Associations Between Digital Alcohol Marketing and Alcohol-related Behaviours: A Systematic Review of Quantitative Studies

Abstract

Background

Digital alcohol marketing has evolved rapidly with the growth of social media. Modern digital environments are driven by user-generated content UGC relating to peer-to-peer sharing, liking, and commenting on alcohol brands, which blurs the line between commercial promotion and authentic social interaction. This systematic review synthesises recent research on the association between digital alcohol marketing and alcohol-related outcomes, while examining the dual roles of UGC and health warnings as empirical countermeasures.

Methods

Following PRISMA guidelines, eight electronic databases (PubMed, Web of Science, Scopus, JSTOR, PsycINFO, Medline, Taylor & Francis Online, ScienceDirect) were systematically searched for quantitative studies published between 2017 and late 2024. Studies were included if they investigated exposure to, and/or engagement with, alcohol marketing in digital contexts, as were studies which examined the role of user-generated messaging and health warnings in alcohol ads. Eligible outcomes were alcohol-related beliefs, attitudes, intentions, and behaviours.

Results

Eighteen quantitative studies met inclusion criteria (10 cross-sectional, 7 experimental, 1 longitudinal). Positive associations with alcohol-related outcomes were reported across all 14 studies (77.8%) assessing exposure to (33.3%, n = 6) and/or engagement with (44.4%, n = 8) digital alcohol marketing, leading to increased drinking frequency and drinking intentions. Across the 6 studies (33.3%) examining UGC and health warnings (with 2 experimental studies overlapping with engagement), emerging evidence demonstrates that while pro-drinking UGC increases ad persuasiveness and normative influence, health warnings lessen the effects of digital alcohol marketing on attitudes and consumption.

Conclusions

Quantitative evidence demonstrates a consistent link between digital alcohol marketing and alcohol-related behaviours across exposure and engagement dimensions. Alcohol control policies must address digital marketing channels, regulate user-generated promotions, and mandate prominent health warnings across online environments. Although positive associations between digital marketing and alcohol-related outcomes predominate, evidence regarding optimal warning designs and platform-specific effects remains limited. Furthermore, the literature remains dominated by US-based research, with a critical shortage of studies conducted in European settings or among older adult populations.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s44197-026-00629-5.

Introduction

Between 2010 and 2019, global per-capita alcohol intake fell from 5.7 L to 5.5 L (a 4.5% decrease), with the highest 2019 figures recorded in the WHO European Region (9.2 L) and the Americas Region (7.5 L; [1]). Despite this slight drop, and in line with global predictions on adult alcohol per-capita consumption, global consumption is expected to increase by 17%, reaching 7.6 L by 2030 [2]. Overall intake remains high, with 59.9% of European adults, 54.1% of adults in the Americas, and 53.8% in the Western Pacific Region reporting as current drinkers [1, 3]. The European region has the highest drinking rates, with more than 50% of males and females aged 15 + consuming alcohol [3].

While good health and well-being is one of the United Nation’s 13 sustainable development goals to be achieved by 2030 [4], high-volume of alcohol use has been linked to significant individual and societal harm, including a causal relationship with various cancers [1, 5]. That said, Canada’s Guidance on Alcohol and Health outlines a strict risk continuum, classifying three or more standard drinks per week as carrying an increased risk of developing several cancers [6]. Whereas the United States, in the 2025–2030 Dietary Guidelines for Americans, removed long-standing quantitative consumption thresholds in favour of a direct mandate to ‘consume less alcohol for better overall health’ [7]. While overall population consumption varies geographically, public health concerns are primarily driven by heavy episodic drinking (binge drinking) and frequent high-volume intake, both of which serve as key indicators of risky alcohol consumption [1, 3].

A central driver of early drinking initiation and risky consumption among youth is exposure to commercial alcohol marketing [8, 9]. Extensive longitudinal evidence demonstrates a causal relationship between exposure to alcohol advertising and both earlier drinking onset and higher consumption levels among underage populations [8–13]. While traditional media [14], and sport sponsorship [15] have historically dominated marketing channels, the rapid expansion of digital media has transformed how consumers encounter alcohol branding and advertising [13, 16, 17].

In digital marketing environments, commercial promotion operates alongside user-generated content UGC, defined as brand-related media, posts, comments, photos, or videos created and shared by individual internet users rather than brand owners [18]. UGC plays a central role in contemporary alcohol marketing by blurring the boundaries between corporate advertising and authentic peer-to-peer social interaction [19]. When users actively participate with brand content (such as liking, sharing, or creating posts about alcohol products), they amplify commercial reach across peer networks. This interactive co-creation of content alters perceived social norms, fostering positive drinking attitudes and normalising alcohol consumption within digital spaces [20]. Given this, the integration of platform algorithms, targeted advertising, and personalised content has advanced the complexity and sophistication of digital marketing [21].

To date, the primary baseline for digital alcohol marketing reviews was established by Noel et al. [13], who evaluated 25 studies published prior to 2017. Noel et al. [13] found that active engagement (such as liking and commenting) is consistently linked to alcohol consumption in both short- and long-term studies, while the effects of passive scrolling remain unclear and somewhat inconclusive. As current research is limited, Noel et al. [13]stressed the need for deeper investigation into these user interactions. Furthermore, the authors pointed out that digital advertising changes incredibly fast. Therefore, they recommended running more long-term studies to see if their early results actually hold up over time. Building upon this, the present systematic review synthesises quantitative evidence published between 2017 and September 2024. Our objectives are twofold: i) to evaluate current quantitative evidence on exposure to and engagement with modern paid and user-generated digital alcohol marketing and (ii) to assess the efficacy of embedded health warnings as potential public health countermeasures. While health warnings on product labels are well-studied, studies examining warnings within alcohol ads remain scarce—an area gaining increasing attention in both academic research [22–24] and policy discussions [25, 26].

Therefore, our overarching aim was to provide an up-to-date systematic review of the literature and in doing so answering the following research questions.

1. What is the current quantitative research evidence on exposure to and engagement with [paid] digital alcohol marketing on individuals’ alcohol use, perceptions, and behaviours across different age groups (e.g. adolescents, young adults, and older adults)?

The main research question is supported by the following sub-questions.

• 2.What is the impact of UGC in conjunction of [paid] digital alcohol marketing on alcohol-related behaviours?
• 3.What is the impact of health warnings when embedded in digital alcohol ads on alcohol-related behaviours?

Methods

Eligibility Criteria

Our systematic review is confined to studies within the domain of consumer research. Unlike econometric research that quantifies population-level behaviours based on advertising expenditures, consumer studies investigate the influence of marketing communications on individual attitudes, drinking intentions, and behaviours. Primary studies using longitudinal, cross-sectional, and experimental methodologies examining the impact of digital alcohol marketing on drinking behaviours were considered for inclusion; primary studies using content analyses or qualitative methodologies were excluded.

We considered a wide range of digital alcohol marketing, including social media channels (Facebook, YouTube, TikTok), branded apps, emails, websites, games, and banner ads [27–29]. Following the approach of previous systematic reviews [13], the variables of interest were exposure to or engagement with paid, owned, and earned digital alcohol marketing as independent variables, and intentions to drink, desire to drink, attitudes toward alcohol, perceptions, beliefs, purchase intentions, and consumption behaviours as dependent variables (outcomes). There were no exclusion criteria based on age demographics.

Studies that only reported exposure to earn media (in isolation - e.g., UGC, peer influence, blogs, influencers; and family use of digital platforms were excluded, as were studies that only examined exposure to traditional media channels, or a combination of both. Studies were also excluded if they examined social media without assessing exposure to digital alcohol marketing or if they focused on social media use more broadly.

Search Strategy

Eight scientific databases were searched: PubMed, Web of Science, Scopus, Jstor, PsycINFO, Medline, Tandfonline, and ScienceDirect up to September 2024. Keywords and set of terms used were based on prior research and validated by our team. The search strategy combined the following sets of terms: “*exposure and engagement,” “*digital alcohol marketing,” “*social media alcohol advertising,” “*drinking behaviour,” “*alcohol consumption” *and/or “intentions, behaviours, attitudes, binge drinking”. Table 1 provides a comprehensive record of the exact search syntax, string combinations, and boolean operators (AND, OR) deployed across all targeted databases.

Table 1.

Systematic Search Strategy and Output Equations

Source	Search Strategy/Syntax
PubMed	(((exposure AND to) OR (engagement AND with))) AND summary((social AND media AND alcohol AND marketing)) OR summary(((facebook) OR (instagram) OR (twitter) OR (tiktok) OR (snapchat))) AND summary(((binge AND drinking) OR (drinking AND intentions) OR (alcohol AND misuse) OR (attitudes))) AND summary((web AND based AND alcohol AND advertising)) OR summary((online AND alcohol AND advertisements)) OR summary((internet AND alcohol AND marketing)) Filters: from 2015–2024
Web of Science	(TS=((exposure AND to) OR (engagement AND with)) AND TS=(social AND media AND alcohol AND marketing) OR TS=((facebook) OR (instagram) OR (twitter) OR (tiktok) OR (snapchat)) AND TS=((binge AND drinking) OR (drinking AND intentions) OR (alcohol AND misuse) OR (attitudes)) AND TS=(web AND based AND alcohol AND advertising) OR TS=(online AND alcohol AND advertisements) OR TS=(internet AND alcohol AND marketing)) AND PY=(2015–2024)
Scopus	(TITLE-ABS-KEY((exposure AND to) OR (engagement AND with)) AND TITLE-ABS-KEY(social AND media AND alcohol AND marketing) OR TITLE-ABS-KEY((facebook) OR (instagram) OR (twitter) OR (tiktok) OR (snapchat)) AND TITLE-ABS-KEY((binge AND drinking) OR (drinking AND intentions) OR (alcohol AND misuse) OR (attitudes)) AND TITLE-ABS-KEY((web AND based AND alcohol AND advertising) OR (online AND alcohol AND advertisements) OR (internet AND alcohol AND marketing))) AND PUBYEAR > 2014
Jstor	(“exposure to” OR “engagement with”) AND (“social media alcohol marketing”) OR (“facebook” OR “instagram” OR “twitter” OR “tiktok” OR “snapchat”) AND (“binge drinking” OR “drinking intentions” OR “alcohol misuse” OR “attitudes”) AND (“web-based alcohol advertising” OR “online alcohol advertisements” OR “internet alcohol marketing”) AND PUBYEAR: [2015 TO 2024]
PsycINFO	TI, AB((exposure AND to) OR (engagement AND with)) AND TI, AB(social AND media AND alcohol AND marketing) OR TI, AB((facebook) OR (instagram) OR (twitter) OR (tiktok) OR (snapchat)) AND TI, AB((binge AND drinking) OR (drinking AND intentions) OR (alcohol AND misuse) OR (attitudes)) AND TI, AB((web AND based AND alcohol AND advertising) OR (online AND alcohol AND advertisements) OR (internet AND alcohol AND marketing)) AND Year Published: 2015–2024
Medline	((“exposure to“[Title/Abstract] OR “engagement with“[Title/Abstract]) AND (“social media alcohol marketing“[Title/Abstract]) OR (“facebook“[Title/Abstract] OR “instagram“[Title/Abstract] OR “twitter“[Title/Abstract] OR “tiktok“[Title/Abstract] OR “snapchat“[Title/Abstract]) AND (“binge drinking“[Title/Abstract] OR “drinking intentions“[Title/Abstract] OR “alcohol misuse“[Title/Abstract] OR “attitudes“[Title/Abstract]) AND (“web-based alcohol advertising“[Title/Abstract] OR “online alcohol advertisements“[Title/Abstract] OR “internet alcohol marketing“[Title/Abstract])) AND (“2015“[Date - Publication] : “2024“[Date - Publication])
ScienceDirect	TITLE-ABSTR-KEY((exposure AND to) OR (engagement AND with)) AND TITLE-ABSTR-KEY(social AND media AND alcohol AND marketing) OR TITLE-ABSTR-KEY((facebook) OR (instagram) OR (twitter) OR (tiktok) OR (snapchat)) AND TITLE-ABSTR-KEY((binge AND drinking) OR (drinking AND intentions) OR (alcohol AND misuse) OR (attitudes)) AND TITLE-ABSTR-KEY((web AND based AND alcohol AND advertising) OR (online AND alcohol AND advertisements) OR (internet AND alcohol AND marketing)) AND YEAR(2015–2024)
Tandofline	The search strategy combined key concepts using Boolean operators. Studies referencing either “exposure to” or “engagement with”, in conjunction with “social media alcohol marketing” or specific platforms (Facebook, Instagram, Twitter, TikTok, Snapchat). We included alcohol-related outcome terms such as “binge drinking”, “drinking intentions”, “alcohol misuse”, and “attitudes”. These were further cross-referenced with digital marketing terms such as “web-based alcohol advertising”, “online alcohol advertisements”, and “internet alcohol marketing”. All terms were searched within the title and abstract fields (or equivalent summary fields, as supported by the platform). Filters were applied to restrict results to English-language articles published from 2015 through 2024

Identification of Studies

Three authors (GF, STS, and VF) conducted the screening of abstracts to identify papers that met the eligibility criteria. Initial search was finished in July 2021, with an update of the search completed in October 2024, ensuring that we have the latest evidence on the topic and update our literature review. Full texts were obtained for relevant papers and were thoroughly reviewed to confirm they met the inclusion criteria. To ensure accuracy, each author independently cross-checked original paper abstracts, which were equally screened by other members of the team (SRK, GL). The complete selection process is outlined in the PRISMA flow diagram (Fig. 1).

Fig. 1.

PRISMA Flow Diagram

Quality Assessment Methods

To assess article relevance, methodological quality, and study variability, we applied the evaluation framework established by Hawker et al. [30: Appendices A and B]. We used the Centre for Evidence-Based Management Survey Scoring System to evaluate overall study quality across various domains, including Abstract and Title, Introduction and Aims, Method and Data, Sampling, Data Analysis, Ethics and Bias, Findings/Results, Transferability/Generalisability, and Implications and Usefulness. The adapted version of the quality appraisal scale by Hawker et al. [30] evaluates nine distinct methodological domains, assigning a score of 1 (Very Poor) to 4 (Good) for each criterion, resulting in a total possible score ranging from 9 to 36. To operationalise these quality scores for evidence synthesis, studies scoring 32–36 were categorised as “High Quality”, scores from 27 to 31 were designated as “Fair/Satisfactory Quality,” and scores of 26 or below were deemed “Low Quality.” This systematic scoring allowed us to explicitly weigh the consistency of evidence against the methodological rigour of the primary studies during data interpretation.

All 18 included studies underwent quality appraisal using the Hawker et al. [30] tool. Given that all studies demonstrated moderate (33.3%, n = 6) to strong (66.7%, n = 12) quality, no studies were excluded based on quality score alone, ensuring a complete synthesis of current literature. Individual study quality ratings are presented in Table 3. The systematic review was conducted as part of a broader evidence synthesis commissioned by the European Commission and followed a predefined protocol developed for this review in particular.

Table 3.

Quality Assessment of Included Studies

Number	Studies	Criteria									Total Score (out of 36)
		1	2	3	4	5	6	7	8	9
1.	Alhabash et al. [38]	3	3	4	4	3	2	3	4	4	30 Strong Quality
2.	Chan et al. [43]	4	4	3	3	3	3	4	3	3	30 Strong Quality
3.	Chen et al. [37]	4	4	4	2	3	2	4	4	4	31 Strong Quality
4.	Critchlow et al. [33]	4	4	4	3	3	3	4	4	4	33 Strong Quality
5.	Gabrielli et al. [31]	3	4	3	2	3	2	4	4	4	26 Moderate Quality
6.	Gupta et al. [39]	3	4	4	3	3	2	4	3	3	29 Moderate Quality
7.	Lou and Alhabash [32]	3	4	4	4	3	2	4	4	3	31 Strong Quality
8.	Malechwanzi et al. [34]	3	3	3	3*1	3	4*2	3	4	3	29 Moderate Quality
9.	McClure et al. [35]	4	3	4	4	3	2	4	4	3	31 Strong Quality
10.	McCreanor et al. [40]	4	4	3	2	5	3	2	2	3	27 Moderate Quality
11.	Noel and Babor [44]	4	4	4	3	4	3	4	3	4	33 Strong Quality
12.	Noel and Lakhan [45]	4	4	4	2	3	2	4	4	4	31 Strong Quality
13.	Noel and Serna [46]	4	3	3	3	3	3	2	2	3	26 Moderate Quality
14.	Noel [47]	4	4	3	4*1	3	1*2	3	4	4	30 Strong Quality
15.	Noel et al. [48]	4	4	4	3	3	3	4	4	4	32 Strong Quality
16.	Roberson et al. [41]	3	4	4	3	3	2	4	4	4	32 Strong Quality
17.	Wang et al. [36]	3	3	4	3	3	2	4	3	4	29 Moderate Quality
18.	Yang and Zhao [42]	3	4	4	3	3	3	4	4	4	32 Strong Quality
Criteria:	1. Abstract and title		Rating Index:		4 points = Good
	2. Introduction and aims				3 points = Fair
	3. Method and data				2 points = Poor
	4. Sampling				1 point = Very Poor
	5. Data analysis
	6. Ethics and bias
	7. Findings/results
	8. Transferability/generalisability
	9. Implications and usefulness

The quality appraisal was conducted independently by two reviewers, and the total score represents the average of the two scores, where applicable. *¹ denotes a study which analysed various datasets in lieu of directly assessing human participants, necessitating an adaptive scoring. In such cases a *² denotes ethics research approval considerations as not having been assessed.

Two researchers conducted the quality assessment independently. Inter-rater reliability was then assessed using Spearman’s rank-order correlation on the reviewers’ initial total quality scores. Agreement was strong (Spearman’s p = 0.945, p < 0.001), exceeding the prespecified threshold of 0.80; no further domain-level agreement analyses were required, and any minor discrepancies in individual domain ratings were resolved through discussion to determine the final quality ratings.

Results

The initial comprehensive search identified 4,460 records. Following the screening of these titles and abstracts, 328 studies were sought for retrieval. Of these, 255 records were excluded, leaving 73 full-text articles for first-round eligibility assessment. Eleven papers met the selection criteria during this phase. An updated search conducted in 2024 yielded an additional 47 records, from which seven met the inclusion criteria. In total, 18 studies were selected for the final review (see Fig. 1).

Study Characteristics

Of the 18 studies included in this review, 11 were conducted in the United States with 1 each carried out in the UK, New Zealand, Hong Kong, Mongolia, Taiwan, Australia, India, and Africa. Included studies were published between 2017 and 2024, with sample populations ranging from adolescents (11 years) to older adults (up to 65 years), and sample sizes varying from 65 [32] to 3,399 participants [33]. See Table 2.

Table 2.

Characteristics of studies measuring exposure to and engagement with digital alcohol marketing, UGC, and health warnings in advertising

Study and Design	Theme(s)*	Sample	Location	Method	Digital media type	Follow-up	Primary outcomes
Gabrielli et al. [31] (Cross-sectional)	Exposure/Awareness	Adolescents 12–17 years (n = 202)	USA	Online survey	Internet ads before videos/movies, display/pop-up ads, brand websites, social media pages	N/A	Lifetime drinking
Malechwanzi et al. [34] (Cross-sectional)	Exposure/Awareness	University students (n = 836)	Kenya	Cross-sectional survey	Alcohol promotions via web-based media	N/A	Early liquor drinking propensity; risky drinking
McClure et al. [35] (Cross-sectional)	Exposure/Awareness	Adolescents 12–17 years (n = 202)	USA	Online survey	Display ads, alcohol commercials, brand-sponsored social media pages	N/A	Ever drinking
Wang et al.[36] (Cross-sectional)	Exposure/Awareness	Students (n = 1277)	Mongolia	Cross-sectional survey	Internet	N/A	Drinking frequency
Chen et al. [37] (Longitudinal)	Exposure/Awareness	Adolescents 13–14 years (n = 1926)	Taiwan	Longitudinal cohort survey	Internet and e-mails	13 months	Drinking frequency; intention to drink
Alhabash et al. [38] (Experimental)	Exposure/Awareness	Young adults 18–20 years (n = 111)	USA	Online experiment	Social media advertisements	N/A	Intention to drink; desire to drink; engagement with alcohol advertisement; attitude towards advertisement
Critchlow et al. [33] (Cross-sectional)	Engagement	Adolescents 11–19 years (n = 3399)	UK	Cross-sectional survey	Social media advertisements	N/A	Alcohol consumption; higher-risk drinking; susceptibility to drink
Gupta et al.[39] (Cross-sectional)	Engagement	Adolescents and young adults 13–25 years (n = 631)	India & Australia	Cross-sectional survey	Facebook, Twitter, YouTube	N/A	Lifetime drinking; drinking behaviour
McCreanor et al. [40] (Cross-sectional)	Engagement	Adolescents 14–20 years (n = 1541)	New Zealand	Cross-sectional survey	Social media use and engagement with alcohol marketing	N/A	Alcohol behaviours, online purchasing, drunkenness, hazardous drinking
Roberson et al.[41] (Cross-sectional)	Engagement	Young adults 18–22 years (n = 682)	USA	Cross-sectional survey	Social media	N/A	Drinking behaviour
Yang & Zhao [42] (Cross-sectional)	Engagement	Students (n = 609)	USA	Cross-sectional survey	Social media	N/A	Intention to binge drink
Chan et al.[43] (Cross-sectional)	UGC	Young adults 18–34 years (n = 675)	Hong Kong	Telephone survey	Social media marketing	N/A	Social media marketing as peer endorsement of drinking
Lou & Alhabash [32] (Experimental)	UGC/Health Warnings	Young adults 18–20 years (n = 65)	USA	Online experiment	Instagram advertisements	N/A	Intention to drink
Noel & Babor [44] (Expérimental)	Engagement/UGC/Health warnings	Young adults 20–24 years (n = 120)	USA	Online experiment	Facebook advertisements	N/A	Desire to drink; advertisement engagement
Noel & Lakhan[45] (Experimental)	UGC/Health warnings	Adults 18–65 years (n = 722)	USA	Online experiment	Facebook advertisements	N/A	Desire to drink
Noel & Serna [46] (Experimental)	Engagement/Health warnings	Adults 18–65 years (n = 723)	USA	Online experiment	Social media	N/A	Alcohol craving; engagement with alcohol-related content
Noel [47] (Experimental)	UGC/Health warnings	Adults 18–65 years (n = 723)	USA	Online experiment	Facebook advertisements	N/A	Intention to purchase alcohol; AUDIT
Noel et al. [48] (Experimental)	Engagement/UGC/Health warnings	Young adults (n = 120)	USA	Online experiment	Facebook advertisements	N/A	AUDIT

Quality Assessment Results

In terms of overall scoring, all but one study achieved 27 or more points out of possible 36. Only one study obtained a score of 34, while the lowest score recorded was 26 (see Table 3). The analysis highlighted strong performance in several domains, including Introduction and Aims (98%), Sampling (98%), Findings/Results (98%), and Implications and Usefulness (98%), reflecting clear findings, well-structured results, and relevant implications for future alcohol marketing policies. Weaker scores were observed in terms of Ethics and Bias (46%) and Data Analysis (78%), with several studies somewhat failing to adequately address statistical implications or researcher bias. However, all included studies provided a clear objective or research question, and consistent data collection methods, therefore all 18 studies had satisfactory quality, with high agreement between authors, confirming the reliability of the appraisal process.

When accounting for study quality (see Table 3), the core finding that exposure to, and engagement with, digital alcohol marketing increases consumption and drinking intentions, is supported by highly robust evidence, as 100% of the cross-sectional, longitudinal, and experimental studies were rated as moderate or high quality (scores 26; Table 3). However, the overall quality score was disproportionately lowered by somewhat weaker performance in Ethics and Bias (46%) and Data Analysis (78%), relating to the internal validity of the evidence base. For example, several experimental and survey designs failed to adequately report or account for potential researcher bias, sample selection biases, or the implications of statistical errors.

Evidence Synthesis

The results are structured around three thematic areas aligned with our research questions. Given that two studies investigated multiple outcomes, these are reported in more than one subsection. Studies were grouped into three themes: (i) exposure to/awareness of digital marketing (n = 6; 4 cross-sectional surveys, 1 longitudinal, 1 experimental); (ii) engagement with digital alcohol marketing (n = 8; 4 cross-sectional surveys, and 4 experimental) (iii) UGC and health warnings (n = 6; 1 cross-sectional, 5 experimental). Of the 18 unique studies included in the review, the thematic distribution is as follows (note that the cumulative percentage exceeds 100% because two studies addressed multiple themes): Exposure/awareness: (33.3%), engagement: (44.4%), and UGC and Health Warnings: (33.3%).

Exposure and Awareness to Digital Alcohol Marketing

The terms exposure to and awareness of digital alcohol marketing are often used interchangeably in the alcohol marketing literature. Surveys typically assess participants’ self-reported awareness of exposure to digital alcohol marketing, while experimental studies involve manipulations, exposing individuals to mock or real-world alcohol advertisements. The outcomes ranged from high-risk use, brand identification, desire to drink, engagement with the ad, frequency and quantity of drinking, and drinking behaviour in general (Table 2). Much of this research focused on alcohol advertising within social media networks (e.g., YouTube, Instagram, Facebook) or online alcohol marketing more broadly (without specifying a medium).

Cross-sectional studies

Four studies [31, 34–36] measured recall and exposure to alcohol ads across branded websites and social media pages [31, 35] with findings suggesting robust associations between the recall of online alcohol ads and subsequent alcohol consumption. Men reported higher exposure to online alcohol marketing [34], and that exposure to paid ads was found to be significantly higher among current drinkers [36].

Longitudinal Studies

We were able to identify only one longitudinal study [37] related to the impact of exposure to alcohol marketing, with findings suggesting a significant positive relationship between exposure to alcohol ads on websites and occasional consumption in the following year amongst adolescents aged between 12 and 14. It was also found that compared to non-drinkers in the study sample, those who consume alcohol demonstrated higher exposure to digital alcohol marketing, with more positive intentions towards drinking; thus, indicating that earlier alcohol initiation may also increase susceptibility to alcohol marketing and greater drinking intentions in the future [37].

Experimental Studies

One study [38] found that exposure to beer ads featuring younger people aged under 25, were more likely to increase self-reported intentions to drink among adolescents than ads featuring adults older than 25, thus suggesting that alcohol ads using social cues trigger social support for drinking alcohol and reinforce social norms via featuring images that promote increased consumption [38].

Engagement with Digital Alcohol Marketing

Cross-sectional studies

Current drinkers were more likely to engage with digital alcohol marketing [33] through user-generated interactions, particularly when these messages are perceived as socially accepted [42]. Recent work [40] reinforced the link between social media engagement and alcohol-related behaviours, highlighting the role of interactive marketing in shaping consumption patterns. It was also found that engagement with social media normalises consumption and leads to increased drinking behaviours [39, 41]. In terms of age group differences, young people aged between 18 and 25 are more likely to engage with digital alcohol marketing and believe that drinking among peers is common and normal than adolescents between the age of 13 and 17 [39, 41].

Experimental Studies

Three studies explored engagement with digital alcohol marketing through experimental designs [44, 46, 48]. Noel and Babor [44] found that exposure to beer advertisements with pro-drinking comments increased both the desire to drink and engagement with the ads, while active involvement with alcohol-related content on Facebook—such as liking or sharing—increased individual’s drinking intentions and alcohol consumption. These findings align with those of Noel et al. [48] who found that ad’s appeal—based on its source, information, and emotional impact was linked to both self-reported alcohol use and engagement (e.g., liking, sharing) on Facebook. Finally, user engagement in the form of likes and comments on social media posts was associated with greater desires to drink alcohol, and socially accepted messages increased user interaction, particularly among those who perceived drinking as normative [46].

UGC and Health Warnings in Alcohol Advertising

Regarding UGC and health warnings, we were able to identify 6 studies, which examined the role of these in shaping the reception of digital alcohol advertisements and associated drinking behaviours.

Cross-Sectional Studies

One cross-sectional study involving 675 Chinese residents aged 18–34 assessed past-month exposure to various forms of social media alcohol marketing, including branded posts, website banners, and UGC. The outcome was the desire to drink in relation to each type of marketing exposure, with findings suggesting that UGC was more strongly associated with the desire to drink than branded marketing content [43], highlighting the influential role of peer-shared content in shaping drinking behaviour.

Experimental Studies

Four experimental studies examined how UGC in digital alcohol ads influences drinking intentions, ad appeal, and behaviour. Noel and Babor [44] found that being exposed to beer ads with pro-drinking comments, increased desire to drink, while a follow-up study [48] reported no significant effect on self-reported alcohol use and drinking behaviour. Noel [47] and Noel and Lakhan [45], compared antidrinking, pro-drinking, and industry-sponsored messages across six ads promoting spirits, wine, and beer brands, with findings suggesting that anti-drinking UGC (e.g., comments against drinking alcohol) decreases purchase intentions [47] and pro-drinking comments (e.g., positive comments about drinking) decreased perceived behavioural control over drinking [45]. The findings indicate that the influence of UGC may vary depending on context or outcome measured.

Building on this, three of the same studies also explored different types of health warnings in digital alcohol ads including health warnings in the form of user-generated comments, self-regulated ‘drink responsibly’ messages [45, 47], and age warnings in social media [32].

Noel and Lakhan [45] measured self-reported drinking intentions after exposure to the US government warning and mental health, pregnancy, minors, liver cirrhosis, cancer, and addiction mock health warnings, across six ads promoting spirits, wine, and beer brands. Compared to the control group, exposure to health warnings in the form of user-generated comments was associated with a decrease in drinking intentions. Whereas pro-drinking, antidrinking, and responsibility messages did not influence intentions to drink [45]. Similarly, Noel [47] tested mental health, drink drive, pregnancy, cancer, and addiction health warnings across spirits, wine, and beer alcohol ads on social media. Again, the warnings were posted as a comment underneath the social media ads and were compared against control groups, pro-drinking, antidrinking comments, and industry-sponsored (responsibility) messages. It was found that warning labels posted as comments did not differ significantly from industry-sponsored messages and that anti-drinking comments decreased purchase intentions significantly more than warning labels [47]. It is noted, however, that the responsibility messages were embedded in the ads, whereas the warning labels were not. That said, both studies measured warning labels which were posted as comments and were not directly embedded in the social media alcohol ads. Furthermore, both studies highlight that the format and placement of health warnings—particularly whether they appear as embedded content or user comments can influence their effectiveness.

In contrast, Lou and Alhabash [32] investigated the effects of warning design formats embedded in alcohol advertisements on social media, involving 65 participants aged 18–20, with findings suggesting that integrated warning formats attracted greater visual attention and led to less favourable attitudes toward both the post and the advertised alcohol brand.

Discussion

The primary aim of the current review was to evaluate the role of digital alcohol marketing in shaping alcohol-related behaviours and consumption patterns. The findings of this systematic review extend the literature baseline established by Noel et al. [13], illustrating that digital alcohol marketing is now a pervasive feature of everyday online environments. Contrasting the two sets of results reveals a clear evolution in how exposure and engagement operate. While Noel et al. [13] concluded that the behavioural impacts of passive online exposure were somewhat inconclusive, our updated review demonstrates a direct link between the self-reported recall of digital alcohol ads and subsequent episodic drinking among young cohorts (e.g. [37]). Furthermore, Noel et al. [13] identified engagement (active participation) as an emerging trend linked to risky consumption. Our findings expand this relationship by highlighting that engagement, specifically through UGC, has become a key driver of influence in the context of alcohol marketing. While the previous review treated interactive features as sub-components of digital platforms, the studies evaluated here establish that peer-shared, pro-consumption UGC is often more strongly associated with drinking desires and shifted social norms than standard corporate-sponsored branding alone [43]. Finally, our systematic review expands on existing research, by evaluating empirical countermeasures, as the previous review by Noel et al. [13] did not consider the role of health warnings in alcohol advertising. Although the evidence on health warnings, particularly for alcohol advertisements and digital media, is still emerging, the few studies included here demonstrate that the formatting and placement of these interventions, such as ad-embedded designs [32] versus warnings posted in comment Sects. [45, 46] can influence ad/brand outcomes and alcohol related behaviours. This brings both theoretical and practical implications for debates on alcohol policy.

Theoretical Implications

In our review, we found positive associations between exposure to and engagement with digital alcohol marketing and increased levels of consumption, consistent with previous reviews showing the significant impact of alcohol marketing on alcohol-related behaviours, particularly among youth. In our review, only four studies focused on older populations with a sample of adults above the age of 30 [32, 41–43] highlighting the need of a greater focus on other demographics, in line with recent calls for research agenda involving adults [49]. The findings reported here are mostly US-based, suggesting the need for further research in other geographical regions, including Europe where approximately one million people die each year due to alcohol consumption [50], and harmful drinking patterns remain prevalent.

Furthermore, a small number of studies, used theories in explaining their study findings [38] or referenced explanatory frameworks in their literature reviews [31]. Several theories have been used as explanatory pathways for the influence of persuasive messages on consumer behaviour, including the Heuristic Advertising Receptivity Model [51], the Message Interpretation Process Model [52, 53] and the Elaboration Likelihood Model [54, 55]. Collectively, these theories suggest that psychological processes to marketing and advertising include both cognitive and affective elements such as alcohol expectancies, normative beliefs, personal characteristics, all of which are thought to be predictors of alcohol consumption. Future studies should apply psychological and behavioural theories, and test which theoretical variables moderate the impact of digital alcohol marketing on drinking-related behaviours.

The studies included in our review predominantly examined exposure to, and engagement with, digital alcohol marketing [31, 34–39, 41, 42], whereas comparatively few investigated how specific advertising content, user-generated content, or health warning characteristics influenced alcohol-related outcomes [32, 43, 45–47]. Based on the findings, however, the content of alcohol ads, UGC, and the use of health warnings appear to be important contextual measures when examining the role of digital alcohol marketing online. The context in which alcohol advertising is viewed, including the presence of UGC or visible health warnings, should be considered when examining the influence of alcohol marketing more broadly and the impact of digital alcohol marketing specifically. Given that alcohol consumption is a social activity, it would be interesting to determine whether contexts matter with respect to content types (e.g. sponsored vs. user-generated) and health warnings (e.g. warnings in alcohol ads vs. warnings embedded as messages or comments).

While empirical research on UGC in digital alcohol ads remains limited, study results in our review varied depending on the type of peer messaging: positive pro-drinking UGC (e.g., enthusiastic comments, likes) significantly increased drinking motivation and ad appeal [44], although others that did not find significant results on alcohol use [48]. Whereas anti-drinking UGC (e.g., critical peer comments) decreased purchase intentions and reduced perceived control over drinking [45, 47]. As UGC is created and disseminated by users rather than companies, it exerts a distinct type of influence, expanding the reach and impact of alcohol marketing beyond that of company-generated content alone. In the context of the elaboration likelihood model, these findings suggest that UGC can be viewed as another contextual factor explaining behaviour, and what stands out in the literature is the need for more data to explain the circumstances under which UGC moderates the impact of alcohol advertising on behavioural outcomes and consumption patterns.

Little research to date has also taken into account the role of health warnings in digital alcohol ads. The findings on the efficacy of health warnings reported here differed depending on warning placement and formatting. For instance, health warnings embedded within the ad graphic [32] reduced brand favourability and drinking intentions, whereas warning labels posted merely as separate user comments below the ad led to somewhat weaker effects compared to industry-sponsored labels [45, 47]. Health warnings, particularly in alcohol ads, have not been used previously, and they are likely to be a new experience for the participants. Despite this, the current findings are in line with previous research demonstrating that health warnings in alcohol ads can be effective on behavioural and ad-related outcomes [24, 56]. Although this prior research does not test warnings specifically in digital settings, when combined with the emerging evidence from the present review, these findings indicate that ad-embedded warnings and UGC can act as countermeasures to the persuasive influence of commercial marketing. The findings reported here may be due to the many permutations that can exist across different warning designs and formats. For instance, health warnings in advertisements are subject to other contextual factors impacting behaviour, such as the content of ads themselves, and research should assess a wider range of theoretical outcomes to include both the warning designs and the content of alcohol ads [23, 24, 56]. Future research could also shed light on the role of multiple warnings embedded in alcohol advertising online.

Practical Implications

Alcohol policy population-based interventions include minimum unit pricing, reducing alcohol availability, content and volume restrictions on alcohol marketing and advertising, many of which are subject to industry opposition. Many countries continue to rely on voluntary agreements with the alcohol industry, although evidence has shown that the use of self-regulatory codes has been unsuccessful in protecting consumers from exposure to alcohol advertising, particularly in the digital space, where violations of these codes are more frequent [57, 58]. Our systematic review illustrates how young people, including adolescents, are exposed to and engaged with digital alcohol marketing, highlighting how existing self-regulation practices are failing. Given that public health aims to protect individuals and promote healthier choices, health promotion strategies recommend either a total ban on alcohol marketing or restrictions on advertising, sponsorship, and promotional activities [59]. For instance, Finland and the Russian Federation have prohibited advertising and promotion of alcohol products on social media [3]. Legislative control measures including advertising content restrictions and mandatory health warnings are relevant to Ireland, where a significant step in alcohol policy has been made through the ratification of the Public Health (Alcohol) Act. For example, Sect. 13 of the Irish Act restricts alcohol advertising to product‑related information and excludes social imagery, aiming to reduce brand salience and perceived ad attractiveness [60].

Similarly, researchers are concerned there is a lack of knowledge and information available to consumers specifically on the health risks associated with alcohol consumption [23, 24, 56]. The implementation of prominent health warnings will inform consumers of the health risks associated with drinking alcohol, and mandating multiple content‑specific health warnings addressing different alcohol‑related health conditions such as cancer, liver disease, cardiovascular harms, and risks during pregnancy, should be a priority for policymakers [24].

Furthermore, the findings reported in our review support calls for mandatory health warnings on alcohol advertising beyond traditional media. While content creators and influencers are not yet required to use health warnings as part of their sponsored alcohol promotion practices, doing so, or even prohibiting influencers promoting alcohol products, could be another useful direction for alcohol policy. Social media influencers should also disclose when they are being paid by a brand to promote alcohol products. In addition, the quantity and exposure to alcohol marketing (both online and offline) should also be regulated with restrictions on media channels, retail hours, and marketing alcohol on specific venues [61].

A recent study with adults in Ireland examined changes in awareness of alcohol marketing one year after the restrictions with respect to children’s branded clothing, public transport, outdoor, and cinema advertising were implemented. Based on two waves of data collection, the study found that the new restrictions on alcohol marketing decreased alcohol marketing awareness [62], demonstrating how regulation can advance public health objectives. Finally, although there are previous studies which have examined consumer reactions to various forms of digital alcohol marketing, we were able to identify only one study that has explored exposure to alcohol marketing in metaverse environments [63]. The study findings suggest that immersive engagement is a novel marketing practice, with new and unknown implications on alcohol-related behaviours. Therefore, the question of how artificial intelligence, including advergames or virtual influencers will be regulated—both broadly and within the alcohol industry—remains open, with significant implications for researchers and policy makers who seek to explore the impact of digital alcohol marketing on drinking behaviour.

Strengths and Limitations

Our review contributes to the growing research evidence on alcohol marketing in digital environments, and it is the first, to our knowledge, highlighting the role of UGC and health warnings in digital alcohol marketing as important contextual factors that warrant attention. The consistency of evidence regarding the power of UGC and digital marketing is methodologically credible, based on high-scoring studies such as Critchlow et al. [33], Noel et al. [48], and Alhabash et al. [38].

Several limitations warrant consideration. First, restricting searches to English-language publications may have omitted relevant studies from non-English speaking regions. Second, reliance on cross-sectional designs and non-probabilistic convenience samples of university students limits causal inference and population generalisability. Third, retrospective self-reported measures of digital exposure are susceptible to recall bias, highlighting the need for objective digital tracking methodologies in future research.

Finally, the summary of findings in our review is based on quantitative evidence with a sample of 18 studies, thus the full impact of digital alcohol marketing, UGC, and health warnings in shaping consumption patterns has not been fully explored here. Qualitative designs can provide more insights on the efficacy of user-generated and health warning messages on consumer behaviour.

Conclusion

The findings of this systematic review warrant close attention from public health practitioners. The evidence emphasises the need for population-based interventions to reduce alcohol-related harm and to restrict exposure to alcohol marketing across digital platforms. Furthermore, regulatory measures, independent of industry influence, are required in assessing the broader impact of alcohol marketing particularly through the lens of upstream social marketing applications. The findings also support efforts to limit online alcohol advertising and point to the potential of counterstrategies, such as implementing health warnings and strengthening public health messaging. Further research is needed to explore the impact of UGC, and alcohol advertising health warnings in digital marketing including metaverse environments.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Original conceptualisation by STS, SRK, GF and NG. Initial draft writing of the main manuscript text led by GF and WvD. Methodology, analysis, writing and revisions of the main manuscript text led by STS and VF. Further revisions and edits - STS, VF, SRK, and GL. All authors reviewed the maniscript.STS - Sandra TricasSRK - Sandra Sandra Radoš KrnelGF- Giulia FerrareseNG- Nils Garnes WvD- Wim van Dalen VF - Vania FilipovaGL - Gorazd Levičnik.

Funding

This article was prepared as part of the AlHaMBRA Project, funded by the European Commission’s European Health and Digital Executive Agency (HaDEA), Service Contract No 20197105 – Title: Support to member states in studies and capacity building activities to reduce alcohol-related harm. Views expressed in this document do not necessarily reflect the views of the European Commission.

Data Availability

No datasets were generated or analysed during the current study.

Declarations

Ethics Approval and Consent to Participate

Not applicable.

Consent for Publication

All authors agreed to the publication of this manuscript.

Competing interests

The authors declare no competing interests.