Parental and peer influences on adolescent alcohol use in China: Findings from a national study
Year: 2025
Type: article
Study design: Observational Study
Abstract: To the Editor: Alcohol consumption is a critical public health challenge in China, impeding progress toward achieving both the Sustainable Development Goals and the Healthy China 2030 initiative.[1] The Global Burden of Disease Study 2021 identifies alcohol as China’s 10th leading risk factor (6th for males) in terms of disability-adjusted life years and the primary cause of death among adolescents.[2] Given that a substantial proportion of Chinese adolescents engage in alcohol consumption, investigating the risk factors underlying this behavior is essential. Using nationally representative data, this study examines how parental and peer behavior and perspectives influent adolescent drinking behaviors. A multi-stage, stratified cluster random sampling method was employed to select participants from seven major regions of Chinese mainland in 2021. One province was randomly selected from each region. Within each province, probability proportional to size sampling (PPS) was used to select five districts and five counties. In each district or county, the PPS method was used to select three middle schools, two academic high schools, and one vocational high school. From each school, one class per grade was randomly chosen. All students in the selected classes participated in a self-administered survey. Detailed methodological procedures have been published previously.[3] The Chinese Center for Disease Control and Prevention Review Board approved the survey (No. 202115). Both school administrators and participants provided consent. In total, 57,336 out of 59,748 eligible students completed the survey, with a 96.0% response rate. The status of alcohol drinking and experiences of drunkenness within the past month, the past year, and over a lifetime were assessed. Alcohol drinking was defined as having consumed at least one alcoholic beverage during a specific timeframe. One standard drink was quantified as equivalent to half a bottle or can of beer, a small glass of white wine, a glass of wine, or a glass of rice wine. Drunkenness was defined as experiencing symptoms resulting from excessive alcohol consumption, including confusion, slurred speech, vomiting, and loss of consciousness, among others. Parental alcohol use was evaluated by the question “Do your father or mother frequently consume alcohol?” (neither/ mother only/ father only/ both). Parental attitudes favorable toward drinking was evaluated by the question “Do your parents allow you to drink alcohol?” (neither allows/ mother allows/ father allows/ both allow). Peer alcohol use was assessed with the question: “Do any of your close friends consume alcohol?” (none/some/most/all). School health education was assessed by asking, “In the past 12 months, have you learned in class that drinking alcohol is harmful to health?” (no/yes). Current cigarette smoking was defined as any cigarettes use within the past 30 days. Similarly, current e-cigarette use referred to any electronic cigarette use within the last 30 days. Participants were asked whether they recalled encountering pro-alcohol advertisements or promotional materials in the past 30 days across television, outdoor billboards, newspapers/magazines, the Internet, and sports events/community events/concerts/fairs. Exposure to at least one promotional format classified participants as exposed, while those reporting zero recall were categorized as unexposed. Anti-alcohol messaging exposure was assessed by asking “In the past 30 days, have you heard or seen messages about the harmful effects of drinking alcohol on health on TV, radio, the Internet, outdoor billboards, posters, newspapers, magazines, or in movies?” (no/yes). Underage purchase was assessed using the question “In the past 30 days, have you been refused to buy alcohol due to your age?” (refused/ not refused/ didn’t buy alcohol in the past 30 days). For complex sampling, a weighted approach was adopted to compute estimates and their 95% confidence intervals (CIs). The study examined associations between adolescent alcohol drinking and drunkenness (lifetime, past year, and past month) and various influencing factors. Multilevel logistic regression analysis was applied, with Level 1 represented individual students, while Level 2 represented schools. Two analytical models were formulated for each relationship: a basic model and a comprehensive model. Model 1 served as the basic model, analyzing parental or peer influences on adolescent alcohol use while controlling for fundamental participant characteristics, including sex, urban-rural residence, geographical location, school type, cigarette use, and e-cigarette use. Model 2 expanded this foundation by incorporating additional variables encompassing parental influences (maternal and paternal drinking patterns and attitudes), peer drinking behaviors, exposure to pro-alcohol and anti-alcohol advertising, school-based health education, and underage alcohol purchasing experiences. Both unadjusted odds ratios (ORs) and adjusted odds ratios (AORs) were calculated. A P-value <0.05 (two-tailed) denoted statistical significance. SAS version 9.4 (Cary, NC, USA), specifically the SURVEYFREQ and GLIMMIX procedures, was used. In 2021, among Chinese adolescents, the estimated percentages were 44.1% (lifetime drinking), 32.7% (past-year drinking), and 11.2% (past-month drinking); corresponding percentages for drunkenness were 11.9%, 5.9%, and 1.6%. Notably, boys reported higher percentages of both drinking and drunkenness than girls. Among students, 67.1% from families with neither parent drank frequently, 28.5% had a father who drank frequently, 1.0% had a mother who drank frequently, and 3.3% had both parents drank frequently [Supplementary Table 1, https://links.lww.com/CM9/C661]. Adolescents whose parents abstained from frequent drinking had a lower percentage of alcohol use (lifetime drinking: 39.6%). The percentage was significantly higher when either the father (55.9%) or mother (56.3%) drank frequently, reaching 72.8% when both parents drank frequently. Similar escalating patterns emerged for past-year and past-month drinking behaviors. The basic model (Mode 1) showed that frequent parental alcohol consumption, regardless of whether it was by the father, mother, or both, was positively associated with adolescent past-year and past-month drinking. Additionally, having only the father as a frequent drinker was positively linked to the adolescents’ lifetime drinking. Comprehensive analysis (Model 2) showed a significant AOR increase for alcohol consumption among adolescents with frequently drinking fathers: AOR 1.3 (1.0–1.8) for past-year drinking and AOR 1.4 (1.1–1.7) for past-month drinking [Supplementary Table 2, https://links.lww.com/CM9/C661]. Compared to adolescents with parents who did not frequently drink (lifetime drunkenness: 10.0%), the prevalence of drunkenness increased to 16.3% when only the father drank frequently, 22.1% when only the mother drank frequently, and 31.0% when both parents drank frequently. Past-year and past-month percentages of drunkenness also increased among adolescents whose parents engaged in frequent drinking. Both Model 1 and Model 2 showed that having either a father or mother who frequently drank increased the likelihood of lifetime drunkenness among adolescents. Model 2 showed an AOR of 1.3 (1.1–1.6) for having a father who frequently drank and an AOR of 4.6 (1.0–20.5) for having a mother who frequently drank [Supplementary Table 3, https://links.lww.com/CM9/C661]. Regarding parental attitudes toward adolescent drinking, 71.1% of students reported that neither parent allowed them to drink, while 21.9% indicated both parents permitted it. High school students were more likely than middle school students to have both parents permit drinking (30.5% vs. 15.4%). Single-parent consent was uncommon, with 5.4% involving only the father’s permission and 1.5% only the mother’s. For lifetime drinking prevalence among adolescents, compared to those whose parents did not allow offspring drinking (27.2%), percentages surged when any parent permitted it: father only (85.7%), mother only (88.0%), or both parents (85.6%). Past-year and past-month drinking showed similar pattern. Both Model 1 and Model 2 revealed that parental favorable attitudes toward offspring drinking, regardless of whether it was from the father, mother, or both, were positively associated with adolescent alcohol drinking across all timeframes. Model 2 demonstrated significant increases in AORs for lifetime drinking among adolescents when only the father allowed drinking (24.6 [4.2–142.5]), only the mother allowed it (21.8 [8.5–56.2]), and both parents (15.3 [8.5–27.5]). A similar trend was observed for past-year and past-month drinking [Supplementary Table 2, https://links.lww.com/CM9/C661]. Regarding lifetime drunkenness among adolescents, compared to those whose parents did not allow offspring drinking (7.7%), percentages were higher when only the father allowed drinking (25.3%), only the mother allowed drinking (27.1%), or both parents allowed drinking (22.4%). This pattern remained consistent for past-year and past-month drunkenness episodes. Model 1 demonstrated that parental permissive attitudes toward offspring drinking were generally associated with increased adolescent drunkenness across all timeframes, with the exception that only mother’s allowance showed no significant association with past-month drunkenness. Model 2’s AORs only revealed associations between mother’s allowance and adolescent lifetime drunkenness (3.2 [1.6–6.5]), past-year drunkenness (3.2 [1.4–7.7]), as well as both parents’ allowance and adolescent past-month drunkenness (2.2 [1.2–3.9]) [Supplementary Table 3, https://links.lww.com/CM9/C661]. Peer drinking patterns varied considerably among participants: 44.9% reported no close friends who consumed alcohol, 43.1% indicated some friends drank, and 12.0% reported that most/all close friends engaged in drinking. Compared to middle school students, the proportion of high school students with some drinking close friends (35.1% vs. 53.7%) and most/all close friends who drank (8.4% vs. 16.9%) was notably higher. Similarly, among academic school students, the percentage of those with most/all close friends who drank (14.8%) was lower than vocational high school students (21.7%) [Supplementary Table 1, https://links.lww.com/CM9/C661]. Compared to adolescents with no close friends who drank (16.2%), the prevalence of lifetime drinking rose steadily to 61.3% for those with some close friends who drank, and escalated to 86.0% for adolescents with most/all close friends who drank. A similar trend was observed for both past-year drinking and past-month drinking. Both Model 1 and Model 2 showed a positive correlation between some or most/all close friends who drank and adolescent drinking across all timeframes. AORs of Model 2 were lifetime (3.8 [3.4–4.3] for some and 6.9 [4.3–10.9] for most/all), past-year (3.6 [3.0–4.4] for some and 7.6 [3.5–16.3] for most/all), and past-month (5.3 [2.0–14.1] for some and 6.0 [3.0–11.7] for most/all) [Supplementary Table 2, https://links.lww.com/CM9/C661]. Compared to adolescents with no close friends who drank (3.3%), the prevalence of lifetime drunkenness increased progressively, rising to 15.1% among those with some close friends who drank, and escalating further to 34.7% among adolescents with most/all close friends who drank. This pattern was consistently observed for both past-year and past-month drunkenness episodes. Model 1 revealed a positive correlation between adolescent drunkenness and having some or most/all close friends who drank, across all timeframes. Model 2 showed AORs of 3.8 (1.9–7.8) and 7.7 (1.7–34.6) for lifetime drunkenness, and 3.5 (1.4–9.0) and 5.4 (2.3–12.9) for past-year drunkenness, respectively, among adolescents with some or most/all drinking close friends [Supplementary Table 3, https://links.lww.com/CM9/C661]. Using nationwide survey data, this study explored how parents and close friends affect adolescent drinking, marking the first use of nationally representative data to evaluate these relationships in China. We found a strong connection between parents’ attitudes towards their children’s drinking and adolescent alcohol use, while the correlation with parents’ own drinking habits is relatively weaker. For instance, adolescents with both permissive parents exhibited substantially elevated risks, with approximately 15-, 10-, and 4-fold increases in lifetime, past-year, and past-month alcohol drinking, respectively. In contrast, adolescents with both parents who frequently drank did not demonstrate significant associations with adolescent drinking across any timeframe. These findings suggest that changing parents’ attitudes may be more effective than altering parental drinking habits. Furthermore, we observed that high school students were more likely to receive permission from both parents to drink compared to middle school students, highlighting the need to help parents of older students maintain restrictive attitudes and effective supervision of their children’s drinking behaviors. To effectively address adolescent alcohol consumption, we must place the family at the heart of intervention initiatives, which has been proved to be effective.[4] By reducing permissive attitudes and strengthening parental supervision, adolescents can be guided toward reduced alcohol drinking. As the drinking among close friends escalates, adolescent alcohol use intensifies, highlighting the influence of peers. It is imperative to guide adolescents toward peer groups with lower drinking prevalence, strengthen alcohol harm reduction education in schools, and promote beneficial social activities. This study had certain limitations. First, the data derived exclusively from in-school adolescent students via self-reporting, excluding school dropouts. Second, causality cannot be inferred because it is a cross-sectional survey. Although self-reported substance use may be influenced by social desirability bias, adolescent alcohol consumption in China has broader social acceptability than strictly prohibited substances such as marijuana likely minimizes this potential bias. In conclusion, parental alcohol use, permissive parental attitudes, and peer alcohol use were linked to adolescent alcohol use. Interventions should focus on family influence, prioritize changing parental attitudes over addressing parental drinking habits, and guide adolescents toward peer groups with lower drinking prevalence. Funding This study was supported by a grant from the National Key Research and Development Program of China (No. 2023YFC3605000). Conflicts of interest None.
Source: Chinese Medical Journal
Language: English
Subfield: Epidemiology
Field: Medicine
Domain: Health Sciences
Open Access status: gold
Estimated APC paid: $2,000
Award: G3262032821
OpenAlex ID: W4417209269