Association between exposure to tobacco control measures and current nicotine use among adolescents in Saudi Arabia: Evidence from the 2022 global youth tobacco survey

Abstract

Objectives

The World Health Organization's six tobacco control strategies (Monitor, Protect, Offer help, Warn, Enforce, and Raise; MPOWER) aim to reduce tobacco use, yet little is known about their impact on individual-level adolescent nicotine use in Middle Eastern countries. This study examined whether exposure to MPOWER policies is associated with current nicotine use among adolescents in Saudi Arabia.

Methods

This cross-sectional analysis used nationally representative data from the 2022 Global Youth Tobacco Survey conducted in schools across Saudi Arabia. The analytic sample included 5092 adolescents aged 13–15 years with complete data on nicotine use and policy exposure. A five-domain MPOWER index (Protect, Offer help, Warn, Enforce, Raise) was constructed from self-reported survey items. The primary outcome was any nicotine product use in the past 30 days. Survey-weighted logistic regression was used to estimate adjusted odds ratios (ORs), controlling for age, sex, school grade, and spending money.

Results

The prevalence of current nicotine use was 11.1 %. MPOWER exposure was significantly lower among users than non-users (2.5 vs. 2.8, p < 0.001). Higher policy exposure was associated with reduced odds of nicotine use (adjusted OR per unit increase = 0.65, 95 % CI: 0.57, 0.73). The association was stronger with the standardised index (adjusted OR = 0.11, 95 % CI: 0.06, 0.21).

Conclusions

Greater individual-level exposure to World Health Organization MPOWER tobacco control measures was inversely associated with adolescent nicotine use in Saudi Arabia. Causal inference cannot be made, and longitudinal studies are warranted.

Highlights

• •This study examines links between tobacco control policy exposure and youth use.
• •A composite policy exposure index was developed from national 2022 survey data.
• •Greater policy exposure was linked to lower odds of current nicotine use.
• •Only 0.2 % of adolescents were exposed to all five policy domains, showing gaps.
• •Findings support stronger youth-targeted tobacco control enforcement in Saudi Arabia.

1. Introduction

Tobacco use is a leading preventable cause of morbidity and premature mortality, with most smokers starting in adolescence (Shubayr et al., 2024; Mutaz et al., 2020; Adebisi et al., 2024a; Adebisi et al., 2024b). Although youth cigarette smoking has declined in many countries, the rise of electronic cigarettes and heated tobacco products has complicated prevention efforts. To accelerate implementation of the Framework Convention on Tobacco Control (FCTC), the World Health Organization (WHO) introduced the MPOWER (Monitor, Protect, Offer help, Warn, Enforce, and Raise) framework in 2008 (De Pinho et al., 2020; Ngo et al., 2017). While designed to reduce traditional tobacco use, MPOWER remains relevant for emerging nicotine products, making youth-focused evaluations critical (Acharya et al., 2024; Ylitörmänen et al., 2023). Yet empirical evidence of its impact on adolescent behaviour is limited, and few studies have applied individual-level survey data to assess associations between MPOWER exposure and youth nicotine use.

Saudi Arabia presents a critical context for evaluating the effectiveness of MPOWER policies. Despite being a signatory to the WHO FCTC, the country continues to experience high levels of tobacco consumption, including emerging forms such as e-cigarettes and waterpipes that are particularly appealing to younger populations (Adebisi et al., 2025a; Monshi et al., 2022a; Monshi et al., 2022b). Recent policy developments, such as public smoking bans, pictorial health warnings, and increases in tobacco taxes, reflect national efforts to align with MPOWER guidelines (Monshi and Ibrahim, 2021). However, these measures have not been uniformly implemented, and enforcement remains variable (Monshi and Ibrahim, 2021). At the same time, population-level data suggest a rising prevalence of nicotine use among Saudi adolescents (Shubayr et al., 2024; Mutaz et al., 2020), raising questions about whether existing tobacco control policies are adequately reaching or influencing this demographic. Understanding the degree to which young people are exposed to MPOWER policies, and whether this exposure correlates with actual behaviour change, is essential for informing policy refinement and prioritisation.

Although global monitoring efforts have tracked country-level adoption of MPOWER policies, far less is known about how these measures translate into adolescents' lived experiences, particularly in low- and middle-income countries. Most evaluations of MPOWER effectiveness rely on ecological or national-level indicators (Watson et al., 2023; Lyle and Hendrie, 2024), which obscure individual variability in policy exposure and behavioural response. Adolescents are not passive recipients of policy; their awareness, attitudes, and environments shape the extent to which such policies exert influence. The Global Youth Tobacco Survey (GYTS), coordinated by the WHO and Centers for Disease Control and Prevention (CDC) (Arrazola et al., 2017; Rachiotis et al., 2020), offers a unique opportunity to bridge this gap by providing harmonised, nationally representative data that capture both tobacco-related behaviours and self-reported exposure to control measures. However, no prior studies in Saudi Arabia have leveraged the GYTS to systematically quantify MPOWER exposure at the individual level or to evaluate its association with current nicotine use across multiple product types. Addressing this gap is particularly urgent given the rise in non-cigarette nicotine consumption and the shifting policy landscape in the Gulf region.

In this study, we used nationally representative data from the 2022 GYTS in Saudi Arabia to quantify adolescents' exposure to five MPOWER policy domains and to examine their association with current nicotine use. Rather than evaluating policy effectiveness, our goal was to describe whether adolescents who reported greater exposure to these domains were less likely to report nicotine use. We constructed a composite MPOWER index using validated survey items and evaluated its psychometric and construct validity. We then tested whether higher reported exposure was associated with lower odds of nicotine use, defined broadly to include combustible cigarettes, waterpipes, e-cigarettes, heated tobacco products, and smokeless tobacco.

2. Method

2.1. Study design and population

This cross-sectional study used data from the 2022 GYTS conducted in Saudi Arabia. The GYTS is a nationally representative, school-based survey designed to monitor tobacco use and exposure to tobacco control policies among adolescents (Arrazola et al., 2017; Rachiotis et al., 2020; Global Youth Tobacco Survey, 2022). It employs a two-stage cluster sampling design to select schools and classrooms, ensuring proportional representation across regions, sex, and grade levels (Global Youth Tobacco Survey, 2022). The survey follows a standardised global protocol developed by the WHO and the U.S. CDC, which allows for comparability across countries and over time. The 2022 Saudi GYTS included a total of 6983 respondents. Of these, 5610 were aged 13 to 15 years, in line with WHO reporting standards. We excluded 518 adolescents in this age range who had missing information on the outcome variable, current nicotine use. The final analytic sample therefore consisted of 5092 adolescents.

2.2. Outcome: any nicotine product use in the last 30 days

The primary outcome was current use of any nicotine product, defined as self-reported use in the past 30 days of cigarettes, waterpipe, electronic cigarettes, heated tobacco products, or smokeless tobacco (e.g., naswar). Respondents answering “yes” to any item were coded as current users; those reporting no use were coded as non-users. Participants with missing data on all items were excluded.

2.3. Exposure: tobacco control measures composite index

To assess adolescents' exposure to tobacco control policies aligned with the World Health Organization's MPOWER framework, we constructed a composite index using data from the 2022 Saudi Arabia GYTS. Because Monitor pertains to national surveillance capacity and is inherently addressed through the GYTS methodology rather than through individual-level exposure, it was not included in the index. We therefore focused on the remaining five pillars, each of which captures policy domains that adolescents can directly experience and report.

Each domain was operationalised using one or more GYTS items, recoded as binary variables (1 = exposed or supportive; 0 = unexposed or not supportive):

• •Protect (P): 1 if the respondent supported smoke-free policies in enclosed or outdoor public places; 0 otherwise.
• •Offer help to quit (O): 1 if a quit attempt was reported for any tobacco product (cigarettes, shisha, e-cigarettes, or smokeless tobacco) in the past 12 months; 0 otherwise.
• •Warn (W): 1 if the respondent had seen anti-tobacco media messages, attended anti-tobacco events, or noticed health warnings on cigarette packaging; 0 if none were reported.
• •Enforce bans on promotion (E): 1 if no exposure was reported to tobacco advertising at points of sale, branded merchandise, brand logos, or free tobacco products; 0 otherwise.
• •Raise taxes (R): 1 if a cigarette pack price was reported in the sample median category or higher (≥20–24 SAR; approximately 5.3–6.4 USD); 0 otherwise. Participants with missing price data were excluded from this domain.

A cumulative MPOWER exposure score was then generated. Two variants were computed:

• 1.Unstandardised score (0–5): Total number of domains to which each respondent was exposed.
• 2.Standardised score (0.0–1.0): The unstandardised score divided by 5, representing the proportion of domains experienced by each adolescent.

The unstandardised score reflects the absolute count of policy domains, while the standardised version enables comparability across subgroups and sensitivity analyses. Using both enhances interpretability and robustness.

Although each domain was operationalised using different self-reported indicators (support, behaviour, or non-exposure), this multidimensional approach aligns with the structure of the MPOWER framework and captures adolescents' awareness, engagement, and environmental exposure to tobacco control policies.

2.4. Covariates

Multivariable models adjusted for a limited set of pre-specified covariates to minimise confounding while avoiding overadjustment. These included age (restricted to 13, 14, or 15 years), sex (male or female), school grade (first, second, or third intermediate), and weekly spending money (categorised as: none, less than 30 SAR [≈ 8 USD], 30–49 SAR [≈ 8–13 USD], 50–99 SAR [≈ 13–27 USD], or 100 SAR or more [≥ 27 USD]). These variables were selected based on their established relevance to adolescent tobacco use in prior studies (Shubayr et al., 2024; Mutaz et al., 2020).

2.5. Statistical analyses

Descriptive statistics were used to summarise sample characteristics and MPOWER exposure by current nicotine use status. Group differences in categorical variables were assessed using chi-squared tests, while comparisons of continuous MPOWER scores were evaluated using independent-samples t-tests. Although MPOWER scores were not strictly normally distributed, t-tests were justified by the Central Limit Theorem given the large sample size.

To assess the psychometric and construct validity of the MPOWER composite index, we conducted multiple analyses. Internal consistency across the five binary components was evaluated using Cronbach's alpha. To examine the dimensionality of the index, we performed principal components analysis (PCA) using the correlation matrix and Kaiser's criterion (eigenvalues >1) to identify the number of relevant components. Factor loadings were reviewed to interpret the underlying dimensions represented by the MPOWER components. The decision to treat the index as an additive composite rather than a latent construct was based on both theoretical considerations and PCA results.

Construct validity was tested by examining bivariate associations between MPOWER scores and theoretically related measures, including attitudes toward smoking harm and current nicotine use. Positive associations with anti-smoking attitudes and negative associations with nicotine use were expected.

To estimate the main associations between MPOWER exposure and current nicotine use, we fitted survey-weighted logistic regression models using Stata (version 18), accounting for complex survey design features—strata, primary sampling units, and final sample weights—declared as svyset PSU [pweight = FinalWgt], strata(Stratum). Both the unstandardised and standardised MPOWER exposure scores were analysed separately. Unadjusted and adjusted odds ratios (ORs) with 95 % confidence intervals (CIs) were reported. Adjusted models included age, sex, school grade, and weekly spending money.

To assess robustness, sensitivity analyses were performed using the standardised MPOWER index, restricting the outcome to cigarette smoking and smokeless tobacco use in the past 30 days—product categories most directly targeted by tobacco control policies in Saudi Arabia.

All statistical tests were two-sided with significance set at p < 0.05. Analyses were conducted using Stata version 18.

2.6. Ethics statement

This study used publicly available, anonymised GYTS data coordinated by the WHO and the U.S. CDC. The survey followed international protocols ensuring voluntary participation, confidentiality, and respondent safety. Because the analysis used secondary anonymised data, no additional ethical approval was required.

3. Results

A total of 5092 adolescents aged 13 to 15 years were included in the analysis. Of these, 565 (11.1 %) reported current use of at least one nicotine product. As shown in Table 1, current use was significantly more prevalent among males than females (60.1 % vs. 39.9 %, p < 0.001). Nicotine use also varied by age, with the highest prevalence among 15-year-olds (36.6 %), compared to 13-year-olds (27.1 %) and 14-year-olds (36.3 %) (p < 0.001). A greater proportion of current users were in the third intermediate grade (36.8 %) compared to non-users (29.3 %, p < 0.001). Weekly spending money was also associated with nicotine use; current users were more likely to report higher discretionary spending, with 14.7 % reporting 100 SAR (26.7 USD) or more per week versus 8.1 % of non-users (p < 0.001). The mean unstandardised MPOWER score was significantly lower among current users than non-users (2.5 vs. 2.8, p < 0.001).

Table 1.

Characteristics of Adolescents Aged 13–15 Years by Current Nicotine Use Status, 2022 Global Youth Tobacco Survey.

Characteristics	Non-user, n (%)	Current User, n (%)	All, n (%)	P-value
Total	4527 (100)	565 (100)	5092 (100)
Sex				P < 0.001
Male	2293 (50.9)	333 (60.1)	2626 (51.9)
Female	2212 (49.1)	221 (39.9)	2433 (48.1)
Age				P < 0.001
13	1587 (35.1)	153 (27.1)	1740 (34.2)
14	1728 (38.2)	205 (36.3)	1933 (38.0)
15	1212 (26.7)	207 (36.6)	1419 (27.9)
Grade				P < 0.001
1st Intermediate	1367 (30.4)	135 (24.5)	1502 (29.8)
2nd Intermediate	1811 (40.3)	214 (38.8)	2025 (40.2)
3rd Intermediate	1314 (29.3)	203 (36.8)	1517 (30.1)
Weekly spending				P < 0.001
None	1041 (23.0)	133 (23.5)	1174 (23.0)
Less than 30 SAR (≈ 8 USD)	1613 (35.6)	154 (27.3)	1767 (34.7)
30–49 SAR (≈ 8–13 USD)	935 (20.7)	96 (17.0)	1031 (20.3)
50–99 SAR (≈ 13–27 USD)	571 (12.6)	99 (17.5)	670 (13.2)
100 SAR or more (≥ 27 USD)	367 (8.1)	83 (14.7)	450 (8.8)
Unstandardised MPOWER score (mean, SD)	2.8 (0.87)	2.5 (0.95)	2.8 (0.89)	P < 0.001

Note: Small discrepancies in totals reflect missing data for specific variables. Missing values were minimal: sex (n = 33), grade (n = 48).

No missing values were observed for age or weekly spending. Percentages are calculated using non-missing denominators. P-values were obtained using chi-squared tests for categorical variables and independent-samples t-tests for the continuous variable (unstandardised MPOWER score).

Exposure to individual MPOWER policy components varied widely among adolescents aged 13–15 years (Table 2). Most participants (88.6 %) supported smoke-free policies in public places, reflecting high exposure to the Protect component. Approximately 70.1 % reported seeing anti-tobacco warnings or media messages, indicating moderate exposure to the Warn domain. Exposure to enforcement of advertising bans was lower, with 60.2 % reporting no contact with tobacco promotion, packaging, or sponsorship (Enforce). Just over half (55.0 %) were classified as exposed to the Raise component, based on reporting a cigarette pack price at or above the sample median. In contrast, only 4.3 % of adolescents had made a quit attempt in the past year.

Table 2.

Exposure to Each Tobacco Control Policy Component Among Adolescents Aged 13–15 Years, Saudi Arabia, 2022 Global Youth Tobacco Survey.

MPOWER Component	Operational Definition	Score = 1 (Exposed)	% Exposed
Protect (P)	Supports smoke-free policies	4510	88.6 %
Offer help (O)	Tried to quit any tobacco in past 12 months	220	4.3 %
Warn (W)	Exposed to warnings/messages	3571	70.1 %
Enforce (E)	No exposure to tobacco promotion	3063	60.2 %
Raise (R)	Reported pack price ≥ median	2798	55.0 %

The distribution of MPOWER composite scores is presented in Table 3. Most adolescents (42.1 %) were exposed to three out of the five policy domains, while only 0.2 % reported full exposure to all five. Approximately 28.8 % were exposed to two domains, and 21.6 % to four. Only 0.6 % of respondents had no exposure to any policy domain. The standardised MPOWER score, expressed as a proportion from 0.0 to 1.0, followed an identical distribution by construction, with scores clustered most frequently at 0.6 (42.1 %) and 0.4 (28.8 %).

Table 3.

Distribution of Composite Tobacco Control Policy Component Scores Among Adolescents Aged 13–15 Years, (n = 5092), 2022 Global Youth Tobacco Survey.

Composite Score	Definition	Frequency (n)	Percent (%)
Unstandardised MPOWER score (0–5)	Count of MPOWER policy domains exposed to
0	No exposure to any domain	29	0.6
1	Exposure to 1 policy domain	343	6.7
2	Exposure to 2 domains	1465	28.8
3	Exposure to 3 domains	2142	42.1
4	Exposure to 4 domains	1102	21.6
5	Full exposure to all 5 domains	11	0.2
Standardised MPOWER Score (0.0–1.0)	Proportion of 5 domains exposed to
0.0	No exposure	29	0.6
0.2	Exposure to 1/5 domains	343	6.7
0.4	Exposure to 2/5 domains	1465	28.8
0.6	Exposure to 3/5 domains	2142	42.1
0.8	Exposure to 4/5 domains	1102	21.6
1.0	Full exposure (all 5)	11	0.2

Note: The standardised MPOWER score is derived by dividing the unstandardised score by 5. The distributions are mathematically equivalent but presented in both formats for interpretability and modelling flexibility.

Fig. 1 displays results from survey-weighted logistic regression models examining the association between exposure to tobacco control measures and current nicotine use among adolescents in Saudi Arabia. In unadjusted models, higher unstandardised MPOWER scores (range: 0–5) were associated with lower odds of nicotine use (OR: 0.63, 95 % CI: 0.56, 0.71). This association remained robust after adjusting for age, sex, school grade, and weekly spending money (adjusted OR: 0.65, 95 % CI: 0.57, 0.73). Results were consistent when using the standardised MPOWER score (range: 0.0–1.0), which represents the proportion of policy domains to which each respondent was exposed. Adolescents with higher standardised scores had lower odds of nicotine use (adjusted OR: 0.11, 95 % CI: 0.06, 0.21). For interpretability, this corresponds to an adjusted OR of 0.80 (95 % CI: 0.75, 0.86) per 0.1 (10 %) increase in the standardised index, indicating that incremental increases in policy exposure were associated with reduced odds of nicotine use. All models accounted for the complex survey design, including clustering, stratification, and sample weights.

Fig. 1.

Association Between Exposure to Tobacco Control Policy Measures and Current Nicotine Use Among Adolescents, Saudi Arabia, 2022 Global Youth Tobacco Survey.

Odds ratios (ORs) and 95 % confidence intervals (CIs) are estimated from survey-weighted logistic regression models.

Unstandardised MPOWER score (0–5) represents the unstandardized additive exposure score (range: 0–5).

Standardised MPOWER score (0.0–1.0) is the standardised average score (range: 0.0–1.0).

Unadjusted models include only the MPOWER variable.

Adjusted models control for age, sex, school grade, and weekly spending money.

All models account for the complex sampling design using strata, primary sampling units, and final survey weights.

Validation of the Tobacco Control Policy Composite Index.

We assessed the psychometric and construct validity of the MPOWER composite index using multiple approaches. Internal consistency across the five binary policy components was modest (Cronbach's α = 0.36). This level of reliability was anticipated, as the MPOWER framework is inherently multidimensional, encompassing distinct but complementary policy domains rather than a single unidimensional construct. Principal components analysis supported this interpretation, with multiple components contributing to the variance. Consistent with established approaches in MPOWER monitoring, we therefore treated the composite score as a simple additive index, designed to capture the cumulative breadth of adolescents' exposure across diverse policy domains. The first two components had eigenvalues greater than one and together accounted for 50.2 % of the total variance. The first component (29.2 %) was defined by strong positive loadings on Enforce, Raise, and Protect, and moderate negative loadings on Offer help and Warn. This component appeared to reflect structural policy exposure. The second component (20.9 %) was characterized by strong loadings on Protect and Warn, suggesting an informational or awareness-oriented dimension.

Construct validity was examined by assessing associations between MPOWER scores and theoretically related constructs. Higher MPOWER scores were positively associated with the belief that smoking is harmful to others (p < 0.001) and negatively associated with current nicotine use (0.65; 95 % CI: 0.57, 0.73; p < 0.001) as stated earlier. These patterns align with expectations and reinforce the index's interpretability as a policy exposure metric. Together, these results support the use of the MPOWER composite index as a multidimensional measure of exposure to tobacco control policies among adolescents.

3.1. Sensitivity analysis

When restricting the outcome to cigarette smoking in the past 30 days (3.2 % prevalence), results were consistent with the main analysis. Higher exposure to the fixed standardised MPOWER index was associated with markedly lower odds of cigarette smoking (aOR = 0.14, 95 % CI: 0.04, 0.44). A similar pattern was observed for smokeless tobacco use (3.7 % prevalence), where higher MPOWER exposure was associated with substantially reduced odds of use (aOR = 0.14, 95 % CI: 0.06, 0.33).

4. Discussion

This study provides robust, nationally representative evidence that greater individual-level exposure to WHO MPOWER policy domains is significantly associated with lower odds of current nicotine use among adolescents in Saudi Arabia. The inverse association observed across both MPOWER scores persisted after adjustment for age, sex, school grade, and spending money. Notably, adolescents exposed to a greater number of policy measures were substantially less likely to report use of any nicotine product, highlighting the potential of comprehensive tobacco control approaches to address the full spectrum of adolescent nicotine use. That only 0.2 % of adolescents reported exposure to all five policy domains underscores the fragmented reach of tobacco control efforts and points to clear opportunities for policy strengthening. Furthermore, the apparently stronger effect observed for the standardised index compared with the unstandardised count score reflects differences in scaling rather than substantive inconsistency. Expressing the effect per 10 % increase in the standardised index provides a more interpretable estimate, indicating that each incremental 10 % rise in policy exposure was associated with roughly a 20 % reduction in the odds of nicotine use. This suggests that even small increases in exposure to MPOWER domains may translate into meaningful behavioural differences among adolescents.

This study adds methodological and empirical value by operationalising MPOWER exposure as an individual-level composite index using GYTS data. Although MPOWER is widely applied for national benchmarking (Arrazola et al., 2017; Rachiotis et al., 2020), it has not previously been adapted for adolescent-level analysis. Our validated index captured both structural (e.g., taxation, advertising bans) and informational (e.g., warnings, media messages) policy dimensions. Exposure was skewed toward moderate coverage, with most adolescents reporting two or three domains, indicating uneven implementation. Very low exposure to cessation support (4.3 %) highlights a neglected intervention area. Using perceived cigarette price as a proxy for taxation offered a behaviourally relevant measure beyond official tax rates.

By situating this analysis in Saudi Arabia, the study makes a unique contribution to global tobacco control research. The Gulf region is undergoing a rapid epidemiological transition, with increasing adolescent uptake of novel nicotine products amid shifting regulatory landscapes (Monshi et al., 2024). Despite Saudi Arabia's commitment to the WHO FCTC (Adebisi et al., 2025a; Monshi et al., 2022a), the pace and consistency of policy enforcement remain uneven. Our findings provide empirical support for the behavioural relevance of MPOWER policies among adolescents in this setting, challenging the assumption that global frameworks require adaptation to be effective in non-Western contexts (Monshi et al., 2022b). At the same time, the low overall policy exposure among Saudi youth raises questions about the fidelity of implementation and the need for better integration of tobacco control within school and community systems. This research, by assessing associations between policy exposure to individual behaviour using a validated instrument, offers a model for other countries seeking to evaluate and refine their tobacco control strategies.

These findings have clear implications for reducing adolescent nicotine use in Saudi Arabia. Comprehensive implementation is critical, as partial exposure to MPOWER domains may be insufficient. Policymakers should expand coverage across all five domains, especially cessation support and taxation. The inverse association observed even with small exposure gains suggests that incremental scaling can still be effective. Youth-focused health communication, particularly in schools, should be prioritised given strong responsiveness to warnings. Finally, routine monitoring of policy exposure through surveys such as the GYTS, with MPOWER metrics embedded in adolescent health surveillance, could strengthen accountability and policy effectiveness.

This study has several notable strengths. It is one of the first to operationalise and validate an individual-level MPOWER exposure index using nationally representative data, allowing for fine-grained analysis of the relationship between policy engagement and behaviour. However, the study also has limitations. Its cross-sectional design precludes causal inference, and the observed associations may reflect unmeasured confounding or reverse causation (Adebisi et al., 2025b; Polosa et al., 2025); for example, adolescents who use nicotine may be less receptive to, or less exposed to, policy messaging. We could not adjust for other important determinants of nicotine use, such as parental and peer smoking, school setting (urban vs. rural), or regional variability in enforcement, raising the possibility of residual confounding. Self-reported measures are subject to recall and social desirability biases, particularly for sensitive behaviours like tobacco use. The proxy measure for taxation (self-reported price) may not fully capture adolescents' economic access to nicotine products, especially in cases of informal or subsidised supply. In addition, all MPOWER components relied on adolescent self-report, and the domains reflected heterogeneous constructs (e.g., support for policies, quit attempts, or non-exposure to promotion), which may introduce misclassification bias. Earlier GYTS waves in Saudi Arabia did not consistently include MPOWER-relevant items, preventing longitudinal assessment of changes in exposure and behaviour over time.

5. Conclusion

This study provides evidence that greater individual-level exposure to WHO MPOWER policy measures is associated with significantly lower odds of nicotine use among adolescents in Saudi Arabia. By constructing and validating a multidimensional exposure index using nationally representative GYTS data, we demonstrate that even partial exposure to key tobacco control domains, such as advertising bans, smoke-free environments, and health warnings, is associated with lower odds of use. As Saudi Arabia and other countries in the region seek to reduce youth nicotine use, this study underscores the value of measuring and maximising policy exposure as a central component of effective tobacco control strategies.

CRediT authorship contribution statement

Najim Z. Alshahrani: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Software, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization.

Ethical approval

Ethical approval was not required for this study, as it involved the secondary analysis of publicly available, de-identified data from the 2022 Global Youth Tobacco Survey, which was conducted with standard ethical oversight by the World Health Organization and the Ministry of Health in Saudi Arabia.

Funding statement

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

This study used secondary data from the 2022 GYTS in Saudi Arabia, publicly available via the WHO NCD Microdata Repository: https://extranet.who.int/ncdsmicrodata/index.php/catalog/966