Study design and data source

This study is a secondary analysis of data from three cross-sectional rounds of the Global Youth Tobacco Survey (GYTS) conducted in Belarus in 2004, 2015, and 2021. The GYTS is a school‑based survey of students aged 13–15 years, employing standardized methodology developed by WHO and the US Centers for Disease Control and Prevention (CDC) to ensure international comparability^5,6. The GYTS protocol was approved by the WHO Ethics Review Committee and by relevant institutional review boards in Belarus. Parental/guardian consent and student assent were obtained prior to participation.

Sampling

A two‑stage cluster sample design was used in each survey round. In the first stage, schools were selected with probability proportional to enrollment size. In the second stage, classes within selected schools were randomly chosen, and all students in those classes were eligible to participate. The total sample across three rounds included 12049 students (2004, n=4012; 2015, n=4018; 2021, n=4019). All participants were school‑attending adolescents aged 13–15 years.

Variables and data collection

Data were collected using a self‑administered, anonymous questionnaire; demographic variables, including sex (male, female) and school grade (7th, 8th, 9th), were collected. Current cigarette smoking was defined as smoking at least one cigarette in the past 30 days; ever smoking was defined as having ever tried or experimented with cigarette smoking, even one or two puffs; frequent smoking was defined as smoking >20 cigarettes in the past 30 days. With regard to electronic cigarette use, current e‑cigarette use was defined as using an electronic cigarette at least once in the past 30 days; awareness was defined as having heard about e-cigarettes. Smokeless tobacco use was defined as using chewing tobacco, snuff, or dipping tobacco in the past 30 days. Cessation behaviors including having tried to quit smoking in the past 12 months (among current smokers) and thinking about quitting due to health warnings, were also collected. Secondhand smoke exposure was assessed, separately for exposure at home and exposure inside public places in the past 7 days.

Other variables included access and availability (including purchasing cigarettes from stores and being refused sale due to age), media and advertising (including exposure to anti‑tobacco messages in mass media), and knowledge and attitudes (including beliefs about health effects of smoking) and susceptibility to initiation (defined as never smokers who are not firm non-smokers).

Statistical analysis

All statistical analyses were performed using SPSS version 25.0 (IBM Corp., Armonk, NY, USA). Data were weighted to adjust for non‑response and selection probability. Prevalence estimates for key indicators were calculated with 95% confidence intervals. To assess trends over the 17-year period, we compared estimates across survey rounds using chi-squared tests for linear trends. Statistical significance was set at p<0.05. Gender differences within each survey round were also examined using chi‑squared tests. Given the cross‑sectional design, multivariable regression models were not applied; however, future studies may benefit from such approaches to control for potential confounding factors.

Sample characteristics

The study sample comprised 12049 adolescents aged 13–15 years across the three survey rounds. In the 2021 round (n=4019), the mean age was 14.1 years (SD=0.8), and the proportion of female participants was 51.2% (n=2058). Similar age and sex distributions were observed in the 2004 and 2015 rounds. Table 1 presents the descriptive characteristics of the sample by survey year.

Trends in cigarette smoking

Over the 17‑year period, current cigarette smoking among adolescents fell from 26.5% (95% CI: 25.2–27.8) in 2004 to 7.5% (95% CI: 6.9–8.1) in 2015 and further to 4.7% (95% CI: 4.2–5.2) in 2021 (p<0.001 for trend). Ever smoking decreased from 62.5% (95% CI: 61.0–64.0) to 20.7% (95% CI: 19.6–21.8, p<0.001). Frequent smoking (>20 cigarettes in the past month) decreased from 8.3% (95% CI: 7.4–9.2) to 1.2% (95% CI: 0.9–1.5, p<0.001). Susceptibility to initiating tobacco use fell from 48.9% (95% CI: 47.4–50.4) in 2004 to 6.5% (95% CI: 5.8–7.2) in 2015 and then increased slightly to 14.6% (95% CI: 13.7–15.5) in 2021 (p<0.001 for overall trend) (Table 1).

E-cigarette and smokeless tobacco use

Data on e-cigarette use were first collected in Belarus in 2021. Current e‑cigarette use prevalence was 15.1% (95% CI: 14.2–16.0), with no statistically significant gender difference (boys: 14.8%, girls: 15.4%, p=0.34). Awareness of e‑cigarettes was high: 92.6% of girls and 89.2% of boys reported having heard about e‑cigarettes (p<0.05). Use increased with grade: prevalence was 7.2% in 7th grade, 14.5% in 8th grade, and 24.5% in 9th grade (p<0.001 for trend).

Smokeless tobacco use decreased from 3.1% (95% CI: 2.6–3.6) in 2004 to 0.6% (95% CI: 0.4–0.8) in 2015 (p<0.001), then increased to 1.7% (95% CI: 1.4–2.0) in 2021 (p<0.001 for 2021 vs 2015).

Gender differences

In 2004, boys had higher rates of current cigarette smoking than girls (31.2%, 95% CI: 29.3–33.1 vs 21.7%, 95% CI: 20.0–23.4, p<0.001). In 2021, this pattern reversed: current any tobacco use was higher among girls (9.0%, 95% CI: 8.2–9.8) than boys (6.1%, 95% CI: 5.4–6.8, p<0.001). Current cigarette smoking was also higher in girls (5.7%, 95% CI: 5.0–6.4) than boys (3.7%, 95% CI: 3.1–4.3, p=0.002).

Age of initiation

The proportion of adolescents who reported trying their first cigarette before the age of 10 years decreased from 31.1% (95% CI: 29.2–33.0) in 2004 to 15.8% (95% CI: 14.4–17.2) in 2015, and further to 11.7% (95% CI: 10.6–12.8) in 2021 (p<0.001 for all comparisons).

Secondhand smoke exposure

Exposure to secondhand smoke at home fell from 75.3% (95% CI: 73.9–76.7) in 2004 to 25.1% (95% CI: 23.8–26.4) in 2015 and to 20.7% (95% CI: 19.6–21.8) in 2021 (p<0.001). Exposure inside public places fell from 90.1% (95% CI: 88.9–91.3) in 2004 to 30.3% (95% CI: 29.0–31.6) in 2015 and to 21.3% (95% CI: 20.2–22.4) in 2021 (p<0.001). Girls reported higher exposure than boys in all settings.

Cessation behaviors

Among current smokers, the proportion who had tried to quit in the past 12 months declined from 77.4% (95% CI: 75.2–79.6) in 2004 to 46.9% (95% CI: 43.7–50.1) in 2021 (p<0.001). The proportion thinking about quitting due to health warnings on cigarette packs fell among girls from 40.8% (95% CI: 37.5–44.1) in 2015 to 8.3% (95% CI: 6.4–10.2) in 2021 (p<0.001). Over 90% of current smokers across all years reported confidence that they could quit if they wanted to.

Access, availability and exposure to anti-tobacco media

Purchasing cigarettes from stores remained the primary source for minors, though this proportion declined from 47.0% (95% CI: 45.1–48.9) in 2004 to 26.4% (95% CI: 24.8–28.0) in 2021 (p<0.001). In 2021, 45.3% (95% CI: 42.1–48.5) of current smokers reported not being refused purchase due to age. Girls were less likely than boys to be refused sale in all survey years (e.g. 2021: 50.1% vs 37.8%, p<0.001).

The proportion of students who had seen anti‑tobacco messages in mass media decreased from 97.2% (95% CI: 96.5–97.9) in 2004 to 70.2% (95% CI: 68.9–71.5) in 2015 and to 49.2% (95% CI: 47.7–50.7) in 2021 (p<0.001).

Knowledge and attitudes towards smoking

The percentage of students who definitely thought that smoking was harmful to others decreased from 69.0% (95% CI: 67.6–70.4) in 2015 to 54.8% (95% CI: 53.3–56.3) in 2021 (p<0.001). Support for smoking bans in enclosed public places decreased from 92.8% (95% CI: 92.0–93.6) in 2015 to 85.2% (95% CI: 84.2–86.2) in 2021 (p<0.001). The belief that smoking helps people feel more comfortable socially increased from 20.1% (95% CI: 18.9–21.3) in 2015 to 31.0% (95% CI: 29.6–32.4) in 2021 (p<0.001).

E-cigarettes and smokeless tobacco

The finding that 15.1% of adolescents reported current e‑cigarette use in 2021 – more than three times the rate of cigarette smoking – represents a significant epidemiological shift. This pattern mirrors a global trend identified in the WHO 2023 report, where the tobacco industry has promoted e‑cigarettes as alternatives to traditional cigarettes⁴. The high awareness (over 90%) and increasing use with grade level, indicate that these products have penetrated the youth market. The observation that half of surveyed students did not perceive e‑cigarettes as harmful highlights a knowledge gap that may be addressed through targeted health education.

The resurgence of smokeless tobacco use in 2021 (1.7%) after a period of decline (from 3.1% in 2004 to 0.6% in 2015) is also concerning. This may be linked to experimentation with novel products or the misconception that ‘smokeless’ implies ‘harmless’. Belarus responded by amending its advertising law in 2021 to ban products similar in appearance or name to tobacco products⁵.

Gender differences

A notable finding is the reversal of the gender gap. The shift from higher smoking rates among boys in 2004 to higher rates among girls in 2021 requires further investigation. Girls showed a slower decline in smoking rates (3.8‑fold vs 8.5‑fold for boys) and higher current use of both tobacco and e‑cigarettes in 2021. The data indicate several potential vulnerabilities for girls: they were less likely to be refused cigarette purchases, reported higher exposure to secondhand smoke, and were 2.5 times more likely to be unaware of e‑cigarette risks. The sharp decline in response to health warning labels among girls (from 40.8% to 8.3% considering quitting) suggests that current messaging strategies may be less effective for this subgroup.

Cessation, media exposure, attitudes and social norms

The decline in quit attempts among current smokers (from 77.4% to 46.9%) coincided with a reduction in anti‑tobacco message exposure (from 97.2% to 49.2%). This temporal association suggests that reduced reach of health promotion campaigns may partially explain lower cessation motivation. The WHO 2023 report emphasizes that maintaining consistent messaging is crucial, as progress can be eroded⁴. Most adolescents who smoke attempt to quit without professional support, indicating a potential role for school‑based or digital cessation interventions.

The observed decrease in perceived harm of smoking to others (from 69.0% to 54.8%) and in support for smoking bans (from 92.8% to 85.2%) suggests a shift in social norms that may undermine past progress. The increase in the belief that smoking helps people feel more comfortable socially (from 20.1% to 31.0%) may indicate a partial normalization of smoking behavior among adolescents.

Limitations

Several limitations should be acknowledged. First, the cross-sectional design of each GYTS round prevents establishing causal relationships between policy implementation and observed trends. Second, data were self‑reported and may be subject to recall bias or social desirability bias, potentially leading to underestimation of tobacco use prevalence. Third, the survey was limited to school‑attending adolescents, excluding out‑of‑school youth who may have higher risk profiles. Fourth, e-cigarette data were collected only in 2021, precluding trend analysis for this product category. Fifth, residual confounding may be present as not all potential confounders (e.g. parental smoking, socioeconomic status) were measured or adjusted for. Future studies using longitudinal designs and multivariable regression models are needed to confirm these findings and to establish causal relationships.

Future research

Given the cross‑sectional nature of this study, further research is required. Longitudinal cohort studies could establish temporal sequences between policy implementation, product availability, and behavioral change. Qualitative studies could explore the social and cultural factors underlying the gender reversal observed in tobacco use. Intervention studies are needed to evaluate the effectiveness of gender-specific prevention programs and novel health messaging strategies targeting e‑cigarettes. Finally, continued monitoring through subsequent GYTS rounds will be essential to track emerging trends and evaluate policy responses.