Abstract
Background
We aimed to evaluate the potential independent and interactive associations of tobacco and alcohol use with suicide ideation, plan, and attempt among adolescents, and the role of legal purchase age restrictions among these associations.
Methods
A pooled cross-sectional analysis was conducted with data from the Global School-based Student Health Survey (GSHS) (2013–2019), the US Youth Risk Behavior Surveillance System (YRBSS) (2021), and a Chinese school-based health survey (2017). Self-reported use of tobacco and alcohol during the past 30 days, and suicide ideation, plan, and attempt during the past year, were collected from questionnaires. Country-level purchase age restrictions on tobacco and alcohol products were obtained from official government websites.
Results
Among the 211,022 adolescents from 58 countries, 46.7% were boys. Compared with adolescents who used neither tobacco nor alcohol, those who used both tobacco and alcohol had the highest odds of suicide ideation (boys: 2.42 [95% CI: 2.25–2.61]; girls: 3.19 [95% CI: 2.98–3.40]; P difference < 0.001), suicide plan (boys: 2.39 [95% CI: 2.21–2.58]; girls: 3.33 [95% CI: 3.11–3.57]; P difference < 0.001), and suicide attempt (boys: 3.24 [95% CI: 3.00–3.51]; girls: 4.03 [95% CI: 3.75–4.33]; P difference < 0.001). In countries with higher suicide plan prevalence, boys who use tobacco and who live in countries with tobacco purchase age restrictions had lower odds of suicide plan (1.86 [95% CI: 1.64–2.12]) than those who lived in countries without restrictions (2.81 [95% CI: 2.32–3.42]). Similarly, girls who use tobacco and who live in countries with legal age purchase restrictions displayed lower odds of suicide plan (2.20 [95% CI: 1.98–2.45]) compared to those who live in countries without restrictions (4.61 [95% CI: 3.65–5.83]). However, our study revealed no subgroup differences in whether countries have legal tobacco or alcohol purchase age restrictions in association with suicide ideation, plan, and attempt risk in countries with a lower prevalence of suicide behaviors.
Conclusions
Tobacco and alcohol consumption are strongly associated with the risk of suicide behaviors among adolescents, particularly among girls. Country-level tobacco control strategies may have the potential to reduce risks of suicide behaviors associated with tobacco use among adolescents.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12916-025-03983-6.
Keywords: Adolescent health, Global health, Suicide behavior, Tobacco use, Alcohol use, Health policy
Background
Suicidal behaviors, including ideation, plan, and attempt, are the leading cause of death and disability among adolescents worldwide [1]. The Sustainable Development Goals (SDGs) emphasize the importance of ensuring healthy lives and promoting well-being for all ages, setting the target (Target 3.4) to reduce premature mortality from non-communicable diseases and promote mental health and well-being by one-third by 2030 [2]. Reducing suicidal behaviors is crucial for preventing premature deaths and achieving this target among adolescents and young people. However, as we approach the halfway mark to the 2030 goals, the global situation remains far from satisfactory and may be exacerbated by the ongoing impact of the pandemic [3, 4]. Suicide ideation, plan, and attempt share common risk factors, and recognizing these shared risk factors is crucial for implementing more effective suicide control strategies across diverse populations [5, 6].
Suicide behaviors are affected by a wide range of sociodemographic, psychiatric, physical, and other influences [7]. Among various risk factors, substance use, particularly tobacco and alcohol use, has been identified as a significant contributor to suicidal behaviors in adolescents from both high-income and low/middle-income countries [8, 9]. Recent studies have shown that tobacco and alcohol use are highly prevalent among adolescents and are often associated with an increased risk of suicidal behaviors [10]. For instance, a cross-sectional study from China and the USA in 2017 reported that adolescents aged 7 to 12 years who reported cigarette and alcohol use were more likely to be at risk for suicidal ideation and attempts [11]. Another large-scale study in the USA in 2019 reported that adolescents inpatients aged 12 to 18 years with alcohol use disorders were at 18% higher odds of suicide-related hospitalization [12]. Despite these findings, there is a substantial research gap regarding the global perspective on the associations between these substances and suicidal behaviors [13]. Most existing studies are region-specific and do not provide a comprehensive analysis of these associations across different countries.
In many countries, policies have been developed to establish and increase legal purchase age restrictions (the minimum age at which individuals are allowed to buy such products) on alcohol and tobacco products, in order to decrease the health damage directly related to them, such as substance independence, injuries, and of course suicidal behaviors [14–16]. A meta-analysis suggested that interventions limiting alcohol use led to a reduction in self-harm suicide attempts in adolescents [17]. Studies in the USA have reported that increasing the minimum age for alcohol consumption is associated with a lower suicide rate among adolescents. These findings suggest that alcohol-related restrictions should be prioritized, especially when tobacco and alcohol use coexist, to reduce the risk of suicide [18, 19]. However, the findings of these studies do not sufficiently elucidate the associations between tobacco and alcohol use and various suicide ideation, plan, and attempt across different countries globally, nor do they explore the potential interactions between these factors. Additionally, few studies have investigated the applicability of policy interventions for reducing the risk of substance-related suicide behaviors, especially from a global perspective [20]. It remains unclear whether these policies are positively related to a decreased risk of suicidal behaviors associated with tobacco and alcohol use, and the consistency of these policies’ effectiveness across different population groups also warrants discussion. More importantly, although it is well known that girls and non-Hispanic adolescents are more affected by these issues compared to boys and Hispanic adolescents [21, 22], it remains unclear whether they benefit equally from these policies.
To bridge these knowledge gaps, we examined the independent and interactive associations of tobacco and alcohol use with suicide ideation, plan, and attempt among adolescents using data from a pooled school-based survey of 211,022 adolescents from 58 countries, and further investigated whether the presence of legal purchase age restrictions on tobacco and alcohol use could reduce the risk of suicide ideation, plan, and attempt associated with tobacco and alcohol use among adolescents.
Methods
Study population and survey design
We performed a pooled analysis of data from three pre-existing surveys that all employed a cross-sectional design and utilized a similar stratified multistage sampling method, along with consistent data curation procedures [23], including the US Youth Risk Behavior Surveillance System (YRBSS) in 2021, the Global School-based Student Health Survey (GSHS) from 2013 through the most recent, and a school-based student health survey conducted in Jiangxi and Jiangsu Provinces of China in 2017. Analysis of deidentified secondary data did not involve an ethical review of human subjects.
The US adolescent data were derived from the YRBSS, which was established in 1991, and aims to monitor six categories of priority health-risk behaviors among young people and adolescents [23], and the latest YRBSS data can be accessed freely from the Centers for Disease Control and Prevention of the USA at https://www.cdc.gov/healthyyouth/data/yrbs/data.htm [24].
The GSHS is a collaborative surveillance project designed to help countries measure and assess behavioral risk factors and protective factors in 10 key areas among adolescents aged 12 to 17. The system has generated comparable data across countries. More than 100 countries worldwide have conducted at least one round of GSHS [25, 26]. The GSHS followed a two-stage sampling design in each of the surveyed countries. In the first stage, schools were randomly selected from a list of all schools in a country with a probability proportional to the number of students enrolled. In the second stage, classes were randomly selected within schools using a random-start systematic equal probability sampling method [27]. GSHS data were collected through self-administered questionnaires with a standardized methodology. The raw data from the GSHS can be downloaded freely from the World Health Organization website (http://www.who.int/ncds/surveillance/microdata) [26, 28]. Specifically, we selected data from countries where the survey was conducted between 2013 and the most recent available year. All countries with available data on tobacco and alcohol use, and suicide ideation, plan, and attempt, were included. If a country conducted more than one survey during the period, the latest survey data were included in the cross-sectional analysis.
The Chinese school-based student health survey was carried out by the Centers for Disease Control and Prevention in Jiangxi and Jiangsu Provinces, aligned with the methodology of the GSHS, and was gathered in 2017. The two provinces were selected based on their geography, gross domestic product (GDP), average standing within China, and Han Chinese population, with one representing coastal areas and inland areas to ensure they are broadly representative.
In this study, 211,102 adolescents aged 12 to 17 years from 58 countries were recruited. The participant inclusion process from each data source is displayed in Fig. 1.
Fig. 1.
Diagram of participant inclusion process for analysis
Exposure variables
The main exposure variables were self-reported use of tobacco products during the past 30 days (yes/no), and self-reported alcohol intake during the past 30 days (yes/no). Tobacco use was assessed using two questions: “During the past 30 days, on how many days did you smoke cigarettes?” and “During the past 30 days, on how many days did you use any tobacco products other than cigarettes?”. Alcohol use was assessed via the following question: “During the past 30 days, on how many days did you have at least one drink containing alcohol?”. The questionnaire details and definitions of exposure are displayed in Additional file 1: Table 1.
Country-level legal purchase age restrictions on tobacco products and alcohol were collected from official national websites or from the World Health Organization (see details from the Additional file 1: supplementary methodology). We recorded whether the countries (or territories) had legal purchase age restrictions at the time when the data were collected, as well as the specific age. The details of countries with or without such restrictions are displayed in Additional file 1: Tables 2 and 3. Regarding legal purchase age restrictions for tobacco use, 49 countries have purchase age restrictions, 6 countries have no purchase age restrictions, and 3 countries have no relevant information. As for legal purchase age restrictions for alcohol use, data from the Global Information System on Alcohol and Health (age limits for on-premise service of beer, wine, and spirits) available at https://www.who.int/data/gho/data/indicators/indicator-details/GHO/age-limits-on-premise-service- indicates that 48 countries have purchase age restrictions, 7 countries have no purchase age restrictions, and 3 countries have no relevant information.
Outcome assessments
Three suicide-related variables, namely suicide ideation, suicide plan, and suicide attempt during the past 12 months, were applied as binary variables (yes or no). Additionally, suicide ideation, plan, and attempt represent stages of suicide behavior, ranging from thinking about suicide without action, to forming a specific method for it, and finally engaging in behaviors with the intent to end one’s life. Questionnaire details and definitions of the outcomes were displayed in Additional file 1: Table 1. A sample size of each country and the prevalence of suicide ideation, suicide plan, and suicide attempt are shown in Additional file 1: Table 4. Since prior research evidence suggests that policies are not uniformly effective across regions with different suicide behavior prevalence rates [29], in stratified analysis, countries were categorized as having low prevalence or high prevalence rates, on the basis of the country-level median prevalence of suicide ideation, plan, and attempt, respectively, to explore the potential effect of age restrictions on alcohol and tobacco purchases on the risk of suicide ideation, plan, and attempt associated with them. The prevalence rates of tobacco and alcohol use with or without legal purchase age are shown in Additional file 1: Tables 5 and 6.
Covariates
The main confounding factors of interest included individual demographic characteristics, lifestyle factors, mental well-being, and country-based sociodemographic factors. The demographic characteristics included self-reported age, sex (either boy or girl), and body mass index (BMI). BMI was calculated from self-reported height and weight at survey time and was converted to z scores according to the World Health Organization (WHO) growth reference [30]. Lifestyle factors included physical activity and sedentary behavior habits [31]. Physical activity was defined as the number of days with at least 60 min of vigorous or moderate physical activity during the past week. Physical inactivity was defined as hours of sedentary behavior on a typical day (less than 1 h; 1 to 2 h; 3 to 4 h; 5 to 6 h; 7 to 8 h; more than 8 h). Mental health indicators included the self-reported presence of feeling lonely (yes/no) and feeling worried (yes/no) during the past 12 months. The country-level Socio-demographic Index (SDI) for the survey year, representing social and economic development, was obtained from the Global Burden of Diseases study (https://www.healthdata.org/research-analysis/gbd).
Statistical analysis
The demographic and behavioral characteristics of the study population were summarized using descriptive statistics. For continuous variables, we reported the mean and standard deviation (SD), and for categorical variables, we provided the frequency (proportion). The prevalence of tobacco and alcohol use among adolescents in 58 countries between 2013 and 2021 was displayed with a world map.
Then, binary logistic regression models were applied to analyze the independent associations of tobacco and alcohol use with suicide ideation, plan, and attempt adjusting for age, body mass index, country, insufficient physical activity (yes/no), long sedentary behavior (yes/no), survey weight, feeling lonely during the past 12 months (yes/no), feeling worried during the past 12 months (yes/no), and SDI. Additionally, we employed logistic regression models to further assess the interactive associations of tobacco and alcohol use with suicide ideation, plan, and attempt, where the exposure variables were categorized on the basis of the combination of tobacco and alcohol use: neither tobacco nor alcohol use (as the reference group), tobacco use only, alcohol use only, and both tobacco and alcohol use. Models were adjusted for age, BMI, country, insufficient physical activity (yes/no), long sedentary behavior (yes/no), survey weight, feeling lonely during the past 12 months (yes/no), feeling worried during the past 12 months (yes/no), and SDI. With an ICC below 0.05, the influence of “country” on the variance in suicide ideation/plan/attempt is minimal, indicating that the “country” variable should be adjusted as a fixed effect, without using a random effects model [32]. The sex differences in each group were examined using the Wald test [33].
To analyze whether the purchase age restrictions played similar roles for both sexes and in countries with different suicide behavior risk (low or high, based on the country-level prevalence of each suicide behavior), the above analyses were conducted by sex and by country-level risk of suicide ideation, plan, and attempt. Participants were divided into four groups: participants who do not use tobacco (or alcohol) from countries without purchase age restrictions (reference group), tobacco (or alcohol) users from countries without purchase age restrictions, participants who did not use tobacco (or alcohol) from countries with purchase age restrictions, and tobacco (or alcohol) users from countries with purchase age restrictions. To analyze whether the purchase age restrictions played similar roles for both sexes and in countries with different suicide behavior risk (low or high, based on the country-level prevalence of each suicide behavior), the above analyses were conducted by sex and by country-level risk of suicide ideation, plan, and attempt. Models were adjusted for age, BMI, insufficient physical activity (yes/no), long sedentary behavior (yes/no), survey weight, feeling lonely during the past 12 months (yes/no), feeling worried during the past 12 months (yes/no), and SDI. The countries without available information on purchase age restrictions (Curacao, Mozambique, and Tokelau for tobacco products; Anguilla, Curacao, and Bahrain for alcohol) were not included in this part of the analysis.
Sensitivity analysis
Several sensitivity analyses were conducted. Firstly, we conducted a two-step random-effects meta-analysis to evaluate the association between the independent associations of tobacco and alcohol use with suicide ideation, plan, and attempt based on different countries. In the first step, we performed binary logistic regression models to calculate the effect estimates for each country, adjusting for age, BMI, insufficient physical activity (yes/no), long sedentary behavior (yes/no), survey weight, feeling lonely during the past 12 months (yes/no), and feeling worried during the past 12 months (yes/no). In the second step, a random effects meta-analysis was further performed to analyze the overall pooled estimated odds among the total participants as well as boys and girls. Secondly, to examine whether the frequency of tobacco and alcohol use contributed to higher odds of suicide ideation, plan, and attempt, interactions between tobacco and alcohol use on suicide ideation, plan, and attempt were conducted in those with moderate tobacco and alcohol users (1–5 days during the past month) and with frequent users (more than 5 days during the past month), respectively. Thirdly, countries with purchase age restrictions were further divided into two groups based on the age limit (≥ 18 years old or < 18 years old) to examine whether the restrictions that did not cover the entire study population (ages 12 to 17) were associated, to some extent, with a reduced odds of suicide ideation, plan, and attempt.
To obtain nationally representative results, sampling weights were taken into consideration (except for China, which did not include weight in the original design). All of the analyses were conducted for the overall participants and by sex group. All the statistical analyses were performed with R (version 4.3.1). Two-sided P values < 0.05 were considered statistically significant.
Results
Participant characteristics
A total of 211,022 adolescents (46.7% boys) from 58 countries were included in our analysis. The demographic and behavioral characteristics of the participants are displayed in Table 1. In total, the pooled prevalence rates for suicide ideation, plan, and attempt were 14.9%, 13.5%, and 11.1%, respectively. Among participants who used tobacco, the pooled prevalence rates of suicide ideation, plan, and attempt were 34.4%, 33.7%, and 36.9%, respectively, for boys, and 27.4%, 27.1%, and 29.0%, respectively, for girls. Among participants who used alcohol, the pooled prevalence of suicide ideation, plan, and attempt were 40.6%, 39.4%, and 41.7% in boys, and 42.7%, 42.8%, and 42.7% in girls, respectively.
Table 1.
Characteristics of the participants included in the analysis
| Characteristics | Sample size | Suicide ideation | Suicide plan | Suicide attempt | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Boys | Girls | Total | Boys | Girls | Total | Boys | Girls | Total | Boys | Girls | |
| BMI, kg/m2, mean (SD) | 98,457 | 112,565 | 21.92 ± 5.15 | 21.43 ± 4.99 | 22.18 ± 5.21 | 21.81 ± 5.11 | 21.29 ± 4.97 | 22.10 ± 5.17 | 21.72 ± 5.10 | 21.21 ± 4.97 | 22.04 ± 5.16 |
| Age, years, mean (SD) | 14.93 ± 1.40 | 14.94 ± 1.42 | 14.93 ± 1.38 | 14.89 ± 1.40 | 14.88 ± 1.43 | 14.90 ± 1.38 | 14.87 ± 1.40 | 14.91 ± 1.42 | 14.85 ± 1.38 | ||
| Tobacco use, N (%) | |||||||||||
| No | 71,137 | 90,330 | 20,071 (70.19) | 6555 (65.61) | 13,516 (72.65) | 18,195 (70.49) | 6233 (66.28) | 11,962 (72.90) | 14,331 (67.89) | 5209 (63.10) | 9122 (70.97) |
| Yes | 18,375 | 12,714 | 8524 (29.81) | 3436 (34.39) | 5088 (27.35) | 7617 (29.51) | 3171 (33.72) | 4446 (27.10) | 6777 (32.11) | 3046 (36.90) | 3731 (29.03) |
| Alcohol use, N (%) | |||||||||||
| No | 62,999 | 75,745 | 15,894 (58.00) | 5515 (59.36) | 10,379 (57.30) | 14,294 (58.39) | 5256 (60.61) | 9038 (57.17) | 11,722 (57.65) | 4519 (58.29) | 7203 (57.26) |
| Yes | 23,032 | 23,680 | 11,508 (42.00) | 3775 (40.64) | 7733 (42.70) | 10,188 (41.61) | 3416 (39.39) | 6772 (42.83) | 8611 (42.35) | 3234 (41.71) | 5377 (42.74) |
| Insufficient physical activitya, N (%) | |||||||||||
| No | 64,103 | 85,582 | 22,438 (73.48) | 7399 (69.05) | 15,039 (75.87) | 20,266 (73.82) | 7008 (69.80) | 13,258 (76.13) | 16,817 (73.70) | 6176 (69.44) | 10,641 (76.43) |
| Yes | 31,720 | 24,101 | 8099 (26.52) | 3317 (30.95) | 4782 (24.13) | 7189 (26.18) | 3032 (30.20) | 4157 (23.87) | 6000 (26.30) | 2718 (30.56) | 3282 (23.57) |
| Daily sedentary behavior ≥ 5 h, N (%) | |||||||||||
| No | 76,948 | 85,505 | 21,013 (69.24) | 7749 (73.00) | 13,264 (67.21) | 19,149 (70.24) | 7378 (74.23) | 11,771 (67.94) | 16,550 (72.91) | 6776 (76.76) | 9774 (70.46) |
| Yes | 18,517 | 23,979 | 9337 (30.76) | 2866 (27.00) | 6471 (32.79) | 8115 (29.76) | 2561 (25.77) | 5554 (32.06) | 6149 (27.09) | 2051 (23.24) | 4098 (29.54) |
| Feeling lonely, N (%) | |||||||||||
| No | 36,071 | 29,193 | 4931 (17.79) | 2470 (25.22) | 2461 (13.72) | 5246 (20.56) | 2720 (28.88) | 2526 (15.69) | 4140 (19.49) | 2246 (27.76) | 1894 (14.40) |
| Yes | 53,007 | 74,410 | 22,794 (82.21) | 7322 (74.78) | 15,472 (86.28) | 20,274 (79.44) | 6697 (71.12) | 13,577 (84.31) | 17,106 (80.51) | 5844 (72.24) | 11,262 (85.60) |
| Feeling worried, N (%) | |||||||||||
| No | 37,969 | 32,118 | 5246 (18.80) | 2639 (26.68) | 2607 (14.48) | 5503 (21.46) | 2890 (30.46) | 2613 (16.18) | 4119 (19.28) | 2191 (26.83) | 1928 (14.61) |
| Yes | 51,462 | 71,848 | 22,651 (81.20) | 7251 (73.32) | 15,400 (85.52) | 20,140 (78.54) | 6599 (69.54) | 13,541 (83.82) | 17,240 (80.72) | 5976 (73.17) | 11,264 (85.39) |
SD standard deviation, BMI body mass index
aInsufficient physical activity time was defined as not reaching the current World Health Organization recommendation of being active for more than 60 min on ≥ 5 days per week
The prevalence of tobacco and alcohol use across 58 countries is depicted in Additional file 1: Fig. S1. Among all the participants, 16.1% reported tobacco use within the past 30 days. The percentage of the population engaged in current tobacco use varied from 3.1 to 34.9%, with Lebanon having the highest prevalence and Cambodia having the lowest. Boys generally presented higher percentages than girls did. Similarly, the percentage of current alcohol use in the population ranged from 0.9% to 58.1%, with Argentina having the highest prevalence and Bangladesh having the lowest.
Meta-analysis of the associations between tobacco use and alcohol use, and suicide ideation, plan, and attempt
The meta-analysis of the associations between tobacco use and alcohol use, and suicide ideation, plan, and attempt are presented in Additional file 1: Fig. S2–S4. Overall, tobacco use was related to an increased odds of suicide ideation, suicide plan, and suicide attempt, with ORs of 2.48 (95% CI 2.32 to 2.66), 2.39 (95% CI 2.24 to 2.56), and 3.46 (95% CI 3.10 to 3.86), respectively. The country with the highest associated odds for suicide ideation, plan, and attempt was Cambodia, with ORs of 6.90 (95% CI 3.90 to 12.21), 4.82 (95% CI 2.85 to 8.17), and 8.05 (95% CI 4.78 to 13.57), respectively.
The associations between alcohol use and elevated odds of suicide ideation, plan, and attempt were significant, with ORs of 1.71 (95% CI 1.54 to 1.89), 1.65 (95% CI 1.51 to 1.80), and 1.85 (95% CI 1.66 to 2.07), respectively. Among the countries, the highest estimates of odds for suicide ideation (OR = 3.69, 95% CI: 2.42 to 5.63) were observed in Honduras. Similarly, Samoa had the greatest odds for suicide plan (OR = 2.72, 95% CI: 1.59 to 4.64), and Nepal for suicide attempt (OR = 6.32, 95% CI: 3.91 to 10.21).
Independent and interactive associations of tobacco and alcohol use on suicide ideation, plan, and attempt
The independent and interactive associations between tobacco use, alcohol use, and suicide ideation, plan, and attempt are presented in Table 2. Girls who reported using tobacco were found to have significantly greater odds of suicide ideation (OR = 2.46, 95% CI 2.34 to 2.59, P < 0.001), suicide plan (OR = 2.47, 95% CI 2.35 to 2.61, P < 0.001), and suicide attempt (OR = 2.94, 95% CI 2.79 to 3.11, P < 0.001) than boys (suicide ideation: OR = 1.97, 95% CI 1.87 to 2.08, P < 0.001; suicide plan: OR = 1.98, 95% CI 1.87 to 2.09, P < 0.001; suicide attempt: OR = 2.39, 95% CI 2.25 to 2.53, P < 0.001, respectively). Similarly, the likelihoods of suicide ideation (OR = 1.67, 95% CI 1.60 to 1.76, P < 0.001), and suicide plan (OR = 1.74, 95% CI 1.66 to 1.83, P < 0.001) among girls who reported alcohol use was also significantly greater than boys (suicide ideation: OR = 1.43, 95% CI 1.35 to 1.51, P < 0.001; suicide plan: OR = 1.37, 95% CI 1.29 to 1.45, P < 0.001, respectively).
Table 2.
Independent and interactive associations of tobacco use and alcohol use on suicide ideation, plan, and attempt
| Tobacco and alcohol use | Suicide ideation | Suicide plan | Suicide attempt | |||
|---|---|---|---|---|---|---|
| OR (95%CI) | P for difference | OR (95%CI) | P for difference | OR (95%CI) | P for difference | |
| Independent associations | ||||||
| Tobacco use a | ||||||
| Boys | 1.97 (1.87,2.08) | < 0.001 | 1.98 (1.87,2.09) | < 0.001 | 2.39 (2.25,2.53) | < 0.001 |
| Girls | 2.46 (2.34,2.59) | 2.47 (2.35,2.61) | 2.94 (2.79,3.11) | |||
| Alcohol use b | ||||||
| Boys | 1.43 (1.35,1.51) | < 0.001 | 1.37 (1.29,1.45) | < 0.001 | 1.66 (1.56,1.77) | 0.082 |
| Girls | 1.67 (1.60,1.76) | 1.74 (1.66,1.83) | 1.78 (1.69,1.88) | |||
| Interactive associations | ||||||
| Neither | ||||||
| Boys | Ref | Ref | Ref | Ref | Ref | Ref |
| Girls | Ref | Ref | Ref | Ref | Ref | Ref |
| Only alcohol use | ||||||
| Boys | 1.28 (1.19,1.38) | < 0.001 | 1.25 (1.15,1.35) | < 0.001 | 1.48 (1.36,1.61) | 0.005 |
| Girls | 1.60 (1.52,1.69) | 1.66 (1.57,1.76) | 1.72 (1.62,1.83) | |||
| Only tobacco use | ||||||
| Boys | 1.75 (1.60,1.91) | < 0.001 | 1.79 (1.63,1.96) | < 0.001 | 2.07 (1.88,2.27) | < 0.001 |
| Girls | 2.75 (2.50,3.02) | 2.80 (2.54,3.08) | 3.51 (3.18,3.87) | |||
| Both use | ||||||
| Boys | 2.42 (2.25,2.61) | < 0.001 | 2.39 (2.21,2.58) | < 0.001 | 3.24 (3.00,3.51) | < 0.001 |
| Girls | 3.19 (2.98,3.40) | 3.33 (3.11,3.57) | 4.03 (3.75,4.33) | |||
Models were adjusted for age, body mass index, country, insufficient physical activity (yes/no), long sedentary behavior (yes/no), survey weight, SDI, feeling lonely during the past 12 months (yes/no), and feeling worried during the past 12 months (yes/no)
aadditionally adjusted for current alcohol use (yes/no)
badditionally adjusted for current tobacco use (yes/no)
Compared with adolescents who used neither tobacco nor alcohol, those who used both tobacco and alcohol had the highest odds of suicide ideation (boys: OR = 2.42, 95% CI 2.25 to 2.61, P < 0.001; girls: OR = 3.19, 95% CI 2.98 to 3.40, P < 0.001; P difference < 0.001), suicide plan (boys: OR = 2.39, 95% CI 2.21 to 2.58, P < 0.001; girls: OR = 3.33, 95% CI 3.11 to 3.57, P < 0.001; P difference < 0.001), and suicide attempt (boys: OR = 3.24, 95% CI 3.00 to 3.51, P < 0.001; girls: OR = 4.03, 95% CI 3.75 to 4.33, P < 0.001; P difference < 0.001).
Additionally, the interactive associations between the frequency of tobacco use and alcohol use, and suicide ideation, plan, and attempt are presented in Additional file 1: Table 7. Generally, we found similar interactive associations in both the moderate-frequency use and high-frequency use groups.
Associations of tobacco and alcohol use with suicide ideation, plan, and attempt, by sex and purchase age restrictions
The interactive associations of legal purchase age restrictions, tobacco use, and alcohol use on suicide ideation, plan, and attempt are presented in Fig. 2. In countries with a lower prevalence of suicide attempts, boys who use tobacco and are from countries with legal tobacco purchase age restrictions had lower odds of exhibiting suicide attempt (OR = 2.79, 95% CI 2.31 to 3.38) compared to those from countries without restrictions (OR = 4.54, 95% CI 3.41 to 6.04; P difference = 0.006).
Fig. 2.
Associations of tobacco and alcohol use with suicide ideation, plan, and attempt among adolescents in 58 countries, by sex, age of purchase restrictions, and the prevalence of suicide behaviors. Note: models were adjusted for age, body mass index, insufficient physical activity (yes/no), long sedentary behavior (yes/no), survey weight, SDI, feeling lonely during the past 12 months (yes/no), and feeling worried during the past 12 months (yes/no). P1 for difference indicates to differences of suicide ideation, plan, and attempt in tobacco use (or alcohol use) from countries with or without legal purchase age restrictions in countries with lower prevalence of suicide ideation, plan, and attempt; P2 for difference indicates to differences of suicide ideation, plan, and attempt in tobacco use (or alcohol use) from countries with or without legal purchase age restrictions in countries with higher prevalence of suicide ideation, plan, and attempt. Lower group refers to below the country-level prevalence of suicide ideation, plan, and attempt, respectively; higher group refers to above the country-level prevalence of suicide ideation, plan, and attempt, respectively. In A, there were no “No tobacco use and no purchase age restrictions” and “Only tobacco use” among the lower group
In countries with a higher prevalence of suicide ideation, boys who use tobacco and from countries with legal age restrictions on tobacco purchases displayed lower odds of suicide ideation (OR = 2.15, 95% CI 1.90 to 2.43) compared to those from countries without such restrictions (OR = 2.93, 95% CI 2.44 to 3.50; P difference = 0.006). Similarly, girls who use tobacco from countries with legal age restrictions on tobacco purchases displayed lower odds of suicide ideation (OR = 2.37, 95% CI 2.14 to 2.63) compared to those from countries without such restrictions (OR = 4.22, 95% CI 3.43 to 5.19; P difference < 0.001). In countries with a higher prevalence of suicide plan, boys who use tobacco and from countries with legal age restrictions on tobacco purchases had lower odds of suicide plan (OR = 1.86, 95% CI 1.64 to 2.12) compared to those from countries without such restrictions (OR = 2.81, 95% CI 2.32 to 3.42; P difference = 0.001). Similarly, girls who use tobacco from countries with legal age restrictions on tobacco purchases displayed lower odds of suicide plan (OR = 2.20, 95% CI 1.98 to 2.45) compared to those from countries without such restrictions (OR = 4.61, 95% CI 3.65 to 5.83; P difference < 0.001).
There were no subgroup differences in countries with or without legal alcohol purchase age restrictions in association with suicide behavior risk in countries with lower suicide behavior prevalence. In countries with higher prevalence, boys who use alcohol and are from countries with legal alcohol purchase age restrictions had higher odds of suicide ideation (OR = 2.93, 95% CI 2.29 to 3.74), suicide plan (OR = 2.13, 95% CI 1.74 to 2.62), and suicide attempt (OR = 2.35, 95% CI 1.92 to 2.87) compared to countries without restrictions (suicide ideation: OR = 1.68, 95% CI 1.14 to 2.47; suicide plan: OR = 1.47, 95% CI 1.09 to 1.98; suicide attempt: OR = 1.43, 95% CI 1.07 to 1.93). In countries with a higher prevalence of suicide attempts, girls who use alcohol and are from countries with legal alcohol purchase age restrictions had higher odds of suicide attempt (OR = 2.11, 95% CI 1.77 to 2.51) compared to countries without restrictions (OR = 1.38, 95% CI 1.05 to 1.80).
Additionally, we compared subgroups based on whether the legal purchase age restrictions for tobacco and alcohol use were set at 18 years. As detailed in Tables 3 and 4, we found similar associations of legal purchase age restrictions with the associations of tobacco, and alcohol use with suicide ideation, plan, and attempt among adolescents, by sex and purchase age restrictions. Specifically, in countries with a higher prevalence of suicide ideation and plan, boys and girls who use tobacco and from countries with legal age restrictions on tobacco purchases displayed lower odds of suicide ideation and plan compared to those from countries without such restrictions (P difference < 0.05). There were no subgroup differences in countries with or without legal alcohol purchase age restrictions in association with suicide behavior risk in countries with lower suicide behavior prevalence both boys and girls (P difference > 0.05). In countries with higher prevalence, boys who use alcohol and are from countries with legal alcohol purchase age restrictions had higher odds of suicide plan (OR = 6.10, 95% CI 2.12 to 17.52), and suicide attempt (OR = 8.15, 95% CI 2.89 to 22.97) compared to countries without restrictions (suicide plan: OR = 1.43, 95% CI 1.05 to 1.94; suicide attempt: OR = 1.39, 95% CI 1.02 to 1.89) among age of purchase restrictions < 18 years. Meanwhile, in countries with higher prevalence, girls who use alcohol and are from countries with legal alcohol purchase age restrictions had higher odds of suicide ideation (OR = 7.06, 95% CI 3.32 to 15.02) compared to countries without restrictions (suicide ideation: OR = 2.58, 95% CI 2.06 to 3.93) among the age of purchase restrictions < 18 years.
Table 3.
Associations between tobacco use and suicide ideation, plan, and attempt among adolescents in 58 countries, by sex, age of purchase restrictions and the prevalence of suicide behaviors
| Tobacco use and purchase age restrictions | Lower group | P1 for difference | Higher group | P2 for difference |
|---|---|---|---|---|
| OR (95%CI) | OR (95%CI) | |||
| Age of purchase restrictions ≥ 18 yearsa | ||||
| Boys | ||||
| Suicide ideation | — | 0.006 | ||
| No tobacco use and no purchase age restrictions | — | Ref | ||
| Only tobacco usea | — | 2.93 (2.44,3.50) | ||
| Only purchase age restrictions | Ref | 1.03 (0.91,1.15) | ||
| Both tobacco use and purchase age restrictionsa | 1.86 (1.71,2.02) | 2.15 (1.90,2.43) | ||
| Suicide plan | 0.658 | < 0.001 | ||
| No tobacco use and no purchase age restrictions | Ref | Ref | ||
| Only tobacco usea | 3.23 (1.99,5.23) | 2.81 (2.32,3.42) | ||
| Only purchase age restrictions | 1.55 (1.17,2.06) | 0.88 (0.78,0.99) | ||
| Both tobacco use and purchase age restrictionsa | 2.84 (2.14,3.78) | 1.86 (1.64,2.12) | ||
| Suicide attempt | 0.006 | 0.775 | ||
| No tobacco use and no purchase age restrictions | Ref | Ref | ||
| Only tobacco usea | 4.54 (3.41,6.04) | 2.99 (2.32,3.86) | ||
| Only purchase age restrictions | 1.23 (1.02,1.48) | 1.14 (0.97,1.34) | ||
| Both tobacco use and purchase age restrictionsa | 2.81 (2.32,3.40) | 2.86 (2.42,3.39) | ||
| Girls | ||||
| Suicide ideation | — | < 0.001 | ||
| No tobacco use and no purchase age restrictions | — | Ref | ||
| Only tobacco usea | — | 4.22 (3.43,5.19) | ||
| Only purchase age restrictions | Ref | 0.95 (0.86,1.04) | ||
| Both tobacco use and purchase age restrictionsa | 3.16 (2.89,3.46) | 2.37 (2.14,2.63) | ||
| Suicide plan | 0.006 | < 0.001 | ||
| No tobacco use and no purchase age restrictions | Ref | Ref | ||
| Only tobacco usea | 2.39 (1.50,3.80) | 4.61 (3.65,5.83) | ||
| Only purchase age restrictions | 1.69 (1.34,2.14) | 0.87 (0.79,0.96) | ||
| Both tobacco use and purchase age restrictionsa | 4.97 (3.91,6.32) | 2.20 (1.98,2.45) | ||
| Suicide attempt | 0.138 | 0.073 | ||
| No tobacco use and no purchase age restrictions | Ref | Ref | ||
| Only tobacco usea | 6.22 (3.96,9.77) | 3.59 (2.82,4.57) | ||
| Only purchase age restrictions | 1.05 (0.89,1.24) | 1.00 (0.88,1.14) | ||
| Both tobacco use and purchase age restrictionsa | 4.31 (3.60,5.15) | 2.79 (2.43,3.20) | ||
Models were adjusted for age, body mass index, insufficient physical activity (yes/no), long sedentary behavior (yes/no), SDI, feeling lonely during the past 12 months (yes/no), and feeling worried during the past 12 months (yes/no). P1 for difference indicates to differences of suicide ideation, plan, and attempt in tobacco use from countries with or without legal purchase age restrictions in countries with lower prevalence of suicide ideation, plan, and attempt; P2 for difference indicates to differences of suicide ideation, plan, and attempt in tobacco use from countries with or without legal purchase age restrictions in countries with higher prevalence of suicide ideation, plan, and attempt. Lower group refers to below the country-level prevalence of suicide ideation, plan, and attempt, respectively; higher group refers to above the country-level prevalence of suicide ideation, plan, and attempt, respectively. In the lower group, there were no “No tobacco use and no purchase age restrictions” and “Only tobacco use” groups
aOnly one country (Dominican Republic) had a restricted purchase age of < 18 years, therefore the stratified analysis was only conducted in countries whose restricted purchase ages were ≥ 18 years
Table 4.
Associations between alcohol use and suicide ideation, plan, and attempt among adolescents in 58 countries, by sex, age of purchase restrictions, and the prevalence of suicide behaviors
| Alcohol use and legal purchase age | Lower group | P1 for difference | Higher group | P2 for difference |
|---|---|---|---|---|
| OR (95%CI) | OR (95%CI) | |||
| Age of purchase restrictions ≥ 18 years | ||||
| Boys | ||||
| Suicide ideation | 0.572 | 0.017 | ||
| No alcohol use and no purchase age restrictions | Ref | Ref | ||
| Only alcohol use | 1.96 (1.47,2.60) | 1.68 (1.14,2.47) | ||
| Only purchase age restrictions | 1.21 (1.01,1.43) | 2.13 (1.66,2.73) | ||
| Both alcohol use and purchase age restrictions | 2.07 (1.71,2.49) | 2.93 (2.30,3.75) | ||
| Suicide plan | 0.579 | 0.035 | ||
| No alcohol use and no purchase age restrictions | Ref | Ref | ||
| Only alcohol use | 1.65 (1.16,2.35) | 1.46 (1.08,1.97) | ||
| Only purchase age restrictions | 1.01 (0.85,1.20) | 1.58 (1.29,1.93) | ||
| Both alcohol use and purchase age restrictions | 1.85 (1.54,2.22) | 2.15 (1.76,2.64) | ||
| Suicide attempt | 0.979 | 0.006 | ||
| No alcohol use and no purchase age restrictions | Ref | Ref | ||
| Only alcohol use | 2.26 (1.57,3.25) | 1.43 (1.06,1.93) | ||
| Only purchase age restrictions | 0.96 (0.79,1.17) | 1.40 (1.15,1.71) | ||
| Both alcohol use and purchase age restrictions | 2.25 (1.82,2.78) | 2.36 (1.93,2.89) | ||
| Girls | ||||
| Suicide ideation | 0.224 | 0.668 | ||
| No alcohol use and no purchase age restrictions | Ref | Ref | ||
| Only alcohol use | 3.02 (2.31,3.95) | 2.90 (2.11,3.99) | ||
| Only purchase age restrictions | 1.21 (1.06,1.38) | 2.04 (1.65,2.51) | ||
| Both alcohol use and purchase age restrictions | 2.50 (2.15,2.90) | 3.15 (2.57,3.87) | ||
| Suicide plan | 0.430 | 0.274 | ||
| No alcohol use and no purchase age restrictions | Ref | Ref | ||
| Only alcohol use | 2.55 (1.68,3.89) | 1.64 (1.27,2.11) | ||
| Only purchase age restrictions | 1.37 (1.20,1.57) | 1.15 (0.98,1.36) | ||
| Both alcohol use and purchase age restrictions | 3.06 (2.62,3.56) | 1.94 (1.64,2.30) | ||
| Suicide attempt | 0.234 | 0.010 | ||
| No alcohol use and no purchase age restrictions | Ref | Ref | ||
| Only alcohol use | 3.97 (2.65,5.95) | 1.38 (1.05,1.80) | ||
| Only purchase age restrictions | 1.26 (1.08,1.47) | 1.18 (1.00,1.41) | ||
| Both alcohol use and purchase age restrictions | 3.04 (2.55,3.62) | 2.10 (1.76,2.51) | ||
| Age of purchase restrictions < 18 years | ||||
| Boys | ||||
| Suicide ideation | — | 0.192 | ||
| No alcohol use and no purchase age restrictions | — | Ref | ||
| Only alcohol use | — | 1.68 (1.14,2.46) | ||
| Only purchase age restrictions | Ref | 2.76 (0.96,7.92) | ||
| Both alcohol use and purchase age restrictions | 1.81 (1.33,2.46) | 3.39 (1.27,9.09) | ||
| Suicide plan | 0.754 | 0.010 | ||
| No alcohol use and no purchase age restrictions | Ref | Ref | ||
| Only alcohol use | 1.67 (1.15,2.42) | 1.43 (1.05,1.94) | ||
| Only purchase age restrictions | 0.73 (0.29,1.86) | 4.29 (1.47,12.53) | ||
| Both alcohol use and purchase age restrictions | 1.43 (0.58,3.52) | 6.10 (2.12,17.52) | ||
| Suicide attempt | 0.182 | 0.001 | ||
| No alcohol use and no purchase age restrictions | Ref | Ref | ||
| Only alcohol use | 2.21 (1.50,3.26) | 1.39 (1.02,1.89) | ||
| Only purchase age restrictions | 0.61 (0.21,1.74) | 3.19 (1.06,9.58) | ||
| Both alcohol use and purchase age restrictions | 1.06 (0.39,2.91) | 8.15 (2.89,22.97) | ||
| Girls | ||||
| Suicide ideation | — | 0.030 | ||
| No alcohol use and no purchase age restrictions | — | Ref | ||
| Only alcohol use | — | 2.85 (2.06,3.93) | ||
| Only purchase age restrictions | Ref | 3.22 (1.44,7.23) | ||
| Both alcohol use and purchase age restrictions | 2.24 (1.65,3.03) | 7.06 (3.32,15.02) | ||
| Suicide plan | 0.068 | 0.931 | ||
| No alcohol use and no purchase age restrictions | Ref | Ref | ||
| Only alcohol use | 2.08 (1.34,3.23) | 1.67 (1.28,2.16) | ||
| Only purchase age restrictions | 2.12 (0.96,4.72) | 0.95 (0.32,2.76) | ||
| Both alcohol use and purchase age restrictions | 4.80 (2.20,10.51) | 1.75 (0.61,4.97) | ||
| Suicide attempt | 0.596 | 0.548 | ||
| No alcohol use and no purchase age restrictions | Ref | Ref | ||
| Only alcohol use | 3.30 (2.15,5.07) | 1.44 (1.09,1.90) | ||
| Only purchase age restrictions | 1.57 (0.63,3.88) | 1.19 (0.36,3.98) | ||
| Both alcohol use and purchase age restrictions | 4.29 (1.80,10.21) | 2.07 (0.65,6.60) | ||
Models were adjusted for age, body mass index, insufficient physical activity (yes/no), long sedentary behavior (yes/no), survey weight, SDI, feeling lonely during the past 12 months (yes/no), and feeling worried during the past 12 months (yes/no). P1 for difference indicates to differences of suicide ideation, plan, and attempt in alcohol use from countries with or without legal purchase age restrictions in countries with lower prevalence of suicide ideation, plan, and attempt; P2 for difference indicates to differences of suicide ideation, plan, and attempt in alcohol use from countries with or without legal purchase age restrictions in countries with higher prevalence of suicide ideation, plan, and attempt. Lower group refers to below the country-level prevalence of suicide ideation, plan, and attempt, respectively; higher group refers to above the country-level prevalence of suicide ideation, plan, and attempt, respectively. In purchase age restrictions (< 18 years) countries, there were no “No tobacco use and no purchase age restrictions” and “Only tobacco use” groups among the lower group
Discussion
On the basis of the most recent data from 58 countries collected between 2013 and 2021, we found that the pooled prevalence rates of suicide ideation, plan, and attempt were 14.9%, 13.5%, and 11.1%, respectively. Both tobacco and alcohol use are significantly associated with an increased odds of suicide ideation, plan, and attempt, with girls exhibiting significantly greater odds than boys. Among the adolescents, those who used both tobacco and alcohol presented greater odds of suicide behaviors compared to those who used neither substance. Additionally, age restrictions on tobacco purchases are associated with decreased odds of suicide behaviors in both boys and girls, especially in countries with a high prevalence of suicide behaviors. On the other hand, alcohol purchase age restrictions are not related to lower odds of alcohol-related suicide behaviors. Our findings indicated that these associations remained robust regardless of whether the age of purchase restrictions was < 18 years or ≥ 18 years old. These findings, from a globalized perspective, suggest that tobacco and alcohol use are associated with adverse mental health outcomes. Age restrictions on tobacco purchases may be associated with a lower risk of tobacco-related suicide behaviors, particularly among girls and in countries with a high prevalence of such behaviors.
Smoking and alcohol use have been found to be associated with suicide ideation, plan, and attempt in numerous population-based studies. One study involving more than 30,000 adolescents from China and the USA reported that both alcohol and tobacco use are associated with increased suicide ideation and attempts [11]. Studies in Korean adolescents reported similar findings and further reported that both conventional and electric smoking were associated with an increase in suicidal behavior, with a greater frequency of smoking associated with a greater risk of suicidal behavior [34, 35]. Since these substance use factors are potentially modifiable risk factors for suicidal behaviors, targeted intervention strategies with prioritization are integral for suicide prevention [36]. Some studies have reported a greater risk of suicide among adolescents who use both tobacco and alcohol [37–39]. One study conducted in the USA revealed that for students who had suicidal ideation or suicide plans, alcohol use was not associated with suicide attempts, whereas cigarette smoking was associated with suicide attempts [19]. Our study contributes to the existing body of related research by providing global validation of the above studies, which were conducted in the USA. Specifically, we confirmed the association between smoking and an increased risk of all suicide-related behaviors, emphasizing the importance of addressing tobacco control in adolescent suicide prevention efforts. In addition, our findings suggest that tobacco use is associated with greater odds of suicide-related behaviors in girls than in boys. Although the self-reported rates of tobacco and alcohol use are lower among girls, considering that they constitute more than 70% of reported suicide behaviors, girls who engage in smoking and drinking behaviors may constitute the highest risk group for suicide mortality. On the one hand, girls may experience more interpersonal problems and other social pressures that are closely related to substance use and mental health risks including suicidal behavior [40, 41]. The hormonal differences between sexes, particularly the effects of estrogen, could also contribute to the greater risk of suicidal behaviors related to tobacco and alcohol use in girls. The interaction between estrogen and nicotine could potentially enhance the mood-altering effects of nicotine, making females more susceptible to its depressive or dysphoric effects, which could contribute to an increased risk of suicide-related behaviors [42, 43]. Therefore, special attention should be given to developing targeted approaches and strategies for intervention efforts in this subgroup.
Research on physiological mechanisms has confirmed that tobacco and alcohol consumption are associated with an increased risk of suicide in adults. Rapid nicotine uptake into the brain through any form of smoking, including inhalation of conventional cigarettes, or electronic cigarettes, or through passive smoking, can facilitate long-term sensitization and short-term impulsivity, both of which impair regulatory action and predispose individuals to negative effects and suicidal behaviors [44]. Further analysis of the 50-year observational study utilizing Swedish medical, criminal, and pharmacy registries revealed a potential association between drinking disorders and suicidal behavior, particularly among women, along with evidence of family aggregation [45]. In addition to reducing suicide-related deaths, strengthening the implementation of the tobacco control framework (SDG Health Target 3.a) and preventing harmful alcohol abuse (SDG Health Target 3.5) are important goals consistent with SDG 3 “Ensuring healthy lives and promote well-being for all at all ages” [46]. Adolescents should undoubtedly be the focus of these initiatives to build a sustainable future for generations. However, until 2022, 2.3 billion people still lived in countries that did not adopt a single tobacco control measure, and passive smoking killed 1.3 million nonsmokers a year [47]. Six of the 58 countries we studied have no age restrictions on purchasing tobacco, and seven have no age restrictions on purchasing alcohol. These countries are located in South America, Southeast Asia, the Western Pacific, Africa, and the Eastern Mediterranean Region, and their adolescent population is unprotected from tobacco and alcohol harm. Inequalities in health promotion policies between countries also contribute to the high prevalence of suicidal behaviors among adolescents in these areas and the extreme inequality of data accessibility. Moreover, sometimes, even if the government imposes age restrictions on the purchase of certain tobacco or alcohol products, adolescents from some countries still have easy access to alternate tobacco and alcohol products, such as cigars and hookahs [48–50]. In our current analysis, the countries with the greatest risks of tobacco-related and alcohol-related suicidal behaviors also had age restrictions on tobacco or alcohol purchases. In addition to these policies, the lack of comprehensive life skills education for young people and some socioenvironmental factors such as conflict conditions and social transition might have a greater influence on suicide behaviors [51–53]. Moreover, in societies where tobacco and alcohol use are culturally embedded, age restrictions alone may not be sufficient to reduce consumption among adolescents and young adults, thereby failing to mitigate associated risks [54]. While our findings suggest that legal purchase age restrictions may be associated with lower suicide risks related to tobacco use among adolescents, it is important to recognize that these restrictions often coexist with broader public health policies, such as anti-tobacco campaigns, school-based prevention programs, and mental health initiatives. For example, in countries like Australia, legal age restrictions are part of a comprehensive tobacco control framework that includes plain packaging laws and public education efforts, which together may amplify the protective effects against tobacco-related harms, including suicide risks. Future studies should aim to disentangle the specific contributions of legal purchase age restrictions from the cumulative impact of these broader public health measures [46].
The main strength of our study is that we used a standardized questionnaire for the measurement of tobacco use, alcohol use, and suicidal ideation, plan, and attempt. This study utilized data from three different datasets including the GSHS, YRBSS, and the China survey. The study timeframe, study design, and questionnaire on exposures and outcomes in the two databases (YRBSS and GSHS datasets) were the same. The study design, population, definitions of exposures and outcomes, and data collection procedures in the China survey were also consistent with the two databases, to ensure comparability across different countries. All of the data had the same design and sampling strategy and therefore provided directly comparable results across the 58 countries. There are several limitations to the present study. First, most of the data (except for data from the USA) were collected prior to the COVID-19 pandemic, during which time adolescents experienced enormous strain specifically. The instant and long-term impacts of pandemics on both substance abuse and psychosocial wellness are difficult to estimate and require longitudinal observations [55]. Second, although we have successfully examined the combined effect of tobacco and alcohol use on suicidal behavior from a global perspective, the cross-sectional design of our study limits our ability to establish causal relationships between these behaviors and suicidal outcomes. Therefore, there is a need for higher-quality longitudinal studies to be conducted in regions where data are lacking to further explore these associations. Third, all exposures and outcome measures were collected through self-administered questionnaires. While questionnaires have demonstrated good reliability and validity among adolescents worldwide, it is important to acknowledge that the possibility of adolescents providing responses that do not accurately reflect their true circumstances cannot be entirely ruled out. Factors such as misunderstanding, intentional concealment, or other motivations might introduce bias to some extent in the study results. Owing to approximately 10% missing data in this study, we opted for a deletion approach to handle these cases. Additionally, the analysis was conducted using data from the GSHS across 58 countries, all countries with available exposure and outcome variables were included, but those with small sample sizes or no valid effect estimates (14 out of 58 countries) were included only in the logistic regression analyses and excluded from two-step meta-analyses. This approach was necessary to maintain the robustness of country-specific results, but might limit the generalizability of our findings to countries excluded. Fourth, the lack of original sampling weights in the Chinese dataset being limited to two provinces, which may limit the understanding for the broader Chinese population. This limitation could affect the findings’ effectiveness and representativeness for the broader Chinese population. Additionally, age restriction data for tobacco and alcohol purchases were collected from government websites during January 2024 and may not reflect the regulations in the country during which the individual data were collected. Further research is needed to track the long-term impact of tobacco and alcohol-related laws and regulations on adolescent suicidal behaviors. Sixth, owing to the absence of individual socio-economic information in the GSHS survey, this study could not further consider the potential confounding effects of socio-economic information on the associations. Finally, the exposures (tobacco and alcohol use) in this study were assessed based on their use in the past 30 days, whereas the outcomes (suicidal ideation, plan, and attempt) were assessed on the basis of their occurrences in the past year. This inconsistency in the timing of the survey questions introduces a limitation, as it is possible that the suicidal behaviors could have preceded substance use. This limitation suggests potential bidirectional associations, such as alcohol use following suicidal ideation as a coping mechanism.
Conclusions
In conclusion, our study provides a novel global perspective of tobacco and alcohol use in adolescent suicide ideation, plan, and attempt. This study revealed that tobacco and alcohol consumption are strongly associated with the risk of suicide behaviors among adolescents, particularly among girls. Additionally, the existence of tobacco purchasing age restrictions may be associated with a lower prevalence of suicide behaviors among adolescents. Country-level tobacco control strategies may have the potential to reduce risks of suicide behaviors associated with tobacco use among adolescents.
Supplementary Information
Additional File 1: Tables S1-S7. Table S1. Questionnaire details and definitions of exposures and outcomes. Table S2. Legal age restrictions on tobacco purchase among included countries. Table S3. Legal age restrictions on alcohol purchase among included countries. Table S4. Sample size of each country and the prevalence of suicide ideation, plan, and attempt. Table S5. The prevalence of tobacco use with or without legal purchase age. Table S6. The prevalence of alcohol use with or without legal purchase age. Table S7. Interactive associations between tobacco use and alcohol use, and suicide ideation, plan, and attempt among different frequency use. Figures S1-S4. Fig. S1. Prevalence of tobacco use and alcohol use among adolescents in 58 countries between 2013 and 2021. Fig. S2. Meta-analysis of tobacco use, alcohol use, and suicide ideation, plan, and attempt among overall adolescents in 58 countries. Fig. S3. Meta-analysis of tobacco use, alcohol use, and suicide ideation, plan, and attempt among boys in 58 countries. Fig. S4. Meta-analysis of tobacco use, alcohol use, and suicide ideation, plan, and attempt among girls in 58 countries.
Acknowledgements
We extend our gratitude to all participants involved in this study. We are grateful to Professor Yongfu Yu (Department of Biostatistics, Key Laboratory for Health Technology Assessment, National Commission of Health, Key Laboratory of Public Health Safety of Ministry of Education, School of Public Health, Fudan University), and Professor Chao Yang (Digital Intelligence Medicine Center, Peking University First Hospital) for his invaluable statistical suggestions on the data analyses.
Data sharing statement
The raw data from the GSHS can be accessed from the World Health Organization website (http://www.who.int/ncds/surveillance/microdata). The latest YRBSS data can be accessed from the Centers for Disease Control and Prevention of the USA at https://www.cdc.gov/healthyyouth/data/yrbs/data.htm. The Chinese data from the student health survey in Jiangxi and Jiangsu Provinces are available upon reasonable request to Professor Zhiyong Zou.
Abbreviations
- BMI
Body mass index
- GDP
Gross domestic product
- GSHS
Global School-based Student Health Survey
- SD
Standard deviation
- SDGs
Sustainable Development Goals
- SDI
Socio-demographic Index
- WHO
World Health Organization
- YRBSS
US Youth Risk Behavior Surveillance System
Authors’ contributions
Z.Z. had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. M.C. and X.W. conducted and are responsible for the data analysis and drafting of the manuscript. Concept and design were Z.Z., Y.J., and W.L. Acquisition, analysis, or interpretation of data were M.C., X.W., J.G., H.W., and J.L. Administrative, technical, or material support were J.L., Z.Z., and Y.J.. Supervision was Z. Z., Y.J., and W.L.
Funding
The present study was supported by the National Natural Science Foundation (No. 82204067), Research on the Construction of a New Public Health Science System and Talent Training Model (No. 201920102401), and the Fundamental Research Funds for the Central Universities (No. 3332024187). The funder has no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; or decision to submit the manuscript for publication.
Data availability
No datasets were generated or analysed during the current study.
Declarations
Declarations
Ethics approval and consent to participate.
Not applicable.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Manman Chen and Xijie Wang contributed equally to this work.
Contributor Information
Zhiyong Zou, Email: harveyzou2002@bjmu.edu.cn.
Yu Jiang, Email: jiangyu@pumc.edu.cn.
References
- 1.Adolescent and young adult health. https://www.who.int/news-room/fact-sheets/detail/adolescents-health-risks-and-solutions. Accessed 4 Dec 2023.
- 2.Goal 3 | Department of Economic and Social Affairs. https://sdgs.un.org/goals/goal3#targets_and_indicators. Accessed 5 Dec 2023.
- 3.United Nations. The Sustainable Development Goals Report 2021. New York: United Nations; 2021.
- 4.United Nations. The Sustainable Development Goals Report 2022. New York: United Nations; 2022.
- 5.Nock MK, Green JG, Hwang I, McLaughlin KA, Sampson NA, Zaslavsky AM, Kessler RC. Prevalence, Correlates, and Treatment of Lifetime Suicidal Behavior Among Adolescents. JAMA Psychiat. 2013;70:300. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Klonsky ED, May AM, Saffer BY. Suicide, Suicide Attempts, and Suicidal Ideation. Annu Rev Clin Psychol. 2016;12:307–30. [DOI] [PubMed] [Google Scholar]
- 7.Favril L, Yu R, Geddes JR, Fazel S. Individual-level risk factors for suicide mortality in the general population: an umbrella review. Lancet Public Health. 2023;8:e868–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Korhonen T, Sihvola E, Latvala A, Dick DM, Pulkkinen L, Nurnberger J, Rose RJ, Kaprio J. Early-onset tobacco use and suicide-related behavior – A prospective study from adolescence to young adulthood. Addict Behav. 2018;79:32–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Baiden P, Mengo C, Boateng GO, Small E. Investigating the association between age at first alcohol use and suicidal ideation among high school students: Evidence from the youth risk behavior surveillance system. J Affect Disord. 2019;242:60–7. [DOI] [PubMed] [Google Scholar]
- 10.Hua LL, Lee J, Rahmandar MH, Sigel EJ. Suicide and Suicide Risk in Adolescents. Pediatrics. 2024. 10.1542/peds.2023-064800. [DOI] [PubMed] [Google Scholar]
- 11.Guo L, Wang W, Du X, Guo Y, Li W, Zhao M, Wu R, Lu C. Associations of Substance Use Behaviors With Suicidal Ideation and Suicide Attempts Among US and Chinese Adolescents. Front Psychiatry. 2021. 10.3389/fpsyt.2020.611579. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Basith SA, Nakaska MM, Sejdiu A, Shakya A, Namdev V, Gupta S, Mathialagan K, Makani R. Substance Use Disorders (SUD) and Suicidal Behaviors in Adolescents: Insights From Cross-Sectional Inpatient Study. Cureus. 2021. 10.7759/cureus.15602. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Cha CB, Franz PJ, M. Guzmán E, Glenn CR, Kleiman EM, Nock MK,. Annual Research Review: Suicide among youth – epidemiology, (potential) etiology, and treatment. J Child Psychol Psychiatry. 2018;59:460–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Toumbourou J, Stockwell T, Neighbors C, Marlatt G, Sturge J, Rehm J. Interventions to reduce harm associated with adolescent substance use. The Lancet. 2007;369:1391–401. [DOI] [PubMed] [Google Scholar]
- 15.Iacobucci G. England plans to raise legal age for buying tobacco each year to create “smoke-free” generation. BMJ. 2023;282:2297. [DOI] [PubMed]
- 16.DeJong W, Blanchette J. Case Closed: Research Evidence on the Positive Public Health Impact of the Age 21 Minimum Legal Drinking Age in the United States. J Stud Alcohol Drugs Suppl. 2014;17:108–15. [DOI] [PubMed]
- 17.Witt K, Chitty KM, Wardhani R, Värnik A, de Leo D, Kõlves K. Effect of alcohol interventions on suicidal ideation and behaviour: A systematic review and meta-analysis. Drug Alcohol Depend. 2021;226:108885. [DOI] [PubMed] [Google Scholar]
- 18.Birckmayer J, Hemenway D. Minimum-age drinking laws and youth suicide, 1970–1990. Am J Public Health. 1999;89:1365–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Wang M, Qin A, Wei Z, Sun L. Differentiating the associations between alcohol use, cigarette smoking, and conditional suicidal behaviors among adolescents. J Affect Disord. 2023;341:112–8. [DOI] [PubMed] [Google Scholar]
- 20.Stockings E, Hall WD, Lynskey M, Morley KI, Reavley N, Strang J, Patton G, Degenhardt L. Prevention, early intervention, harm reduction, and treatment of substance use in young people. Lancet Psychiatry. 2016;3:280–96. [DOI] [PubMed] [Google Scholar]
- 21.Spiller HA, Ackerman JP, Spiller NE, Casavant MJ. Sex- and Age-specific Increases in Suicide Attempts by Self-Poisoning in the United States among Youth and Young Adults from 2000 to 2018. J Pediatr. 2019;210:201–8. [DOI] [PubMed] [Google Scholar]
- 22.Uddin R, Burton NW, Maple M, Khan SR, Khan A. Suicidal ideation, suicide planning, and suicide attempts among adolescents in 59 low-income and middle-income countries: a population-based study. Lancet Child Adolesc Health. 2019;3:223–33. [DOI] [PubMed] [Google Scholar]
- 23.Centers for Disease Control and Prevention. Methodology of the youth risk behavior surveillance system - 2013. MMWR Recommendations and Reports. 2013;62:1–23. [PubMed] [Google Scholar]
- 24.U.S. Centers for Disease Control and Prevention. Youth Risk Behavior Surveillance System (YRBSS). 2021. https://www.cdc.gov/yrbs/data/index.html. Accessed 16 Nov 2023.
- 25.World Health Organization. Noncommunicable Disease Surveillance, Monitoring and Reporting. 2021. https://www.who.int/teams/noncommunicable-diseases/surveillance. Accessed 10 Aug 2022.
- 26.World World Health Organization. Global school-based student health survey (GSHS). 2018. https://iris.who.int/bitstream/handle/10665/275390/GSHS.pdf?sequence=2018. Accessed 10 Aug 2022.
- 27.Ismail S, Odland ML, Malik A, Weldegiorgis M, Newbigging K, Peden M, Woodward M, Davies J. The relationship between psychosocial circumstances and injuries in adolescents: An analysis of 87,269 individuals from 26 countries using the Global School-based Student Health Survey. PLoS Med. 2021;18:1–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.World Health Organization (2023) Global school-based student health survey. https://www.who.int/teams/noncommunicable-diseases/surveillance/systems-tools/global-school-based-student-health-survey.
- 29.Grucza RA, Plunk AD, Krauss MJ, Cavazos-Rehg PA, Deak J, Gebhardt K, Chaloupka FJ, Bierut LJ. Probing the Smoking-Suicide Association: Do Smoking Policy Interventions Affect Suicide Risk? Nicotine Tob Res. 2014;16:1487–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Onis de, Mercedes,. Development of a WHO growth reference for school-aged children and adolescents. Bull World Health Organ. 2007;85:660–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Khan A, Lee E-Y, Rosenbaum S, Khan SR, Tremblay MS. Dose-dependent and joint associations between screen time, physical activity, and mental wellbeing in adolescents: an international observational study. Lancet Child Adolesc Health. 2021;5:729–38. [DOI] [PubMed] [Google Scholar]
- 32.Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2013. 10.4324/9780203771587. [Google Scholar]
- 33.Altman DG. Statistics Notes: Interaction revisited: the difference between two estimates. BMJ. 2003;326:219–219. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Lee J, Kim TH, Min S, Kim MH, Park KC, Moon JS, Ahn JS. Depressive symptoms and suicidal behaviours in adolescent non-daily smokers compared to daily smokers and never-smokers in Korea: National cross-sectional study. PLoS ONE. 2018;13:e0207182. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Kim SH, Jeong SH, Park EC, Jang SI. Association of Cigarette Type Initially Smoked with Suicidal Behaviors among Adolescents in Korea from 2015 to 2018. JAMA Netw Open. 2021;4:1–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Breet E, Goldstone D, Bantjes J. Substance use and suicidal ideation and behaviour in low- and middle-income countries: a systematic review. BMC Public Health. 2018;18:549. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Wang PW, Yen CF. Adolescent substance use behavior and suicidal behavior for boys and girls: a cross-sectional study by latent analysis approach. BMC Psychiatry. 2017;17:392. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Abio A, Owusu PN, Posti JP, Bärnighausen T, Shaikh MA, Shankar V, Lowery Wilson M. Cross-national examination of adolescent suicidal behavior: a pooled and multi-level analysis of 193,484 students from 53 LMIC countries. Soc Psychiatry Psychiatr Epidemiol. 2022;57:1603–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Jung M. The relationship between alcohol abuse and suicide risk according to smoking status: A cross-sectional study. J Affect Disord. 2019;244:164–70. [DOI] [PubMed] [Google Scholar]
- 40.Miranda-Mendizabal A, Castellví P, Parés-Badell O, et al. Gender differences in suicidal behavior in adolescents and young adults: systematic review and meta-analysis of longitudinal studies. Int J Public Health. 2019;64:265–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Jafari A, Mahdizadeh M, Peyman N, Gholian-Aval M, Tehrani H. Exploration the role of social, cultural and environmental factors in tendency of female adolescents to smoking based on the qualitative content analysis. BMC Womens Health. 2022;22:38. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Torres OV, O’Dell LE. Stress is a principal factor that promotes tobacco use in females. Prog Neuropsychopharmacol Biol Psychiatry. 2016;65:260–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Peltier MR, McKee SA. The role of sex hormones in targeting stress-induced tobacco craving, stress-reactivity, and smoking with guanfacine among women who smoke. Addiction Neuroscience. 2023;7:100084. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Swann AC, Graham DP, Wilkinson AV, Kosten TR. Nicotine Inhalation and Suicide: Clinical Correlates and Behavioral Mechanisms. Am J Addict. 2021;30:316–29. [DOI] [PubMed] [Google Scholar]
- 45.Edwards AC, Ohlsson H, Sundquist J, Sundquist K, Kendler KS. Alcohol Use Disorder and Risk of Suicide in a Swedish Population-Based Cohort. Am J Psychiatry. 2020;177:627–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.United Nations Goal 3 Ensure healthy lives and promote well-being for all at all ages. https://sdgs.un.org/goals/goal3#targets_and_indicators.
- 47.World Health Organization. WHO report on the global tobacco epidemic, 2023: protect people from tobacco smoke. Geneva: World Health Organization; 2023.
- 48.Pesko MF, Hughes JM, Faisal FS. The influence of electronic cigarette age purchasing restrictions on adolescent tobacco and marijuana use. Prev Med (Baltim). 2016;87:207–12. [DOI] [PubMed] [Google Scholar]
- 49.Camenga DR, Kong G, Cavallo DA, Liss A, Hyland A, Delmerico J, Cummings KM, Krishnan-Sarin S. Alternate Tobacco Product and Drug Use Among Adolescents Who Use Electronic Cigarettes, Cigarettes Only, and Never Smokers. J Adolesc Health. 2014;55:588–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Sugarman DE, Carey KB. The relationship between drinking control strategies and college student alcohol use. Psychol Addict Behav. 2007;21:338. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Marahatta K, Samuel R, Sharma P, Dixit L, Shrestha B. Suicide burden and prevention in Nepal: The need for a national strategy. WHO South East Asia J Public Health. 2017;6:45. [DOI] [PubMed] [Google Scholar]
- 52.Sultana M, Gow J, Mosharaf P, Rahman H, Koly KN, Rahman MA, Huq N, Rawal LB, Kumar M, Mahumud RA. Parental role and peer support in adolescent suicidal behavior in eight South-East Asian countries. J Affect Disord. 2023;332:221–30. [DOI] [PubMed] [Google Scholar]
- 53.Jegannathan B, Kullgren G. Gender differences in suicidal expressions and their determinants among young people in Cambodia, a post-conflict country*. BMC Psychiatry. 2011;11:47. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54.Chentsova VO, Bravo AJ, Pilatti A, Pautassi RM, Mezquita L, Hogarth L, Team C-CAS. Age of First Use, Age of Habitual Use, and Problematic Alcohol Use: a Cross-cultural Examination Among Young Adults in Seven Countries. Int J Ment Health Addict. 2023. 10.1007/s11469-023-01016-y. [Google Scholar]
- 55.Otu A, Charles CH, Yaya S. Mental health and psychosocial well-being during the COVID-19 pandemic: the invisible elephant in the room. Int J Ment Health Syst. 2020. 10.1186/S13033-020-00371-W. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Additional File 1: Tables S1-S7. Table S1. Questionnaire details and definitions of exposures and outcomes. Table S2. Legal age restrictions on tobacco purchase among included countries. Table S3. Legal age restrictions on alcohol purchase among included countries. Table S4. Sample size of each country and the prevalence of suicide ideation, plan, and attempt. Table S5. The prevalence of tobacco use with or without legal purchase age. Table S6. The prevalence of alcohol use with or without legal purchase age. Table S7. Interactive associations between tobacco use and alcohol use, and suicide ideation, plan, and attempt among different frequency use. Figures S1-S4. Fig. S1. Prevalence of tobacco use and alcohol use among adolescents in 58 countries between 2013 and 2021. Fig. S2. Meta-analysis of tobacco use, alcohol use, and suicide ideation, plan, and attempt among overall adolescents in 58 countries. Fig. S3. Meta-analysis of tobacco use, alcohol use, and suicide ideation, plan, and attempt among boys in 58 countries. Fig. S4. Meta-analysis of tobacco use, alcohol use, and suicide ideation, plan, and attempt among girls in 58 countries.
Data Availability Statement
No datasets were generated or analysed during the current study.