Our study has two main components:

We conducted a nationwide web-based survey targeting 1500 adults aged between 20 and 74 years. Our research sampling design was based on a stratified multistage sampling design using population data categorized by region, age, and sex provided by the National Statistical Office of Korea, thereby ensuring that survey participants represented the entire national population. Those who clicked on the link received an explanation about the study’s purpose.

Adequate explanations were provided to all participants, and informed consent was obtained before they participated in the survey. Among those who were invited, 21.8% of the participants consented to the research, resulting in a total of 1500 responses that were used for analysis.

The sociodemographic factors of survey participants collected in the study included age, gender, education level, household income, marital status, and residential area. Using a structured questionnaire, the participants’ frequency of alcohol consumption, the quantity consumed per drinking session, and the frequency of exposure to drinking alcohol scenes in social media, including press and web-based platforms, were assessed over the previous year. Additionally, to assess opinions on appropriate levels of restrictions on alcohol-related social media, we posed the following question: “Do you believe that alcohol scenes on social media platforms such as YouTube should be restricted? If you think it’s necessary, to what extent do you think it should be limited?” The response comprised the following options: (1) do not believe any restrictions on viewing are necessary; (2) limitations to a warning level cautioning about scenes featuring alcohol before viewing; (3) restriction for viewing only upon adult certification; (4) restrictions such as pixelation even with adult certification; and (5) restrictions prohibiting the inclusion of drinking alcohol scenes in videos.

Our target sample size was 1500 participants, which would allow for the description of responses with a 2.5% margin of error at a 95% confidence level. A descriptive analysis of the participants’ baseline characteristics as well as each item in the questionnaire was conducted, using frequencies and percentages. In addition, the study analyzed the relationship between sociodemographic factors and the frequency of exposure to scenes depicting alcohol consumption on social media platforms. It also conducted a regression analysis on perceptions of appropriate restriction levels for drinking scenes in social media and sociodemographic factors.

Using YouTube, the keywords “sulbang (alcohol-drinking shows),” “alcohol,” “drinking broadcast,” and “drinking” were searched. Results were sorted using YouTube’s default search option of “relevance-based ranking.” An analysis by an internet search engine revealed that over 90% of users typically select from the first 3 pages of search results, so 300 videos were chosen based on keywords collected through relevance-based rankings. A total of 318 videos were analyzed after excluding unrelated overlapping videos (eg, news) and videos with less than 10,000 views. Permission from YouTube was not required to conduct this study, as all the data used in our paper was publicly available, and no special access was needed for data collection.

Data collected for each video included video ID, upload date, views, likes, dislikes, comments, and video length. Video Power Index was calculated as ([Likes * 100 / [Likes+Dislikes]] * [Views/Day] / 100). Video content analysis was evaluated based on the Media Drinking Scene Guidelines developed by the Ministry of Health and Welfare and the Korea Health Promotion Institute.

The guidelines were first developed in 2017 and revised in 2023. They include the 12 recommendations set out below.

The final YouTube video content coding was independently analyzed by 2 trained coders. The intercoder reliability was assessed using Cohen kappa (κ=0.92), yielding a value above 0.8, which indicates a high level of agreement. The coders underwent training sessions to ensure consistency and adherence to the coding protocol.

This study was approved by the National Cancer Center Ethics Board (project number: 2023-0284-0004). Before participation, all respondents were informed about the study’s purpose, procedures, potential risks, and benefits. Informed consent was obtained online before the survey began, and participation was entirely voluntary. No compensation was provided to participants. Access to the data was restricted to authorized researchers only. To ensure privacy and confidentiality, all collected data were fully anonymized and deidentified before analysis.

Table 1 shows the relationship between sociodemographic variables and the frequency of exposure to depictions of alcohol consumption on social media platforms. Although no statistically significant variances based on gender were observed, a higher percentage of women than men reported viewing social media content depicting alcohol consumption more than 4 times a week. There were no significant differences in age or marital status, but the younger age groups tended to view more social media containing alcohol consumption. Conversely, a significant difference was observed for educational level. Specifically, the group with higher education levels had the highest proportion of individuals viewing alcohol-related content more than 4 times a week (n=60, 16.9% for up to high school graduates; n=194, 19.0% for college graduates; and n=33, 26.2% for those with graduate degrees or higher; P<.001).

Significant differences in exposure to alcohol-related content in media were observed based on income levels. An analysis focusing on the frequency of individuals who viewed social media featuring alcohol consumption more than 4 times a week indicated that the lowest income group ( US $2000 per month) exhibited a higher tendency to see content associated with drinking (P<.001). The study did not find a significant association between self-reported health status and exposure to alcohol-related content on social media; however, a significant correlation was observed between smoking and alcohol consumption. Specifically, heavy drinkers were significantly more likely to view alcohol-related content on social media platforms (P<.001).

Figure 1 illustrates the opinions regarding the regulation of scenes depicting alcohol consumption on social media platforms, including web-based platforms such as YouTube. Among the respondents, 16.9% (n=253) said that viewing restrictions were unnecessary, while 29.9% (n=449) said that they believed cautionary warnings about alcohol-related scenes before viewing were essential. Additionally, a significant proportion (376/1500, 25.1%) felt that restrictions requiring adult certification for viewing were necessary, while 8.3% (n=124) of respondents advocated for stricter measures, such as pixelation, even with adult certification. Finally, 19.9% (n=298) of respondents supported restrictions that would prohibit the inclusion of alcohol consumption scenes in videos. Notably, women expressed a stronger need for restrictions compared with men, with 22.2% (n=164) of women believing that restrictions on drinking videos were necessary, compared with 17.6% (n=134) of men. The data also revealed that 21.4% (n=163) of men and 12.2% (n=90) of women believed that there should be no restrictions on drinking videos, further indicating that women were more in favor of restrictions on such videos.

Table 2 shows the results of an analysis conducted on the factors influencing perceptions of restrictions on drinking scenes. The analysis considered demographic factors such as age and gender, as well as the association between the amount of alcohol consumed and the level of exposure to drinking scenes on social media with perceptions of restrictions on drinking scenes. The perception of the level of restrictions on drinking scenes in social media was found to be positively associated with age, with a coefficient of 0.21984, and a P value of <.001. Accordingly, the analysis confirmed that older age groups believe that scenes depicting drinking on social media should have greater restrictions.

Regarding gender differences, women were significantly more likely than men to advocate for stricter limitations on depictions of drinking on social media (coefficient: 0.37827; P<.001), although no significant variances were observed for education level, marital status, household income, or self-reported health status.

However, in relation to the quantity of alcohol consumed, there was a stronger tendency to support less stringent restrictions on drinking scenes as alcohol consumption increased (coefficient: −0.3652; P<.001). The exposure level to drinking scenes showed the strongest correlation (coefficient: 1.0278; P<.001), with individuals who were more frequently exposed to drinking scenes tending to hold stronger opinions regarding the necessity for increased restrictions on these scenes.

The analysis encompassed 318 YouTube videos, with a Like ratio of 97.9% (range: 32.7‐100.0), indicating a prevalent positive reception among viewers. The median Video Power Index value stood at 939.6 (ranging from 10.4 to 84,821.7; Table 3). Examination of alcohol-related content within drinking scenes, in accordance with the Media Drinking Scene Guidelines, revealed that approximately 89.0% (n=283) of the content glorified drinking (Table 4). Videos showing positive portrayals often suggest that alcohol helps with stress relief and can rejuvenate a fatigued body after just one drink. Videos featuring celebrities or notable figures consuming alcohol constituted 27.7% (n=88) of the total, while 92.8% (n=295) depicted either binge drinking or drunkenness. Videos that undermined the autonomy of others and promoted drinking accounted for 15.7% (n=50) of the total, with 42.8% (n=136) depicting a distorted drinking culture, often emphasizing high alcohol tolerance or portraying alcohol as a catalyst for sexual advances. Conversely, just 0.6% (n=2) of the videos were age-restricted to viewers younger than 19 years, and only 31.1% (n=98) included warnings.

Alcohol consumption is a major global health risk factor and social burden [26,27], yet it has been positively perceived as a means of stress relief and fostering social bonds in everyday life [14]. This perception has normalized alcohol use across various cultures, and exposure to drinking scenes has increased, particularly due to the rise in social media platform usage, such as YouTube, following the COVID-19 pandemic [28,29]. In South Korea, based on previous studies that suggested repeated exposure to drinking scenes could lead to the normalization of alcohol use and the glorification of its effects, 10 media drinking scene guidelines were introduced in 2017 to regulate exposure to drinking scenes in media. This effort was further enhanced in 2023 with updates that included age restrictions and warning labels, especially considering the pronounced impact on adolescents and young adults [21,22]. However, these guidelines remain legally unenforceable.

This study aims to emphasize the importance of implementing practical regulatory measures for social media platforms by analyzing sulbang (alcohol-themed broadcasting) content on YouTube and examining public opinion on regulating alcohol-related broadcasts on social media. The study explores the correlations between sociodemographic factors, frequency of exposure to drinking scenes on YouTube, and the global scope of regulations on such scenes. It also evaluates whether the increased sulbang content on YouTube aligns with the media drinking scene guidelines. The findings reveal that women, individuals with higher education levels, or those with lower incomes were more frequently exposed to drinking content. Although there was no significant difference between age groups, younger individuals tended to watch more alcohol-related social media content. Additionally, individuals with higher alcohol consumption reported greater engagement in alcohol-related content on social media platforms. Regarding opinions on regulating YouTube drinking content, women and older adults showed a stronger tendency to support stricter regulations. There was also a positive correlation between exposure to drinking scenes and belief in the necessity of regulation (greater exposure was associated with stronger support for regulation) and a negative correlation between alcohol consumption and belief in the necessity of regulation (higher consumption was associated with lower support for regulation). This may suggest that heavy drinkers perceive drinking as more normalized due to repeated exposure, which aligns with Bandura’s social learning theory, explaining how observed behaviors can shape perceptions and habits [17,18]. Finally, an evaluation of YouTube (sulbang) content based on the media drinking scene guidelines showed that many videos portrayed alcohol consumption positively or emphasized binge drinking and distorted drinking cultures. However, relatively few videos included age restrictions or warning labels. These findings demonstrate a lack of enforcement mechanisms in web-based media regulations and highlight the need for future legally binding regulations.

Previous studies have widely demonstrated that exposure to drinking scenes in media increases positive perceptions of alcohol consumption and drinking behavior [30-32]. This study aligns with those findings, particularly emphasizing the tendency of binge drinkers to watch alcohol-related content more frequently and to have lower awareness of the need for regulation. This suggests that repeated exposure may reinforce positive attitudes toward drinking and lead to resistance against regulation [33]. According to 2021 high-risk drinking statistics in South Korea, while the high-risk drinking rate among men has decreased, it has increased among women, particularly among men aged 40-50 years and women aged 20-30 years [34]. This supports previous findings that women are more emotionally influenced by drinking scenes in media [35,36]. These results underline the need for tailored regulatory and educational strategies targeting specific demographic groups.

Currently, most countries enforce offline alcohol regulations, such as taxation and age restrictions on sales, but regulations for alcohol-related content on social media remain nonbinding recommendations. Considering the negative impacts of exposure to drinking scenes and the necessity of regulation, strong legal measures, such as age restrictions, broadcasting time regulations, and warning messages before broadcasts, should be swiftly implemented to effectively control the exposure to alcohol content on social media.

This study has several limitations. First, the analysis focused solely on YouTube as the research platform. However, YouTube is the most widely used video-sharing platform worldwide and records the highest viewership among social media platforms. Moreover, it provides videos of sufficient length to analyze drinking behaviors. Therefore, analyzing only YouTube allowed for the collection of a sufficiently large data set. Second, there is no validated psychometric tool specifically designed to assess exposure to alcohol-related content and regulatory attitudes. As a result, a separate questionnaire was designed for this study. Although the survey instrument was not a previously validated tool, considerable effort was made to ensure the accuracy and reliability of the study results. To improve representativeness, a stratified sampling method was adopted based on region, age, and gender. Additionally, structured response options and clear definitions of key concepts were provided to minimize ambiguity and enhance response reliability. Another limitation is that this study indirectly assessed exposure to drinking scenes through a survey rather than directly measuring actual exposure. To address this, the survey items related to alcohol consumption were carefully structured to be as specific as possible. Finally, since this study is a cross-sectional study, it has limitations in explaining causal relationships. These limitations have been explicitly acknowledged in the conclusion, and future research should use data analytics and experimental methodologies to achieve more objective and direct evaluations.

This study aimed to analyze sulbang (alcohol-themed broadcasting) content on YouTube and investigate public opinion regarding the regulation of alcohol-related broadcasts on social media platforms like YouTube. The findings indicated that individuals with higher tendencies toward alcohol consumption were more likely to frequently watch sulbang content and generally opposed regulatory measures. However, the majority of survey participants agreed on the necessity of regulating (sulbang) content. Additionally, an evaluation of sulbang content based on media drinking scene guidelines showed that most videos portrayed alcohol consumption positively or included binge drinking, while very few videos featured age restrictions or warning labels. Considering the potential influence of such content in shaping attitudes toward alcohol consumption and the possible impacts of increased exposure, it is essential to consider regulatory approaches for alcohol-related content on social media.

This study highlights the potential association between positively portrayed alcohol consumption content and public attitudes but cannot confirm causal relationships due to its reliance on cross-sectional data. To address this limitation, further research using longitudinal or experimental methodologies is needed. Such studies would play a crucial role in assessing the long-term effects of exposure and evaluating the effectiveness of proposed regulatory measures.