Results

The prevalence of ever use of waterpipe and e-cigarettes varied widely across EU MS (Fig. 1). For waterpipe, prevalence was 31.2% (95% CI: 29.1–33.3%) in 2012, 35.0% (32.8–37.4%) in 2014, 29.1% (26.8–31.4%) in 2017, 31.8% (29.6–34.2%) in 2020, and 14.9% (13.2–16.8%) in 2023. Among minors, who are not legally permitted to purchase tobacco- or nicotine-containing products, prevalence declined from 25.2% in 2012 to 4.0% in 2023 (Table 1). Country-level prevalence ranged from 1.7% in Portugal (2023) to 75.5% in Sweden (2014). The greatest reduction between 2012 and 2023 was observed in Denmark (−42.2%), whereas Italy recorded minimal change (−0.1%) (Supplementary Table S4).

Fig. 1: Prevalence of ever waterpipe and e-cigarette use among young people across European Union Member States in 2012 and 2023. (a) Prevalence of ever waterpipe use (2012). (b) Prevalence of ever waterpipe use (2023). (c) Prevalence of ever e-cigarette use (2012). (d) Prevalence of ever e-cigarette use (2023). Regions with no data or excluded from the analyses are coloured in grey.

 | 2012 | 2014 | 2017 | 2020 | 2023
Weighted prevalence (%) | aPRs (95% CI) | Weighted prevalence (%) | aPRs (95% CI) | Weighted prevalence (%) | Weighted prevalence (%) | aPRs (95% CI) | Weighted prevalence (%) | aPRs (95% CI)
Ever use of waterpipe |  |  |  |  |  |  |  |  | 
Gender |  |  |  |  | REF. Year |  |  |  | 
Woman | 25.7% | 1.10 (0.97–1.24) | 30.1% | 1.15 (1.01–1.30) | 23.1% | 27.7% | 1.24 (1.02–1.52) | 12.9% | 0.51 (0.41–0.64)
Man | 36.3% | 1.04 (0.94–1.15) | 39.6% | 1.09 (0.98–1.21) | 34.7% | 35.7% | 1.08 (0.90–1.30) | 16.9% | 0.56 (0.44–0.72)
Age (years) |  |  |  |  |  |  |  |  | 
15–17 | 25.2% | 1.39 (1.07–1.81) | 24.2% | 1.36 (1.07–1.73) | 19.0% | 14.5% | 1.02 (0.71–1.46) | 4.0% | 0.44 (0.29–0.68)
18–24 | 35.3% | 1.05 (0.95–1.17) | 38.9% | 1.15 (1.01–1.32) | 31.7% | 32.1% | 1.10 (0.90–1.34) | 16.8% | 0.56 (0.44–0.70)
25–30 | 27.9% | 1.00 (0.89–1.13) | 33.6% | 1.03 (0.94–1.13) | 28.6% | 36.5% | 1.25 (1.06–1.49) | 15.7% | 0.55 (0.44–0.70)
Area of residence (exploratory) |  |  |  |  |  |  |  |  | 
Rural | 25.2% | 0.93 (0.77–1.12) | 33.5% | 1.12 (0.93–1.33) | 27.2% | 26.3% | 1.13 (0.97–1.32) | 9.7% | 0.53 (0.40–0.70)
Urban | 33.9% | 1.12 (1.02–1.23) | 35.6% | 1.12 (1.02–1.24) | 29.8% | 33.9% | 1.18 (0.96–1.44) | 16.8% | 0.54 (0.43–0.68)
Difficulty paying bills |  |  |  |  |  |  |  |  | 
Almost Never/Never | 33.2% | 1.06 (0.94–1.18) | 35.7% | 1.11 (0.98–1.25) | 30.8% | 31.3% | 1.06 (0.89–1.26) | 12.8% | 0.46 (0.37–0.57)
Most of the Time/ From Time to Time | 28.3% | 1.08 (0.94–1.24) | 34.0% | 1.12 (0.98–1.28) | 26.3% | 33.0% | 1.36 (1.09–1.70) | 18.8% | 0.71 (0.54–0.93)
Ever use of E-cigarettes |  |  |  |  |  |  |  |  | 
Gender |  |  |  |  |  |  |  |  | 
Woman | 9.8% | 0.46 (0.39–0.56) | 15.9% | 0.76 (0.65–0.88) | 19.1% | 20.1% | 1.20 (1.01–1.43) | 25.5% | 1.41 (1.18–1.69)
Man | 14.5% | 0.48 (0.41–0.56) | 21.8% | 0.76 (0.67–0.86) | 28.0% | 26.7% | 1.11 (0.99–1.25) | 26.2% | 1.11 (0.95–1.29)
Age (years) |  |  |  |  |  |  |  |  | 
15–17 | 8.6% | 0.44 (0.32–0.61) | 12.7% | 0.77 (0.57–1.03) | 18.3% | 16.6% | 1.02 (0.80–1.29) | 16.5% | 1.28 (0.93–1.75)
18–24 | 13.0% | 0.47 (0.39–0.56) | 19.8% | 0.74 (0.63–0.87) | 25.9% | 24.5% | 1.15 (0.99–1.33) | 31.5% | 1.28 (1.12–1.46)
25–30 | 12.4% | 0.48 (0.41–0.57) | 19.9% | 0.77 (0.69–0.86) | 22.5% | 24.4% | 1.17 (1.02–1.34) | 21.4% | 1.17 (0.96–1.42)
Area of residence (exploratory) |  |  |  |  |  |  |  |  | 
Rural | 11.4% | 0.40 (0.32–0.50) | 18.6% | 0.69 (0.59–0.80) | 24.5% | 21.4% | 0.97 (0.82–1.14) | 24.0% | 1.17 (0.94–1.46)
Urban | 12.6% | 0.50 (0.43–0.59) | 19.1% | 0.78 (0.68–0.89) | 23.3% | 24.3% | 1.23 (1.09–1.39) | 26.5% | 1.27 (1.11–1.46)
Difficulty paying bills |  |  |  |  |  |  |  |  | 
Almost Never/Never | 12.7% | 0.50 (0.43–0.59) | 17.1% | 0.76 (0.66–0.87) | 22.4% | 21.7% | 1.16 (1.03–1.31) | 25.1% | 1.38 (1.20–1.60)
Most of the Time/ From Time to Time | 11.5% | 0.44 (0.36–0.54) | 22.2% | 0.75 (0.65–0.88) | 25.7% | 27.3% | 1.15 (0.99–1.34) | 27.2% | 1.06 (0.89–1.27)

Prevalence of ever e-cigarette use increased from 12.2% (10.8–13.8%) in 2012 to 25.8% (23.7–28.2%) in 2023 (Fig. 1). Among 15–17-year-olds, a similar upward pattern was observed, with prevalence raising from 8.6% in 2012 to 16.5% in 2023 (Table 1). Country-level prevalence ranged from 1.2% (0.3–12.7%) in Sweden (2012) to 58.5% (49.8–66.8%) in Latvia (2023). Most countries observed increases, ranging from 0.9% in Hungary to 50.8% in Sweden. Only Portugal and Romania recorded declines of 2.0% and 10.0%, respectively (Supplementary Table S5).

Prevalence of current waterpipe use was 2.5% (1.8–3.5%) in 2017, 5.1% (4.1–6.4%) in 2020, and 1.8% (1.2–2.7%) in 2023 (Fig. 2). Among minors, prevalence declined from 5.4% to 0.1% over the same period (Table 2). Estimates ranged from 0.4% in Finland (2020) and the Netherlands (2023) to 18.5% in Luxembourg (2020). France and Portugal recorded minimal decreases (<0.1%) while Latvia had the greatest decline (−7.5%) from 2017 to 2023. In contrast, Poland and Austria observed increases of 1.3% and 2.1%, respectively (Supplementary Table S6).

Fig. 2: Prevalence of current waterpipe and e-cigarette use among young people across European Union Member States in 2017 and 2023. (e) Prevalence of current waterpipe use (2017). (f) Prevalence of current waterpipe use (2023). (g) Prevalence of current e-cigarette use (2017). (h) Prevalence of current e-cigarette use (2023). Regions with no data or excluded from the analyses are coloured in grey.

Prevalence of current e-cigarette use increased across EU, rising from 1.6% (1.1–2.3%) in 2017 to 5.3% (4.0–6.8%) in 2023 (Fig. 2). Similarly, among 15–17-year-olds, prevalence rose from 1.2% to 4.4% (Table 2). Country-level prevalence ranged from 0.4% in Ireland and Spain (2017) to 18.0% in Estonia (2023). Latvia recorded the largest increase in prevalence (+13.8%), whereas Sweden showed the smallest change (<0.1%). Declines were observed only in Austria (−2.5%) and Greece (−1.3%) (Supplementary Table S7). Prevalence could not be estimated for several countries for either product due to very low estimates or lack of observations.

 | 2017 | 2020 | 2023
Weighted prevalence (%) | Weighted prevalence (%) | aPRs (95% CI) | Weighted prevalence (%) | aPRs (95% CI)
Current use of waterpipe
Gender | REF. Year |  |  |  | 
Woman | 1.6% | 3.3% | 2.30 (1.46–3.63) | 2.0% | 1.19 (0.57–2.47)
Man | 3.4% | 6.7% | 1.85 (1.18–2.90) | 1.6% | 0.47 (0.31–0.72)
Age (years) |  |  |  |  | 
15–17 | 5.4% | 4.3% | 1.50 (0.77–2.93) | 0.1% | 0.08 (0.01–0.76)
18–24 | 2.5% | 5.7% | 2.16 (1.29–3.64) | 2.2% | 0.57 (0.30–1.07)
25–30 | 1.8% | 4.6% | 1.92 (1.14–3.22) | 1.7% | 1.00 (0.54–1.84)
Area of residence |  |  |  |  | 
Rural | 2.1% | 3.9% | 1.63 (0.96–2.78) | 0.3% | 0.23 (0.11–0.48)
Urban | 2.6% | 5.5% | 2.07 (1.39–3.07) | 2.3% | 0.82 (0.49–1.40)
Difficulty paying bills |  |  |  |  | 
Almost Never/Never | 2.9% | 4.2% | 1.57 (1.01–2.44) | 1.1% | 0.60 (0.28–1.30)
Most of the Time/ From Time to Time | 1.9% | 6.9% | 2.76 (1.66–4.60) | 3.0% | 0.81 (0.41–1.63)
Current use of E-cigarettes
Gender |  |  |  |  | 
Woman | 0.7% | 1.7% | 2.12 (1.15–3.91) | 6.6% | 8.47 (4.57–15.71)
Man | 2.4% | 3.9% | 1.41 (1.00–1.97) | 4.0% | 2.52 (1.63–3.94)
Age (years) (exploratory) |  |  |  |  | 
15–17 | 1.2% | 2.6% | 1.54 (0.64–3.68) | 4.4% | 4.48 (2.30–8.74)
18–24 | 1.4% | 3.2% | 1.64 (1.05–2.58) | 6.2% | 4.58 (2.67–7.83)
25–30 | 1.9% | 2.5% | 1.52 (0.94–2.44) | 4.3% | 3.64 (2.15–6.17)
Area of residence (exploratory) |  |  |  |  | 
Rural | 1.6% | 3.3% | 1.48 (0.76–2.88) | 5.9% | 3.29 (1.82–5.94)
Urban | 1.6% | 2.7% | 1.75 (1.22–2.49) | 5.0% | 4.82 (2.91–7.99)
Difficulty paying bills (exploratory) |  |  |  |  | 
Almost Never/Never | 1.4% | 2.5% | 1.30 (0.89–1.90) | 4.7% | 4.38 (3.11–6.18)
Most of the Time/ From Time to Time | 1.8% | 3.6% | 2.00 (1.15–3.47) | 6.4% | 3.61 (1.66–7.86)

The Poisson regression models showed that, compared with 2017, ever waterpipe use was significantly lower in 2023 (aPR: 0.52, 0.42–0.66) and higher in 2014 (aPR: 1.12, 1.02–1.24), while e-cigarette use was significantly lower in pre-2017 waves (2012 aPR: 0.47, 0.40–0.54; 2014 aPR: 0.75, 0.67–0.84), and higher in 2020 (aPR: 1.16, 1.03–1.31) and in 2023 (aPR: 1.24, 1.07–1.44). Current waterpipe use was higher in 2020 (aPR: 1.96, 1.16–3.28) relative to 2017, while e-cigarette use increased in both 2020 (aPR: 1.64, 1.15–2.34) and 2023 (aPR: 4.27, 2.73–6.67) (Fig. 3).

Fig. 3: Adjusted prevalence ratios for prevalence of ever and current waterpipe and e-cigarette use across waves relative to 2017. aPR: Adjusted prevalence ratio. CI: Confidence interval.

Trends in ever waterpipe use varied across sociodemographic subgroups in earlier waves (Table 1). Among women, prevalence decreased in 2017 compared with 2014 (aPR: 1.15, 1.01–1.30) and increased again in 2020 (aPR: 1.24, 1.02–1.52). By age, higher prevalence relative to 2017 occurred earlier among younger groups and later among older groups: increases were observed in 2012 and 2014 among 15–17-year-olds, in 2014 among 18–24-year-olds, and in 2020 among 25–30-year-olds. Changes over time were broadly similar by financial circumstances, except in 2020, when prevalence increased among individuals reporting financial difficulty compared with 2017 (aPR: 1.36, 1.09–1.70). In 2023, all subgroups showed a statistically significant decline in prevalence relative to 2017 (aPR range: 0.44–0.71).

Across all sociodemographic subgroups, the prevalence of ever e-cigarette use was significantly lower in 2012 and 2014 than in 2017 (aPR range: 0.40–0.78) (Table 1). Post-2017 trends differed across groups, with women, urban residents, and those without financial difficulties observing greater relative increases in prevalence than their counterparts in 2020 and 2023. By age, an increase was observed in 2020 among 25–30-year-olds (aPR: 1.17, 1.02–1.34) and in 2023 among 18–24-year-olds (aPR: 1.28, 1.12–1.46), but not among other age groups.

Most subgroups recorded higher prevalence of current waterpipe use in 2020 relative to 2017, with varied magnitudes (Table 2). Relative increases were observed among both genders, those with and without financial difficulties, 18–24- and 25–30-year-olds, and urban residents. By 2023, statistically significant declines were observed among men, the youngest subgroup, and rural residents, but not among their counterparts.

Between 2017 and 2023, current e-cigarette use increased substantially across all subgroups (aPR range: 2.52–8.47) (Table 2). Among women, prevalence doubled in 2020 (aPR: 2.12, 1.15–3.91) and increased eightfold by 2023 (aPR: 8.47, 4.57–15.71), whereas among men prevalence increased 2.52-fold in 2023 (95% CI: 1.63–3.94).

Sensitivity analyses yielded estimates consistent in direction and magnitude with the primary analyses for both products at the EU level, with minor variations across alternative definitions and after exclusion of the 2020 wave; some estimates, particularly for waterpipe current use, lost statistical significance, whereas increasing trends in e-cigarette use remained robust (Supplementary Tables S8–S10).