the tobacco industry’s gendered business practices.^{4 10 11} In this study, we use gender as a comprehensive term for socially constructed identities relevant to tobacco control, including binary categories (ie, men and women), commonly used in survey-based data.

The role of gender as a determinant of tobacco use is known to vary significantly across time and country contexts. For example, while men generally smoke at higher rates worldwide, countries such as Kuwait, Timor-Leste and South Korea, characterised by decreasing smoking rates among men between 1990 and 2015, saw these rates increase among women over the same period.¹²

In the European region, while smoking prevalence among men decreased by 13.6% between 2000 and 2020, women only experienced a 4.9% decrease, narrowing the smoking gender gap.⁵ These shifts highlight the need for global tobacco control research to adopt approaches that are responsive to the nuanced ways in which gender norms affect tobacco use. To do so, we must first understand the degree to which gender is currently integrated in contemporary health research.

Methods

This study assesses the extent to which gender-sensitive approaches have been used in tobacco control research since the FCTC’s adoption using the European Institute for Gender Equality’s Gender Impact Assessment (EIGE-GIA) toolkit,¹³ which was designed to support the implementation of gender-responsive policies. We first searched PubMed using search terms ‘gender’ AND ‘framework convention on tobacco control’ to purposively sample studies addressing gender in some way, which resulted in the identification of 72 studies. These were supplemented by 25 additional studies identified through a reverse snowball sampling strategy of articles citing the WHO’s 2018 Gender-responsive tobacco control: evidence and options for policies and programmes.¹⁴ Although the WHO has previously discussed the importance of gender in tobacco control,^{15 16} this 2018 report offers the most recent guidance to the Conference of the Parties for integrating gender in tobacco control policies.

Inclusion criteria required studies to be peer-reviewed original research on tobacco smoking using nationally representative data published after 2005 in English, resulting in 39 population-level studies and four review studies (43 studies total) for qualitative coding (online supplemental table S1). Each study was assessed against the EIGE-GIA toolkit’s two core gender-sensitive analysis criteria, which are (1) to specify the current situation of the target group and (2) to assess the impact of a policy intervention on this group. We further divided these core criteria into five subcriteria that could be applied to tobacco control evaluations and added two subcriteria to account for studies that did not specify an intervention but still included a gendered analysis, as illustrated in figure 1.

Figure 1. Process flow chart showing the application of the EIGE Gender Impact Assessment (GIA) toolkit criteria. EIGE, European Institute for Gender Equality.

Summary of findings

Of the 43 studies included for analysis, 40 met core criteria 1 of clearly specifying a target group and detailing the tobacco-relevant issue of interest. Of these, 31 focused on adults, 8 on youth and 1 on all ages. Most studies (35 of 40) disaggregated their target group by sex. For instance, one study by Gredner et al¹⁷ examined the impact of tobacco control policies on lung cancer rates for men and women aged 35 and over in Europe, clearly identifying the target group to assess the differential impact of the intervention.¹³

Although most studies identified a target group, only 16 of the 40 studies assessed a specific tobacco control intervention. One example is a study by Winkler et al that explored the impact of MPOWER measures in reducing smoking prevalence in 13 West African countries, noting a significant effect from smoking cessation programmes in men and none in women.¹⁸ Although the remaining studies (24 out of 40) did not evaluate the impact of a specific tobacco control intervention, many still discussed the implications of their findings for tobacco control policies. For example, Adia et al’s study of demographic factors associated with smoking prevalence among men and women in Samoa did not formally assess a tobacco control intervention, but the authors identified higher odds of smoking among unmarried men working as labourers compared with their married counterparts, and this finding, alongside others, was used to recommend tailored tobacco control interventions.¹⁹

Regardless of whether a specific tobacco control intervention was evaluated, most studies (22 of 40) failed to assess the gender-sensitive implications of their findings. One study evaluating the effects of global MPOWER implementation on adult smoking rates found a significant association between MPOWER composite scores and decreasing smoking prevalence only for men, but did not explore what may have contributed to their failure to produce impacts among women.²⁰ In total, 12 of the 23 studies that did not evaluate a specific tobacco control policy put forward gender-sensitive implications of their studies, while only 6 of the 15 studies that evaluated the gendered impacts of an intervention or policy discussed the gender-sensitive implications of their findings.

Discussion

As guided by the EIGE-GIA toolkit, we identify recurring gaps in gender-sensitive interpretations of health research findings to inform the design of tobacco control policies and interventions.¹³ Our assessment reveals that even among health research evaluating tobacco control interventions that consider gender, most studies (25 out of 43) fail to go beyond the use of basic sex-disaggregated data. While analysing sex-disaggregated data in tobacco control evaluations provides important initial insights, it falls short of capturing the nuanced ways gendered dynamics influence tobacco consumption.

Failure to integrate gender-sensitive approaches in the study of tobacco control not only limits the effectiveness of interventions but also risks reinforcing harmful gender norms.⁴ For example, despite advances in implementing tobacco control measures across Europe, progress has not been uniform for men and women.⁵ In Poland, the creation of the Polish Quitline cessation clinics has contributed to a decline in smoking prevalence among men, but smoking rates among women have plateaued, largely due to targeted marketing strategies by tobacco companies that are proving difficult to counter.²¹ Similarly, in Pakistan and Bangladesh, many women face barriers to accessing smoking cessation supports, including a lack of awareness of existing services and hesitancy to seek professional help to quit, while men are often pressured to smoke due to cultural norms and structural sexism.^22–24 Research that elucidates the determinants of smoking behaviours for men and women and the ways these are influenced by structural sexism is therefore crucial to informing gender-sensitive tobacco control interventions that effectively address these disparities.

Despite the urgent need for gender-sensitive tobacco control research, several of the studies we reviewed failed to meaningfully engage with gender, with three studies choosing to control for gender.^25–27 Such techniques reflect a perspective that gender should be treated as a confounding variable rather than a critical determinant that shapes health outcomes. Reducing gender to a control variable obscures the reality that people of different genders experience health issues and are impacted by health interventions differently.^{28 29}

Thus, gender-sensitive research should go beyond basic comparisons between men and women to explore complex contextual factors, including cultural norms, affordability and regulatory environments, that uniquely shape smoking behaviours. For example, cigarette affordability often affects smoking decisions differently for men and women, with women who smoke in low-income settings being more responsive to changes in cigarette prices than men, who often maintain their purchasing habits despite price increases.^{30 31} These differences highlight the need for tobacco control policies to be informed by robust, gender-responsive evidence, which can be supported by frameworks such as the EIGE-GIA toolkit and the WHO gender-responsive assessment scale.^{13 32} While we recognise the limitations in accessing high-quality gender-related tobacco data, we join calls for a shift in the way sex and gender are conceptualised and operationalised in health research, including through qualitative and mixed-methods studies, which were under-represented as compared with quantitative studies.¹⁰ Integrating gender as a dynamic system of power and social positioning rather than as a binary sex category is essential to advancing equity in tobacco control research and policymaking. Moreover, this integration must be rooted in an understanding of how gender-based oppression shapes tobacco use patterns and reinforces inequitable health outcomes.

Conclusions

Gender has not been sufficiently integrated into tobacco control research since the adoption of the WHO FCTC. Using the EIGE-GIA toolkit, we find that most health research fails to incorporate a gender-sensitive approach to translate the implications of their findings for tobacco control interventions and policies. These gender-blind approaches risk reinforcing harmful gender norms and perpetuating gender inequities in health. Sex-disaggregated approaches alone do not fully capture the role gender plays in shaping smoking behaviours. Understanding these persistent disparities requires research that fully integrates a gender lens in exploring the broader social, economic and cultural contexts to produce actionable recommendations for tobacco control interventions. Employing a gender-responsive approach benefits everyone by promoting equitable and effective tobacco control interventions.

Data availability statement

The full dataset produced by this study can be found in the supplementary appendix.