Understanding the associations between eHealth activities and support for alcohol restrictions among drinkers—the roles of alcohol-related health risk awareness and self-efficacy

Abstract

Objective

Excessive alcohol consumption is a significant health concern in the United States. Alcohol restriction policies can help control the harms caused by alcohol use. eHealth activities have been proposed to bolster public support for such policies. However, the psychological mechanisms underlying this relationship remain unexplored. This study aims to investigate the association between eHealth activities and support for alcohol restriction policies by influencing the specific awareness of certain diseases (cancer and heart disease), and how this relationship can be moderated by self-efficacy, among three groups of drinkers (i.e., light, moderate, and heavy drinkers).

Method

This study employed a secondary analysis of the Health Information National Trends Survey (HINTS) in 2020, with a sample size of 1388. A moderated mediation analysis was conducted to examine the relationships among focal variables.

Result

This study reported a positive indirect relationship between eHealth activities and support for alcohol restriction policies mediated by drinkers’ perception that alcohol consumption causes cancer. Moreover, a positive moderating effect of self-efficacy was found in the relationship between eHealth activities and belief that alcohol causes cancer. However, for beliefs about alcohol causes heart disease, neither the indirect effect nor the moderating effect of self-efficacy was significant. In subgroup analysis, the indirect pathway from eHealth activities to support for alcohol restriction policies, via increased belief that alcohol causes cancer, was significant among light and moderate drinkers. Self-efficacy moderated the link between eHealth activities and beliefs about alcohol-related cancer and heart disease in moderate drinkers.

Conclusion

This study indicates the role of eHealth activities to promote alcohol restriction policies by raising drinkers’ belief that alcohol causes cancer. Moreover, self-efficacy is crucial since it can strengthen the impact of eHealth activities on the belief that alcohol causes cancer. Finally, it is essential to correct the misconception about alcohol's benefits for heart disease.

Keywords

eHealth alcohol restriction alcohol restriction policies self-efficacy moderated mediation

Introduction

Excessive alcohol consumption is a significant health issue associated with over 60 health conditions, such as high blood pressure, heart disease, and cancer, resulting in an estimated 4% of the global disease burden.¹ Despite the numerous health risks associated with alcohol consumption, the drinking problem in the United States remains critical. According to the 2019 report of the National Institute of Alcohol Abuse and Alcoholism, nearly 15 million individuals aged 12 and above had alcohol use disorder.²

The severity of alcohol-related problems has led to widespread concern among scholars and governments, prompting a focus on developing alcohol restriction policies and investigating their associations with reducing alcohol use and related harms.^3–5 However, the lack of public support may undermine the effectiveness of such policies,⁶ attracting attention from both academia and industry.^7–9

This study targeted drinkers as the audience, since they are directly affected by alcohol restriction policies,¹⁰ but often resist such policies that restrict their alcohol consumption.^11,12 Developing interventions to garner their support can lead to actual health behavioral changes and successful implementation of such policies.¹³

This study aims to address the following key research gaps. First, while previous studies investigating factors influencing alcohol-related behaviors have primarily focused on drinking behaviors themselves, support for alcohol restriction policies remains relatively underexplored.¹⁴ Given the importance of such support, it is significant to investigate this area further. Second, scholars have investigated how individual variables, such as gender and age,^9,15 income,¹⁶ and education,¹⁷ are associated with public support for alcohol restriction policies. However, as most of these factors are non-modifiable, it is of interest to investigate modifiable factors to provide implications for controlling the harms of alcohol use. Previous studies have found that awareness of health risks associated with alcohol use may be a relevant factor in understanding public support for alcohol restriction policies.^7,8,18 However, few studies have specifically examined what types of risks are associated with public support. To be specific, previous research has predominantly examined cancer-related risk perceptions and their influence on alcohol attitudes and behaviors,^7,8 heart disease represents an equally significant alcohol-related health risk that has received limited attention in policy support research. Differentiating between disease-specific risk beliefs responds to emerging evidence that public perceptions and misinformation about alcohol's health risks vary substantially by disease type.¹⁹

Third, although eHealth interventions have been shown to influence alcohol risk perceptions,²⁰ their potential as tools for enhancing support for alcohol restriction policies remains unexplored. This study examines whether eHealth activities relate to policy support through their influence on perceived drinking risks, thereby specifying this indirect relationship and identifying which risk types are most influential for future intervention development. Fourth, we aim to investigate how to optimize eHealth effectiveness by incorporating self-efficacy as a moderator to identify when and for whom these activities are most effective in shaping alcohol-related risk beliefs. While self-efficacy's moderating role in eHealth-health belief relationships has been established,²¹ its specific influence on alcohol-related disease risk beliefs warrants further investigation. This approach provides a theoretically significant bridge between individual-level digital health engagement and collective policy preferences through domain-specific cognitive risk beliefs and motivational factors, extending the Social Cognitive Theory to the alcohol control context. Finally, this study examines model relationships across light, moderate, and heavy drinking subgroups, an area previously unexplored in the literature. These findings will identify optimal intervention targets for different drinking levels, support precision policy advocacy, and reveal how eHealth intervention effectiveness varies according to alcohol consumption intensity.

With the rapid development of information and communication technologies (ICTs), eHealth activities are considered an effective intervention in influencing people's health-related behavioral intentions, especially in alcohol-related issues.²² eHealth activities (abbrv. EHA) are defined as the use of modern electronic information and technologies, including email, text messaging, push notifications, websites, and mobile-based applications, for health-related purposes.²³ Based on the results of the Health Information National Trends Survey (HINTS) in 2020, this study investigates the association between eHealth activities and support for alcohol restriction policies among drinkers, by influencing two specific types of health risks of alcohol use (i.e., the belief that alcohol causes cancer, abbrv. BAC; and the belief that alcohol causes heart diseases, abbrv. BAH), and how such mediation relationships depend on a modifiable individual factor (i.e., self-efficacy).

It is widely acknowledged that eHealth activities can influence health behavioral change through cognitive processes.^20,24 In the context of alcohol-related issues, people can gain awareness of the negative consequences of drinking from eHealth activities, and further perform relevant behavioral changes.^20,25 Among the various risks of drinking, those related to cancer and heart diseases are particularly important for encouraging people to engage in alcohol restriction behaviors,^8,10,26 especially among drinkers.⁷ However, previous studies have found that many people have relatively low levels of understanding of the risks that alcohol causes cancer^7,10,27 and heart diseases.^19,28–30

eHealth activities, among various interventions, have been found to effectively inform people about beliefs that alcohol causes cancer and heart diseases.^31–34 Hence, the first hypothesis:

H1: eHealth activities are positively associated with drinkers’ beliefs that alcohol causes cancer (H1a) and heart diseases (H1b).

People who are aware of the risks of drinking are more likely to engage in alcohol restriction activities (e.g., abstaining from alcohol or not driving after drinking). Especially for drinkers, who are directly affected by such policies, and may see those policies as intrusive to their freedoms and enjoyment.^11,12,35

Alcohol consumption contributes to various cancers (e.g., mouth, breast, colon, esophagus, etc.) even at low levels and harms cardiovascular health, causing hypertension, arrhythmias, atrial fibrillation, and heart diseases.^7,36–41

Awareness that alcohol consumption can cause cancer and heart diseases significantly increases support for alcohol restriction policies (e.g., taxation, marketing, labeling) across regions like Australia, England, Denmark, Canada, and the U.S.^{7,8,10,42–44} Thus, the second and third hypotheses are:

H2: Drinkers’ beliefs that alcohol causes (H2a) cancer and (H2b) heart diseases are positively associated with their support for alcohol restriction policies.

H3: The positive relationship between eHealth activities and support for alcohol restriction policies is mediated by the beliefs that alcohol causes (H3a) cancer and (H3b) heart diseases.

Self-efficacy refers to the confidence a person has about their capacity to perform certain behaviors that may be relevant to certain outcomes.⁴⁵ In the context of public health, it can be defined as the confidence to manage one's health behaviors.

Previous studies indicated that self-efficacy can enhance the effects of eHealth activities.^46–48 The underlying mechanism of the moderating role of self-efficacy may result in its impact on learning accomplishment. According to the social cognitive theory, people with high self-efficacy are likely to benefit more from learning and information processing activities, as they are more engaged and less prone to distractions, leading to better performance outcomes.^49–51

In the context of this study, eHealth activities can be regarded as a process of learning by using digital tools to gain new knowledge of health.^52–54 Beliefs that alcohol causes cancer and heart diseases can also be considered as new knowledge that can be absorbed by participating in eHealth activities. Drinkers with high self-efficacy in health management are more motivated to engage in eHealth activities, to persevere in the face of relevant challenges, and to use effective learning strategies,^55–58 resulting in the acquisition of more health information and knowledge. Therefore, we assume that drinkers with high self-efficacy may gain more belief that alcohol causes cancer and heart diseases from eHealth activities, which may further generate stronger support for alcohol restriction policies, and propose the following two hypotheses:

H4:. Self-efficacy moderates the relationship between eHealth activities and the beliefs that alcohol causes (H4a) cancer and (H4b) heart diseases, such that the relationships are stronger for those with high self-efficacy.

H5: Self-efficacy moderates the indirect relation between eHealth activities and support for alcohol restriction policies through the beliefs that alcohol causes (H5a) cancer and (H5b) heart diseases, such that the mediating relationships are stronger for those with high self-efficacy.

In alcohol-related research, consumption levels consistently serve as a critical factor influencing drinking-related behaviors, risk perceptions, and intervention effectiveness.^59,60 Heavy drinkers demonstrate a pronounced tendency to portray alcohol consumption positively, emphasizing perceived benefits such as stress relief, anxiety reduction, and alleviation of loneliness, and even endorsing misconceptions about cardiovascular benefits.^61,62 This positive framing often stems from their engagement in defensive information processing: a psychological mechanism where individuals with higher alcohol consumption show significantly greater levels of message derogation, defensive avoidance, and reduced threat perception when confronted with alcohol-related health risks.⁶³ This defensive processing pattern manifests as a self-serving bias that protects their sense of identity and reduces cognitive dissonance between knowledge of risks and continued drinking behavior.^64,65 Heavy drinkers frequently employ psychological denial as a primary defense mechanism, downplaying both the quantity consumed and its impact while justifying their behavior through external attributions such as stress or social pressures rather than acknowledging personal responsibility.^66,67

Regarding alcohol restriction policies, heavy drinkers typically show lower support levels, particularly when they perceive such policies as ineffective or unlikely to influence other heavy drinkers’ behavior.⁶⁸ This resistance partly stems from their defensive processing patterns and impaired executive function, which limits their ability to process risk information effectively even when self-efficacy is high.⁶⁹ However, recent research also reveals a nuanced picture: some heavy drinkers, particularly those who have experienced negative consequences or attempted to reduce consumption, may actually rate their personal risk higher than lighter drinkers.⁷⁰ In such cases, these individuals might become more supportive of alcohol restriction policies, viewing policy restrictions as validating their personal recognition of alcohol's risks rather than as threats to their freedom. This suggests that among heavy drinkers, experiential learning from negative consequences can override defensive processing patterns, potentially creating windows of opportunity for policy support and intervention effectiveness.

The effectiveness of health interventions is also influenced by participants’ drinking levels. Research demonstrates that interventions in primary care and group settings produce moderate, significant reductions in alcohol consumption among harmful drinkers.⁷¹ Heavy drinkers typically achieve greater absolute reductions in consumption, while lighter drinkers may demonstrate better relative outcomes in meeting low-risk drinking guidelines. This differential response pattern may stem from variations in self-efficacy, as individuals with less severe consumption patterns maintain greater confidence in their ability to control drinking, thereby facilitating better intervention engagement and outcomes.⁷² Additionally, altered reward sensitivity regarding alcohol consumption may contribute to these different response patterns.⁷³

Digital health interventions present a more complex picture regarding drinking level effects. Some research indicates that heavy drinkers demonstrate lower success rates following web-based alcohol interventions.^74,75 However, other studies reveal that heavy drinkers are significantly more likely to express concern about their drinking after viewing alcohol media interventions and may be more inclined to engage in protective behaviors.^76,77 This apparent contradiction may be explained by heavy drinkers’ direct experiential evidence of alcohol's harmful effects, which creates cognitive resonance when receiving alcohol-related information, potentially strengthening intervention impacts.

To conclude, alcohol consumption levels might influence multiple dimensions of alcohol-related intervention effectiveness and behavioral responses. However, the research literature presents mixed findings, with some studies reporting positive effects while others document negative outcomes across these domains. Given these conflicting findings, this study addresses the following research question:

RQ1: To what extent do the conditional indirect relationships between eHealth activities and support for alcohol restriction policies through alcohol-related health risk beliefs (cancer and heart disease), moderated by self-efficacy, vary across light, moderate, and heavy drinkers?

Methods

Data source and participants

To test the hypotheses, this study conducted a secondary analysis of the HINTS administered in 2020 (HINTS 5 Cycle 4). HINTS is a nationally representative survey, conducted by National Cancer Institute (NCI) to examine American public's health information use, health beliefs and health behaviors. The HINTS 5 survey has received ethical approval from Westat IRB (Project number: 6048.14). A total of 3865 completed responses were collected with a response rate of 37%. For the purpose of this study, we specifically focused on drinkers, identified as those who reported consuming alcohol at least once per week. From the initial dataset, 1600 participants were identified. Listwise deletion for missing data was used to generate a complete dataset. The final sample size was 1388.

Based on categorizations provided by previous studies,^78,79 we categorized participants into three subgroups in terms of their drinking levels: light, moderate and heavy drinkers. Light drinkers were defined as individuals consuming ≤3 drinks per week; moderate drinkers included men who have＞3 drinks ≤14 drinks per week and women who consumed ＞3 drinks to ≤7 drinks per week; heavy drinkers were men consuming＞14 drinks per week and women currently using＞7 drinks per week. One alcoholic drink–equivalent is described as containing 14 g (0.6 fluid ounce) of pure alcohol. The following are reference beverages that are 1 alcoholic drink–equivalent: 12 fluid ounces of regular beer (5% alcohol), 5 fluid ounces of wine (12% alcohol), or 1.5 fluid ounces of 80 proof distilled spirits (40% alcohol). After excluding cases with erroneous responses, the final sample comprised 578 light drinkers, 504 medium drinkers, and 290 heavy drinkers.

Detailed information on demographics is presented in Table 1.

Table 1.

Sample characteristics.

Variable	Statistics
Age, mean (SD)	53 (16.14)
Gender, N (%)
Male	681 (49.1%)
Female	707 (50.9%)
Education, N (%)
Less than 8 years	12 (0.9%)
8 through 11 years	22 (1.6%)
12 years or completed high school	169 (12.2%)
Post high school training other than college (vocational or technical)	90 (6.5%)
Some college	288 (20.7%)
College graduate	446 (32.1)
Postgraduate	361 (26.0%)
Political viewpoint, N (%)
Very liberal	98 (7.1%)
Liberal	239 (17.2%)
Somewhat liberal	144 (10.4%)
Moderate	453 (32.6%)
Somewhat conservative	165 (11.9%)
Conservative	219 (15.8%)
Very conservative	70 (5.0%)
Household annual income, N (%)
$0 to $9999	40 (2.9%)
$10,000 to $14,999	40 (2.9%)
$15,000 to $19,999	40 (2.9%)
$20,000 to $34,999	115 (8.3%)
35,000 to $49,999	162 (11.7%)
$50,000 to $74,999	235 (16.9%)
$75,000 to $99,999	200 (14.4%)
$100,000 to $199,999	394 (28.4%)
$200,000 or more	162 (11.7%)
Drinking levels
Light drinkers (≤3 drinks/week)	578 (41.6%)
Moderate drinkers (>3 drinks to ≤14 drinks/week for men; >3drinks to ≤7 drinks/week for women)	504 (36.3%)
Heavy drinkers (＞14 drinks/week for men; ＞7 drinks for women)	290 (20.9%)
Missing or erroneous	16 (1.2%)
N	1388 (100.0%)

Measurement

Ten variables were used to test the hypotheses. eHealth activities (EHA) was measured with four items, drawn from some previous research.^58,80 Respondents were asked whether in the past 12 months they had used a computer, smart phone or other electronic means to (i) look for health or medical information, (ii) communicate with a doctor or a doctor's office, (iii) look up medical test results, and (iv) make appointments with a health care provider. Responses to these four questions (0 = “no,” 1 = “yes”) were summed together for analysis (M = 2.48, SD = 1.41).

Self-efficacy utilized a single item, similar to prior studies.^81,82 Respondents rated their self-efficacy with the question “Overall, how confident are you about your ability to take good care of your health?.” A five-point Likert scale was used (1 = “never confident” to 5 = “completely confident”). The higher score represented higher level of self-efficacy (M = 3.94, SD = 0.76).

Belief that alcohol causes cancer (BAC) comprised three items. Respondents were asked to report how much drinking the following types of alcohol (i) beer, (ii) wine and (iii) liquor affect the risk of getting cancer, on a 5-point scale (1 = “decreases risk a lot” to 5 = “increases a lot”). Moreover, we recoded the answer of score 6 (“don’t know”) into score 3 (“No effect”) (M = 3.28, SD = 0.60, Cronbach's α = 0.86).

Belief that alcohol causes heart disease (BAH) was also measured by three items. Participants were asked to answer how much they believe consuming the following types of alcohol (i) beer, (ii) wine and (iii) liquor affect the risk of getting heart disease. The responses were handled similarly to BAC (M = 3.34, SD = 0.66, Cronbach's α = 0.82).

Support for alcohol restriction policies (SAR) was measured using three items, adapted from.⁷ Participants were asked to report to what extent would they support or oppose (i) banning outdoor advertising of alcohol such as on billboards and bus stop, (ii) requiring specific health warnings on alcohol containers and (iii) requiring alcohol containers to show the recommended drinking guidelines for keeping health risks low. Each item was answered on a five-point Likert scale (1 = “strongly oppose” to 5 = “strongly support”) (M = 3.54, SD = 0.86, Cronbach's α = 0.83).

Socio-demographics were used as controls to reduce confounding effects, including age, gender, education, annual household income range, and political viewpoint.

Data analysis

Moderated mediation was tested using SPSS PROCESS Model 7.⁸³ Min-Max normalization was introduced to convert focal variables into a common measurement scale, 0–1, with regression coefficients generated denoted as percentage coefficients (b_p). Relationships were analyzed with 95% confidence intervals (CIs) using 5000 bootstrapped samples.

Results

Table 2 presents the descriptive statistics for the key variables (in original and 0–1 percentage scales). Zero-order Pearson correlations are reported in Table 3.

Table 2.

Descriptive statistics of the independent, dependent, mediating and moderating variables.^a,b

Variables	(Original scale) Mean ± SD	(New scale) Mean ± SD
Dependent variable
SAR (support for alcohol restriction policies)	3.54 ± 0.86	0.63 ± 0.22
Independent variable
EHA (eHealth activities)	2.48 ± 1.41	0.62 ± 0.35
Mediating variables
BAC (belief that alcohol causes cancer)	3.28 ± 0.60	0.57 ± 0.15
BAH (belief that alcohol causes heart diseases)	3.34 ± 0.66	0.59 ± 0.17
Moderating variable
SEF (self-efficacy, five levels)	3.94 ± 0.76	0.73 ± 0.19

The original scale of the variables has five levels (1–5); EHA (0–4).

The new scale has two levels (0–1).

Table 3.

Zero-order Pearson correlations among focal variables.

Variables	1	2	3	4	5
1. EHA	—	.142***	.113***	.084**	.088**
2. BAC		—	.584***	.191***	.001
3. BAH			—	.171***	−.053*
4. SAR				—	.018
5. SEF					—

* P < .05, ** P < .01, *** P < .001.

Hypotheses testing results are presented in Table 4 and Figure 1. H1a predicting a positive relationship (a₁ path) between EHA and BAC was supported by significant results (B = .026, b_p = .026, 95% CI [.006, .046], P < .01). However, H1b predicting a positive relationship between EHA and BAH (a₂ path), was not (B = .012, b_p = .012, 95% CI [−.011, .034], P = .307). H2a suggested BAC positively associates with SAR, controlling for EHA (b₁ path), which was supported (B = .184, b_p = .186, 95% CI [.096, .275], P < .001). Similarly, H2b suggesting BAH positively correlates with SAR, controlling for EHA (b₂ path), was supported (B = .103, b_p = .101, 95% CI [.021, .181], P < .05).

Figure 1.

Associations between EHA and SAR, mediated by BAC and BAH, moderated by SEF. Note: Path indicators are percentage coefficients (bp); * P < .05; ** P < .01; *** P < .001.

Table 4.

Regression analyses of mediation and moderation relationships. ^a−c

Relationships	Coefficient	SE	95% CI	P
Mediation pathways
a₁ path: EHA → BAC	B = .026, b_p = .026	.010	[.006, .046]	.01
a₂ path: EHA → BAH	B = .012, b_p = .012	.011	[−.011, .034]	.307
b₁ path: BAC → SAR	B = .184, b_p = .186	.046	[.096, .275]	.001
b₂ path: BAH → SAR	B = .103, b_p = .101	.041	[.021, .181]	.05
Indirect relationship
a₁ $\times$ b₁ path: EHA → BAC → SAR	B = .005, b_p = .005	.002	[.001, 010]
a₂ $\times$ b₂ path: EHA→ BAH → SAR	B = .001, b_p = .001	.001	[−.001, 004]
Conditional indirect relationships
Low SEF → a₁ $\times$ b₁ path	B = .0003, b_p = .0003	.003	[−.005, .006]
Medium SEF → a₁ $\times$ b₁ path	B = .005, b_p = .005	.002	[.001, .010]
High SEF → a₁ $\times$ b₁ path	B = .009, b_p = .010	.004	[.003, .017]
Low SEF → a₂ $\times$ b₂ path	B = .001, b_p = .001	.002	[−.002, .005]
Medium SEF → a₂ $\times$ b₂ path	B = .001, b_p = .001	.001	[−.001, .005]
High SEF → a₂ $\times$ b₂ path	B = .001, b_p = .001	.002	[−.002, .006]
Direct relationship
d path: EHA → SAR	B = .007, b_p = .008	.017	[−.026, .041]	.655
Interaction effect
SEF $\times$ EHA→ a₁ path	B = .032, b_p = .126	.050	[.028, .224]	.05
SEF $\times$ EHA→ a₂ path	B = .002, b_p = .007	.056	[−.104, .117]	.906
Conditional relationships of the a₁ and a₂ paths
Low SEF → a₁ path	B = .002, b_p = .002	.014	[−.026, .029]	.903
Medium SEF → a₁ path	B = .026, b_p = .026	.010	[.006, .045]	.05
High SEF → a₁ path	B = .050, b_p = .050	.014	[.023, .077]	.001
Low SEF → a₂ path	B = .010, b_p = .010	.016	[−.020, .041]	.512
Medium SEF → a₂ path	B = .012, b_p = .012	.011	[−.011, .033]	.308
High SEF → a₂ path	B = .013, b_p = .013	.015	[−.018, .043]	.409
Index of moderated mediation
SEF → a₁ $\times$ b₁ path	B = .006, b_p = .023	.013	[.002, .053]
SEF → a₂ $\times$ b₂ path	B = .0002, b_p = .0007	.006	[−.011, .015]
Total relationship
c path: EHA → SAR	B = .007, b_p = .021	.017	[−.022, .046]	.233

All models controlling for gender, age, education, annual household income, political viewpoint.

B represents unstandardized coefficient.

Percentage coefficient (b_p) used in this table.

H3a proposed that EHA was positively and indirectly associated with SAR through BAC (a₁, b₁, and a₁*b₁). This indirect relationship was significant (B = .005, b_p = .005, 95% CI [.001, .010]), thus supporting H3a. H3b proposed BAH as an additional mediator, but the indirect relationship of EHA on SAR via BAH was not statistically significant (B = .001, b_p = .001, 95% CI [−.001, .004]), leading to the rejection of H3b.

H4a proposed that self-efficacy strengthens the relationship between EHA and BAC (a ₁ path). The interaction effect of EHA $\times$ self-efficacy was significant (B = .032, b_p = .126, 95% CI [.028, .224], P < .05). A simple slope test (Figure 2) showed that for people with high self-efficacy, the relationship between EHA and BAC was significant and positive (B = .050, b_p = .050, 95% CI [.023, .077], P < .001); for medium self-efficacy, the relationship was significant but weaker (B = .026, b_p = .026, 95% CI [.006, .045], P < .05); for low self-efficacy, the relationship was weaker and not significant (B = .002, b_p = .002, 95% CI [−.026, .029], P = .903). Thus, H4a was supported, showing stronger relationship between EHA and BAC at higher self-efficacy. H4b assumed self-efficacy moderates the positive relationship between EHA and BAH (a₂ path), but the relationship was not significant (B = .002, b_p = .007, 95% CI [−.104, .117], P = .906), leading to the rejection of H4b.

Figure 2.

The EHA–BAC association moderated by SEF. Note: Vertical and horizontal axes are both on 0–1 percentage scales (ps); SEF, self-efficacy.

H5a stated that self-efficacy strengthens the indirect relationship between EHA and SAR through BAC. The moderated mediation index was significant (B = .006, b_p = .023, 95% CI [.002, .053]). Conditional mediation analysis revealed that the relationship of EHA-BAC-SAR (a₁ $\times$ b₁ path) was significant and positive (B = .009, b_p = .010, 95% CI [.003, .017]) for high self-efficacy, significant but weaker for medium self-efficacy (B = .005, b_p = .005, 95% CI [.001, .010]), and non-significant for low self-efficacy (B = .0003, b_p = .0003, 95% CI [−.005, .006]). Thus, H5a was supported. For H5b, self-efficacy did not significantly moderate the EHA-BAH-SAR (a₂ $\times$ b₂ path) relationship (B = .0002, b_p = .0007, 95% CI [−.011, .015]), thus rejecting H5b.

For answering the RQ1, we conducted moderated mediation analysis in the three subgroups. In the light drinker subgroup, EHA was positively associated with both BAC (B = .043, b_p = .043, 95% CI [.006, .079], P < .05) and BAH (B = .049, b_p = .049, 95% CI [.007, .092], P < .05), indicating a positive link between EHA and alcohol-related health risk beliefs. BAC was positively associated with SAR (B = .200, b_p = .200, 95% CI [.049, .351], P < .001) while the association between BAH and SAR was found to be non-significant (B = .072, b_p = .072, 95% CI [−.060, .204], P = .283). The direct relationship between EHA and SAR proved statistically insignificant (B = .019, b_p = .019, 95% CI [−.035, .073], P = .485). Mediation analysis revealed that EHA positively influenced SAR indirectly through BAC (B = .009, b_p = .009, 95% CI [.000, .020]), while the path through BAH was insignificant (B = .004, b_p = .004, 95% CI [−.003, .012]). Regarding moderation effects, self-efficacy did not significantly moderate the relationship between EHA and either BAC (B = .034, b_p = .137, 95% CI [−.014, .082], P = .165) or BAH (B = .011, b_p = .045, 95% CI [−.044, .067], P = .688).

In the moderate drinker subgroup, the relationship between EHA and BAC was significant and positive (B = .052, b_p = .052 95% CI [.012, .091], P < .05), whereas the association between EHA and BAH was not significant (B = .032, b_p = .032, 95% CI [−.011, .076], P = .142). Meanwhile, BAC was positively associated with SAR (B = .236, b_p = .236, 95% CI [.096, .375], P < .01), while the association between BAH and SAR was insignificant (B = .101, b_p = .101, 95% CI [−.027, .229], P = .121). Consistent with findings from the lighter drinker group, the direct relationship between EHA and SAR was insignificant (B = −.003, b_p = −.003, 95% CI [−.056, .051], P = .924). EHA exerted a positively indirect effect on SAR through the mediation of BAC (B = .012, b_p = .012, 95% CI [.002, .028], P < .01). Furthermore, self-efficacy significantly moderated the pathway between EHA and BAC (B = .058, b_p = .233, 95% CI [.006, .110], P < .05.). A simple slope test (Figure 3) showed that the relationship was significant and positive among people with high self-efficacy (B = .094, b_p = .094, 95% CI [.040, .149], P < .001), significant but weaker for those with medium self-efficacy (B = .050, b_p = .050, 95% CI [.010, .089], P < .05), and non-significant for individuals with low self-efficacy (B = .005, b_p = .005, 95% CI [−.052, .062], P = .859). The moderated mediation index was also significant (B = .014, b_p = .055, 95% CI [.001, .032]). In addition, self-efficacy significantly and positively moderated the relationship between EHA and BAH (B = .064, b_p = .256, 95% CI [.007, .121], P < .05). A simple slope test (Figure 4) showed that for individuals with high self-efficacy, this relationship was significant and positive (B = .079, b_p = .079, 95% CI [.019, .139], P < .01); however, it was non-significant for both medium (B = .030, b_p = .030, 95% CI [−.013, .079], P = .172) and low self-efficacy groups (B = −.019, b_p = −.019, 95% CI [−.082, .043], P = .550). Despite this moderation, self-efficacy did not significantly moderate the indirect effect of EHA on SAR through BAH, as indicated by a non-significant index of moderated mediation (B = .007, b_p = .026, 95% CI [−.002, .020]).

Figure 3.

Associations between eHealth activities and belief that alcohol causes cancer among moderate drinkers.

Figure 4.

Associations between eHealth activities and belief that alcohol causes heart diseases among moderate drinkers.

In the heavy drinker subgroup, the relationships between EHA and both BAC (B = .039, b_p = .039, 95% CI [−.016, .094], P = .168) and BAH (B = .041, b_p = .041, 95% CI [−.020, .103], P = .185) were found to be non-significant, indicating that EHA is not associated with alcohol-related risk beliefs in this group. Additionally, the association between BAC and SAR was non-significant (B = .100, b_p = .100, 95% CI [−.070, .270], P = .249), whereas BAH was positively associated with SAR (B = 173, b_p = .173, 95% CI [.020, .325], P < .05). The direct relationship between EHA and SAR was also non-significant (B = −.017, b_p = −.017, 95% CI [−.083, .049], P = .605), and the indirect effects of EHA on SAR through either BAC (B = .004, b_p = .004, 95% CI [−.003, .017]) or BAH (B = .007, b_p = .007, 95% CI [−.003, .024]) were likewise non-significant. Moreover, self-efficacy did not moderate the relationships between EHA and BAC (B = .044, b_p = .178, 95% CI [−.017, .105], P = .158) or between EHA and BAH (B = .019, b_p = .074, 95% CI [−.050, .086], P = .593).

Discussion

This study aimed to investigate the relationship between eHealth activities and support for alcohol restriction policies through belief that alcohol causes cancer and heart diseases among drinkers, while also exploring the moderating role of self-efficacy in this mediation process. Results showed that eHealth activities can be an effective means to promote support for alcohol restriction policies by enhancing drinkers’ belief that drinking causes cancer, but not by enhancing the belief that alcohol causes heart diseases. This mediation relationship was stronger for drinkers with high self-efficacy believing alcohol causes cancer but not for those linking it to heart disease.

eHealth as an effective intervention for drinkers’ support for alcohol restriction policies

The study found a positive relationship between eHealth activities and support for alcohol restriction policies among drinkers through the mediating effect of the belief that alcohol causes cancer, which supports previous studies conducted in various contexts.^7,10,27,43 Previous studies have highlighted the need to identify modifiable factors that predict support for alcohol restriction policies and explore how media campaigns can enhance such factors.^8,10,84 Following the study results, eHealth activities can be considered an effective intervention to inform drinkers about important modifiable cancer risk factors and further increase support for alcohol restriction policies.^85,86

To further explore the mechanisms of how eHealth activities affect support for alcohol restriction policies, this study conducted a post-hoc analysis examining the effectiveness of each eHealth activity. Findings revealed that all four types of eHealth activities (i.e., looking for health information, communicating with a doctor, looking up medical test results, and making appointments with healthcare providers) can affect support for alcohol restriction policies through beliefs that alcohol causes cancer. This finding supports the importance of beliefs that alcohol can cause cancer in the relationship between eHealth and people's support for alcohol restriction policies. In contrast, regarding beliefs that alcohol causes heart disease, only Making appointments with healthcare providers demonstrated effectiveness as an intervention (B = .015, bp = .004, 95% CI [.002, .032]). This differential pattern aligns with existing evidence regarding public misinformation and neglect of alcohol-related heart disease risks.¹⁹ There are several explanations for why appointment scheduling uniquely enhances recognition of alcohol-related cardiovascular risks. First, appointments function as gateways to systematic health assessment, representing standardized opportunities for comprehensive healthcare interactions.⁸⁷ Unlike other eHealth activities that focus on specific, isolated health issues, appointment scheduling signals upcoming systematic healthcare services that may encompass comprehensive alcohol screening and cardiovascular risk evaluation, thereby enhancing beliefs about alcohol's impact on heart disease. Second, appointments establish formal therapeutic relationships that facilitate sensitive health discussions, which research indicates are particularly effective for changing health beliefs.⁸⁸ Unlike information seeking or casual online consultations, appointment scheduling demonstrates patient readiness for health behavior discussions and creates an environment where providers can systematically address alcohol-related risks while ensuring patient receptivity.⁸⁹ Finally, information received during formal healthcare appointments may carry greater source credibility and persuasive power compared to self-directed online health information seeking or informal online provider consultations due to its formal nature.⁹⁰

Regarding the moderating role of self-efficacy, among the four eHealth activities, only Looking for health information demonstrated no significant interaction effect with self-efficacy on beliefs that alcohol causes cancer (B = .092, bp = .093, 95% CI [−.006, .191], P = .067). This differential pattern can be explained by two key mechanisms. First, according to the Risk Perception Attitude framework, self-efficacy more strongly moderates behaviors that involve active and motivated engagement.⁹¹ Unlike structured activities such as making appointments or communicating directly with healthcare providers, general health information seeking represents a more passive, habitual, or curiosity-driven behavior that does not necessarily require high confidence in one's ability to manage health to initiate or sustain.⁹² Structured eHealth activities demand greater cognitive and motivational investment; therefore, individuals with high self-efficacy can extract more information, knowledge, and beliefs through such activities.⁹³ Conversely, health information seeking, which is typically less motivated and cognitively demanding, may be more influenced by external factors or general interest rather than internal motivational resources like self-efficacy. The second explanation concerns source credibility differences. The three activities demonstrating significant self-efficacy interactions all involve direct engagement with established medical authorities or formal healthcare systems that possess inherent credibility. In contrast, “looking for health information” typically involves exposure to unregulated, variable-quality online sources that lack consistent credibility verification.^94,95 When accessing information from credible sources, individuals can focus their cognitive resources entirely on information processing, allowing self-efficacy to enhance this focused processing.^96,97 However, when seeking health information online, individuals must simultaneously evaluate source credibility and process information content, creating a cognitive burden that overwhelms the benefits self-efficacy typically provides.⁹⁸ Specifically, during eHealth information seeking, the additional cognitive resources required for credibility evaluation may create a ceiling effect, whereby even highly self-efficacious individuals cannot overcome fundamental credibility deficits to benefit from health information when they cannot determine its validity and reliability, resulting in the insignificant moderating role of self-efficacy.

The role of health beliefs in shaping support for alcohol restriction policies

Within the sample, the levels of belief that alcohol causes cancer and heart diseases are relatively low. Both in the general sample and drinkers, more than 60% of the participants indicated that consuming different types of alcohol may decrease the risk of, have no effect on, or that they have no idea about the influence of alcohol on the risk of getting cancer or heart disease. This result aligns with the current literature, which suggests that although many studies support the risks of alcohol consumption in relation to cancer and heart disease, there is a common misunderstanding among drinkers about the health benefits of alcohol.^7,10,27,99 This is likely due to the alcohol industry's efforts to downplay the potential risks associated with drinking, which reduces public awareness of alcohol-related health risks.^100,101

An important finding of this study is the strong relationship between the belief that alcohol causes cancer and support for alcohol restriction policies, which suggests that drinkers have a significant concern about the relationship between alcohol use and cancer.¹⁰² It is also reflected in the indirect-only mediation model, which is supported by the insignificant direct relationship,¹⁰³ further suggesting that the only mechanism underlying the relationship between eHealth activities and support for alcohol restriction policies is via cancer risk factors. Thus, although the effect size for the association between eHealth activities and belief that alcohol causes cancer is relatively small, the implications of this study cannot be ignored. As scholars¹⁰⁴ pointed out, the main goal of mediation analysis is to understand the underlying mechanisms; thus, the practical significance of the findings does not rely solely on effect size. For the current study, given the insignificant direct relationship, making the mediating factor (i.e., belief that alcohol causes cancer) becomes critical for explaining the relationship between eHealth activities and support for alcohol restriction policies. Especially when the literature indicates that drinkers are unlikely to believe in the harm of alcohol unless they are personally suffering from symptoms of harm from alcohol consumption,¹⁰⁵ policymakers need to disseminate information and knowledge about the link between drinking and cancer to increase support for alcohol restriction policies. In particular, the information should be personally relevant, such as addressing age-related illness and its association with alcohol consumption.¹⁰⁵

However, results also show that perceptions of risk related to heart diseases fail to mediate the relationship between eHealth activities and support for alcohol restriction policies. This suggests that drinkers may be more attuned to health risks related to cancer. This finding is particularly interesting given the World Health Organization reported more deaths from cardiovascular diseases due to alcohol consumption (i.e., 474, 000 deaths, globally) than deaths from cancer due to alcohol consumption (i.e., 401,000 deaths, globally) in 2019.¹⁰⁶ A possible reason for this discrepancy may be the widespread cancer campaigns and heightened awareness about the curability of cancer contribute to a higher level of concern and perceived vulnerability.¹⁰⁷ While information about the risks of alcohol consumption on heart diseases is mixed and inconsistent.¹⁰⁸ For instance, scholars¹⁹ found that people have misconceptions that drinking can lower the risk of coronary heart diseases. Such existing misinformation might be used by drinkers to justify their alcohol consumption, leading to decreased support for alcohol restriction policies.¹⁰⁹ The widespread misperception that alcohol, particularly wine, is beneficial for heart health can be traced back to the 1980s when Klatsky et al.¹¹⁰ first reported an association between moderate alcohol consumption and a lower risk of coronary heart disease. This finding was later popularized through the concept of the “French Paradox,” which suggested the French population, despite overindulged in rich and fatty food, had relatively low rates of heart disease due to regular and moderate wine consumption.¹¹¹ However, later research questions that argument, suggesting that what looks like drinking-related benefits might come from light and moderate drinkers often having a state of overall well-being that relies on higher incomes, healthier diet with longer relaxed meals, and exercising more than non-drinkers and heavy drinkers do.^112,113

A major contributor to the spread of this myth, the alcohol industry has frequently supported positive health claims with studies and lobbied the government to allow them on wine labels.¹¹³ Traditional media further amplified these messages. In the United States, the initial spread of the French Paradox narrative was accompanied by a marked increase in alcohol sales, particularly wine. More recently, alcohol brand placements in television shows and films have surged, making alcohol one of the most frequently featured product categories in 2025.^114,115 Advertisers disseminate misinformation about the purported healthy heart effects of alcohol through popular culture in traditional media, such as television shows portraying fictional characters consuming alcohol as high socioeconomic status, attractive, and glamorous.^{108,114,116,117}

Moreover, alcohol-related advertising may appear as often as every three minutes to those who are participating in eHealth activities, such as looking up health information online,¹¹⁸ making it challenging to distinguish between reliable health information and commercial content. This persistent exposure to misinformation can lead drinkers to justify their behavior and reduce public support for alcohol restriction policies.¹⁰⁹

Given the growing use of gray-area advertising approaches and the reliance on voluntary industry self-regulation, reassessing advertising policies, particularly in traditional media and social media, is urgent.¹¹⁹ It is crucial to deliver consistent and correct messages about the risks of alcohol consumption on heart disease, counteract misconceptions, and enhance health-related outcomes.¹⁰⁸ For instance, the effectiveness of public health campaigns in Canada, such as the “Alcohol can cause cancer” label, in raising awareness of alcohol-related cancer risks and reducing alcohol consumption¹²⁰ suggests that the government, the industry, and the market collectively should consider similar initiatives in the United States to conquer the misconception of alcohol consumption on heart health.

The moderating role of self-efficacy

The significant moderating role of self-efficacy indicates that individuals with high self-efficacy in health management are more likely to gain greater information about how alcohol causes cancer through eHealth activities, and consequently, increase their support for alcohol restriction policies, compared to those with low self-efficacy. This is consistent with previous research suggesting that eHealth activities can facilitate the process of health knowledge learning, and that self-efficacy can enhance the effectiveness of this process.^49,121 For individuals with high self-efficacy, using eHealth as a tool may increase their willingness and confidence in their ability to gain and comprehend useful information on alcohol and cancer risks, leading to greater support for alcohol restriction policies.¹²² It extends the social cognitive theory to the context of alcohol restriction, and also suggests that professionals develop interventions to enhance self-efficacy in individuals who are less confident in their ability to manage their health, such as providing education and support to help them gain the necessary skills and knowledge.

Differential effects of drinking levels on eHealth outcomes

Findings from the subgroup analysis reveal distinct patterns across drinking groups. The indirect pathway linking engagement in eHealth activities to support for alcohol-restriction policies via heightened perceptions of belief that alcohol causes cancer was significant among light and moderate drinkers, but it disappeared for heavy drinkers. In this latter group, eHealth activities did not shift risk perceptions, and even when beliefs were formed, those beliefs failed to translated into policy support. Such resistance echoes previous evidence that heavy drinkers defensively downplay risk information^63,71 and that impaired goal-directed control further disrupts the conversion of perceived risks into behavioral change.¹²³

The moderating role of self-efficacy displayed a similarly nuanced pattern. It did not moderate the link between eHealth activities and risk perceptions for either light or heavy drinkers, yet it significantly strengthened the connection between eHealth activities and beliefs about alcohol-related cancer and heart disease risks among moderate drinkers. Among light drinkers, elevated baseline self-efficacy for health management may produce a ceiling effect that constrains variance necessary for meaningful moderation.¹²⁴ Additionally, optimism bias leads light drinkers to perceive risk information as personally irrelevant, diminishing their motivation for behavioral change regardless of self-efficacy levels, thus neutralizing self-efficacy's potential moderating influence on eHealth-derived risk perceptions.^125,126

Heavy drinkers typically demonstrate low self-efficacy that remains resistant to enhancement due to alcohol's neuroadaptive effects.¹²⁷ Their compromised executive functioning may overwhelm standard eHealth effectiveness, leaving even high self-efficacy individuals within this group lacking sufficient cognitive resources for effective risk information processing.⁶⁹ Furthermore, heavy drinkers’ defensive processing patterns lead them to externalize alcohol-related risks, convincing themselves that such consequences affect others rather than themselves.⁶³ In this context, higher self-efficacy may paradoxically reinforce defensive responses, as individuals believe they can manage potential risks, ultimately rendering the moderating relationship non-significant.

The significant moderating role of self-efficacy among moderate drinkers suggests they operate within a psychologically optimal range for processing eHealth interventions. Unlike light drinkers who dismiss risk relevance, moderate drinkers acknowledge threat applicability while also avoiding the denial and cognitive impairment characteristic of heavy drinkers.^128,129 Their drinking patterns demonstrate greater behavioral plasticity, enabling self-efficacy to effectively facilitate both risk information acceptance and subsequent alcohol restriction behaviors.

Implications

The study possesses several theoretical implications. First, it demonstrates that eHealth activities enhance support for alcohol restriction policies primarily through increasing drinkers’ beliefs that alcohol causes cancer, while revealing a critical gap in the case of heart disease perceptions. This differential impact highlights a concerning misperception issue where drinkers remain largely unaware of alcohol's cardiovascular risks. Second, the study extends the Social Cognitive Theory to the alcohol control context by demonstrating how self-efficacy serves as a crucial moderator in the eHealth-policy support relationship. Individuals with higher self-efficacy in health management are better positioned to process and internalize cancer-related risk information from eHealth activities, leading to stronger support for restrictive policies. Third, the subgroup analysis further refines this theoretical framework by showing that eHealth interventions fail to influence risk perceptions or policy support among heavy drinkers, while self-efficacy moderates the eHealth–risk belief relationship mainly in moderate drinkers. This highlights a psychologically optimal zone where cognitive engagement and motivation align, suggesting that interventions targeting moderate drinkers with enhanced self-efficacy may be most effective.

This study also has practical implications. First, healthcare providers should prioritize cancer-focused alcohol risk communication in clinical settings as the primary pathway for building policy support. Findings suggest that formal healthcare appointments are uniquely effective for enhancing cancer risk awareness, with appointment scheduling serving as the most powerful eHealth activity for shifting alcohol-cancer beliefs. Providers should leverage these structured interactions to systematically discuss alcohol's effects across multiple cancer types. Given the important mediating role of beliefs that alcohol causes cancer, and the low level of awareness among US residents, policymakers are encouraged to disseminate information and knowledge about the link between drinking and cancer to increase support for alcohol restriction policies.

Second, healthcare providers must actively counteract persistent myths about alcohol's cardiovascular benefits, as the widespread belief represents a significant barrier to policy support. They should deploy clear, unambiguous messages using consistent, evidence-based statements across clinical consultations and media campaigns to directly refute the long-standing myth and highlight that even low-level drinking increases risks of heart diseases. Public health campaigns should separate alcohol-heart messaging from cancer messaging to avoid diluting the effectiveness of cancer-focused interventions, since heart disease beliefs fail to translate into policy support despite cardiovascular risks existing alongside cancer risks.

Third, healthcare professionals should routinely assess health management self-efficacy before implementing alcohol risk interventions, as low self-efficacy individuals show minimal benefit from standard eHealth activities. Health behavior programs should incorporate evidence-based self-efficacy enhancement techniques including mastery experiences, vicarious learning, and verbal persuasion. Digital health platforms should integrate self-efficacy building features rather than simply providing information access.

Fourth, based on results of the subgroup analysis, light drinkers represent an optimal target for straightforward cancer-focused eHealth interventions about alcohol use, as they show responsiveness to both cancer and heart disease risk information through eHealth activities, though only cancer beliefs translate to policy support. Moderate drinkers require intensive self-efficacy building as a prerequisite for effective eHealth interventions, as this group demonstrates the strongest interaction between self-efficacy and eHealth effectiveness, requiring interventions that combine structured self-efficacy enhancement with drinking risk education for maximum effectiveness. Heavy drinkers demonstrate resistance to standard eHealth interventions and require fundamentally different approaches. Resource allocation strategies should prioritize moderate drinkers receiving self-efficacy-enhanced interventions, followed by light drinkers receiving standard health education, while heavy drinkers require specialized, intensive interventions rather than general prevention programs.

Limitations

Several limitations should be noted. First, the use of cross-sectional data makes it difficult to determine causal relationships among variables. Future studies could employ longitudinal designs or experimental methods to replicate the current findings. Second, this study relied on secondary data from the HINTS survey, which, while robust and nationally representative, imposes constraints on the originality of research questions and the precision of measurement. For example, some constructs, such as self-efficacy, were measured using single-item indicators, eHealth activities were assessed within broad, predefined categories, and crucial factors such as health literacy were not included. These limitations may restrict the depth of theoretical elaboration. It is crucial that future research considers primary data collection that incorporates validated, multi-item scales and more targeted measures to facilitate a more nuanced understanding of the examined relationships. Third, we only tested the conditional effect of self-efficacy in health management. Future research should also examine self-efficacy in other domains relevant to eHealth activities, such as technology use. Fourth, for the two variables related to beliefs about the risks of alcohol, we recoded responses of “don’t know” as “no effect,” which may result in the loss of valuable information. Future studies are encouraged to explore better methods for integrating such responses into the analysis to capture a more accurate representation of participants’ beliefs. Fifth, eHealth activities were measured as general eHealth use rather than alcohol-specific eHealth activities in this study. Future research is encouraged to explore alcohol-specific eHealth use in relation to health beliefs and behaviors. Sixth, some effect sizes in this study are small, which indicates there might be additional potential predictors that need exploration. Future studies are encouraged to investigate different types of interventions and their associations with risk beliefs regarding alcohol use.

Conclusions

In the United States, excessive alcohol consumption and relevant restriction policies are growing social concerns. eHealth activities can be a significant intervention to encourage the public to support such policies. Specifically, the association between eHealth activities and support for alcohol restriction policies may be indirect through the mediated association with beliefs that alcohol causes cancer and may also be conditional regarding self-efficacy in health management. This study provides empirical evidence for the effectiveness of eHealth activities in improving alcohol restriction outcomes.

Footnotes

ORCID iDs

Zijun Chloe Wang

Minqin Ma

Ming Milano Li

Xinshu Zhao

Song Harris Ao

Ethical approval

This study used secondary data. The HINTS data meet established ethical standards and have obtained ethics approval.

Contributorship

Zijun Chloe Wang: writing—original draft preparation. Minqin Ma: Writing—reviewing and editing. Xi Xia: writing—reviewing and editing. Ming Milano Li: writing—reviewing and editing. Xinshu Zhao: funding acquisition. Song Harris Ao: writing—original draft preparation.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the University of Macau [grant numbers CRG2021-00002-ICI, ICI-RTO-0010-2021, CPG2022-00004-FSS, SRG2018-00143-FSS], Macau Higher Education Fund [grant number HSS-UMAC-2020-02], and the Fundamental Research Funds for the Central Universities, Sun Yat-sen University [grant number 17000-13230003].

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

References

Brand

Saisana

Rynn

, et al. Comparative analysis of alcohol control policies in 30 countries. PLoS Med 2007; 4: e151.

SAMHSA . 2019 National survey on drug use and health. Rockville: Center for Behavioral Health Statistics and Quality, 2020.

Rajendram

Lewison

Preedy

. Worldwide alcohol-related research and the disease burden. Alcohol Alcohol 2006; 41: 99–106.

Foster

Gmel

Mohler-Kuo

. Light and heavy drinking in jurisdictions with different alcohol policy environments. Int J Drug Policy 2019; 65: 86–96.

Greenfield

Cook

Karriker-Jaffe

, et al. The relationship between the US state alcohol policy environment and individuals’ experience of secondhand effects: alcohol harms due to others’ drinking. Alcohol Clin Exp Res 2019; 43: 1234–1243.

Room

Babor

Rehm

. Alcohol and public health. Lancet 2005; 365: 519–530.

Bates

Holmes

Gavens

, et al. Awareness of alcohol as a risk factor for cancer is associated with public support for alcohol policies. BMC Public Health 2018; 18: 1–11.

Buykx

Gilligan

Ward

, et al. Public support for alcohol policies associated with knowledge of cancer risk. Int J Drug Policy 2015; 26: 371–379.

Tobin

Moodie

Livingstone

. A review of public opinion towards alcohol controls in Australia. BMC Public Health 2011; 11: 1–9.

10.

Weerasinghe

Schoueri-Mychasiw

Vallance

, et al. Improving knowledge that alcohol can cause cancer is associated with consumer support for alcohol policies: findings from a real-world alcohol labelling study. Int J Environ Res Public Health 2020; 17: 398.

11.

Casswell

Thamarangsi

. Reducing harm from alcohol: call to action. Lancet 2009; 373: 2247–2257.

12.

Giesbrecht

Greenfield

Anglin

, et al. Changing the price of alcohol in the United States: perspectives from the alcohol industry, public health, and research. Contemp Drug Probl 2004; 31: 711–736.

13.

Greenfield

Bond

, et al. Risks of alcohol use disorders related to drinking patterns in the US general population. J Stud Alcohol Drugs 2014; 75: 319–327.

14.

Watson

Stapleton

Buykx , et al. Changes in public support for alcohol policies in NSW, Australia, 2013–2019. Public Health Res Pract 2022; 32: e31452118.

15.

Diepeveen

Ling

Suhrcke

, et al. Public acceptability of government intervention to change health-related behaviours: a systematic review and narrative synthesis. BMC Public Health 2013; 13: 1–11.

16.

Wilkinson

Room

Livingston

. Mapping Australian public opinion on alcohol policies in the new millennium. Drug Alcohol Rev 2009; 28: 263–274.

17.

Selbæk

, et al. Factors associated with consumption of alcohol in older adults-a comparison between two cultures, China and Norway: the CLHLS and the HUNT-study. BMC Geriatr 2017; 17: 1–10.

18.

Pechey

Burge

Mentzakis

, et al. Public acceptability of population-level interventions to reduce alcohol consumption: a discrete choice experiment. Soc Sci Med 2014; 113: 104–109.

19.

Arora

ElSayed

Beger

, et al. The impact of alcohol consumption on cardiovascular health: myths and measures. Glob Heart 2022; 17: 45.

20.

Carrà

Crocamo

Bartoli

, et al. Impact of a mobile eHealth intervention on binge drinking in young people: the digital–alcohol risk alertness notifying network for adolescents and young adults project. J Adolesc Health 2016; 58: 520–526.

21.

Benight

Shoji

Yeager

, et al. Predicting change in posttraumatic distress through change in coping self-efficacy after using the my trauma recovery eHealth intervention. JMIR Ment Health 2018; 5: e10309.

22.

Smit

Lokkerbol

Riper

, et al. Modeling the cost-effectiveness of health care systems for alcohol use disorders: how implementation of eHealth interventions improves cost-effectiveness. J Med Internet Res 2011; 13: e1694.

23.

Cristian

. Breast cancer and gynecological cancer rehabilitation. Amsterdam: Elsevier Health Sciences, 2020.

24.

Magsamen-Conrad

Wang

Tetteh

, et al. Using technology adoption theory and a lifespan approach to develop a theoretical framework for eHealth literacy: extending UTAUT. Health Commun 2020; 35: 1435–1446.

25.

Wilson

Malone

Hunt

, et al. Problematic alcohol use and hippocampal volume in a female sample: disentangling cause from consequence using a co-twin control study design. Psychol Med 2018; 48: 1673–1684.

26.

Chambers

Copson

Dutey-Magni

, et al. Alcohol use and breast cancer risk: a qualitative study of women’s perspectives to inform the development of a preventative intervention in breast clinics. Eur J Cancer Care 2019; 28: e13075.

27.

Buykx

Gavens

, et al. Public awareness of the link between alcohol and cancer in England in 2015: a population-based survey. BMC Public Health 2016; 16: 1–12.

28.

Piano

. Alcohol’s effects on the cardiovascular system. Alcohol Res 2017; 38: 219–241.

29.

Naimi

Stockwell

Zhao

, et al. Selection biases in observational studies affect associations between ‘moderate’ alcohol consumption and mortality. Addiction 2017; 112: 207–214.

30.

Greenfield

Kerr

, et al. Externalities from alcohol consumption in the 2005 US National Alcohol Survey: implications for policy. Int J Environ Res Public Health 2009; 6: 3205–3224.

31.

Ginossar

. Predictors of online cancer prevention information seeking among patients and caregivers across the digital divide: a cross-sectional, correlational study. JMIR Cancer 2016; 2: e5108.

32.

Park

Baeg

. Association of eHealth literacy with cancer information seeking and prior experience with cancer screening. Comput Inform Nurs 2014; 32: 458–463.

33.

Shim

Kelly

Hornik

. Cancer information scanning and seeking behavior is associated with knowledge, lifestyle choices, and screening. J Health Commun 2006; 11: 157–172.

34.

Hides

Quinn

Chan

, et al. Telephone-based motivational interviewing enhanced with individualised personality-specific coping skills training for young people with alcohol-related injuries and illnesses accessing emergency or rest/recovery services: a randomized controlled trial (QuikFix). Addiction 2021; 116: 474–484.

35.

Holmila

Mustonen

Österberg

, et al. Public opinion and community-based prevention of alcohol-related harms. Addict Res Theory 2009; 17: 360–371.

36.

Rehm

Gmel

Sr Gmel

, et al. The relationship between different dimensions of alcohol use and the burden of disease—an update. Addiction 2017; 112: 968–1001.

37.

Bagnardi

Rota

Botteri

, et al. Alcohol consumption and site-specific cancer risk: a comprehensive dose–response meta-analysis. Br J Cancer 2015; 112: 580–593.

38.

Zhao

Stockwell

Roemer

, et al. Is alcohol consumption a risk factor for prostate cancer? A systematic review and meta–analysis. BMC Cancer 2016; 16: 1–13.

39.

Corliss

. A closer look at alcohol’s effect on heart health, www.health.harvard.edu/heart-health/a-closer-look-at-alcohols-effect-on-heart-health (2023, accessed 13 August 2024).

40.

World Heart Federation . Cardiovascular health for everyone, www.world-heart-federation.org/world-heart-summit/world-heart-summit-2022/ (2022, accessed 13 August 2024).

41.

Roerecke

Rehm

. Alcohol consumption, drinking patterns, and ischemic heart disease: a narrative review of meta-analyses and a systematic review and meta-analysis of the impact of heavy drinking occasions on risk for moderate drinkers. BMC Med 2014; 12: 1–11.

42.

Christensen

ASP

Meyer

MKH

Dalum

, et al.

Can a mass media campaign raise awareness of alcohol as a risk factor for cancer and public support for alcohol related policies?

Prev Med 2019; 126: 105722.

43.

Seidenberg

Wiseman

Eck

, et al. Awareness of alcohol as a carcinogen and support for alcohol control policies. Am J Prev Med 2022; 62: 174–182.

44.

Calnan

Millar

Mongan

. Support for evidence-based alcohol policy in Ireland: results from a representative household survey. Eur J Public Health 2023; 33: 323–330.

45.

Bandura

. Self-efficacy: toward a unifying theory of behavioral change. Adv Behav Res Therapy 1978; 1: 139–161.

46.

Schwarzer

R, editor

. Self-efficacy: thought control in action . London: Routledge, 1992.

47.

Schwarzer

. Some burning issues in research on health behavior change. Appl Psychol 2008; 57: 84–93.

48.

Zhang

Han

Dang

, et al. User acceptance of mobile health services from users’ perspectives: the role of self-efficacy and response-efficacy in technology acceptance. Inform Health Soc Care 2017; 42: 194–206.

49.

Bandura

. Self-efficacy: the exercise of control. New York: Worth Publishers, 1997.

50.

Zhu

Chen

, et al. How does internet information seeking help academic performance?–the moderating and mediating roles of academic self-efficacy. Comput Educ 2011; 57: 2476–2484.

51.

Nie

Lau

Liau

. Role of academic self-efficacy in moderating the relation between task importance and test anxiety. Learn Individ Differ 2011; 21: 736–741.

52.

Lam

Amon

Nguyen

, et al. The effect of e-health contents on health science students’ attitude toward the efficiency of health ICT in care provision. In: Health informatics: building a healthcare future through trusted information. Amsterdam: IOS Press, 2012, pp.99–104.

53.

Keep

Janssen

McGregor

, et al. Mapping eHealth education: review of eHealth content in health and medical degrees at a metropolitan tertiary institute in Australia. JMIR Med Educ 2021; 7: e16440.

54.

Airola

. Learning and use of eHealth among older adults living at home in rural and nonrural settings: systematic review. J Med Internet Res 2021; 23: e23804.

55.

Kim

Xie

. Health literacy in the eHealth era: a systematic review of the literature. Patient Educ Couns 2017; 100: 1073–1082.

56.

Pan

Sun

Chow

IHS

. Leader-member exchange and employee creativity: test of a multilevel moderated mediation model. Hum Perform 2012; 25: 432–451.

57.

Schwarzer

Luszczynska

. How to overcome health-compromising behaviors: the health action process approach. Eur Psychol 2008; 13: 141–151.

58.

Marrie

Leung

Tyry

, et al. Use of eHealth and mHealth technology by persons with multiple sclerosis. Mult Scler Relat Disord 2019; 27: 13–19.

59.

Marzan

Callinan

Livingston

, et al. Alcohol consumption, heavy episodic drinking and the perpetration of antisocial behaviours in Australia. Drug Alcohol Depend 2022; 235: 109432.

60.

Spada

Wells

. Metacognitions about alcohol use in problem drinkers. Clin Psychol Psychother 2006; 13: 138–143.

61.

Kelly

Olanrewaju

Cowan

, et al. Alcohol and older people: a systematic review of barriers, facilitators and context of drinking in older people and implications for intervention design. PLoS One 2018; 13: e0191189.

62.

Wiers

Van Woerden

Smulders

, et al. Implicit and explicit alcohol-related cognitions in heavy and light drinkers. J Abnorm Psychol 2002; 111: 648.

63.

Morris

Tattan-Birch

, et al. Look away now! defensive processing and unrealistic optimism by level of alcohol consumption. Psychol Health 2024; 12: 1–9.

64.

Kolind

. Young people’s alcohol use is still strongly related to social inclusion. Addiction 2022; 117: 1216.

65.

Pennington

Shaw

Adams

, et al. Where's the wine? Heavy social drinkers show attentional bias towards alcohol in a visual conjunction search task. Addiction 2020; 115: 1650–1659.

66.

Anthenelli

Grandison

. Effects of stress on alcohol consumption. Alcohol Res 2012; 34: 381–382.

67.

Frone

. Work stress and alcohol use: developing and testing a biphasic self-medication model. Work Stress 2016; 30: 374–394.

68.

Lovatt

Eadie

, et al. Public attitudes towards alcohol control policies in Scotland and England: results from a mixed-methods study. Soc Sci Med 2017; 177: 177–189.

69.

Montgomery

Fisk

Murphy

, et al. The effects of heavy social drinking on executive function: a systematic review and meta-analytic study of existing literature and new empirical findings. Hum Psychopharmacol 2012; 27: 187–199.

70.

Bujalski

Stokwiszewski

. Alcohol risk perception and the risk target. Results from a survey among the Polish adult population. Addict Res Theory 2022; 30: 253–261.

71.

Alvarez-Bueno

Rodriguez-Martin

Garcia-Ortiz

, et al. Effectiveness of brief interventions in primary health care settings to decrease alcohol consumption by adult non-dependent drinkers: a systematic review of systematic reviews. Prev Med 2015; 76: S33–S38.

72.

Woods

Porges

Bryant

, et al. Current heavy alcohol consumption is associated with greater cognitive impairment in older adults. Alcohol Clin Exp Res 2016; 40: 2435–2444.

73.

King

Hasin

O’Connor

, et al. A prospective 5-year re-examination of alcohol response in heavy drinkers progressing in alcohol use disorder. Biol Psychiatry 2016; 79: 489–498.

74.

Andrade

de Lacerda

Gomide

, et al. Web-based self-help intervention reduces alcohol consumption in both heavy-drinking and dependent alcohol users: a pilot study. Addict Behav 2016; 63: 63–71.

75.

Radtke

Ostergaard

Cooke

, et al. Web-based alcohol intervention: study of systematic attrition of heavy drinkers. J Med Internet Res 2017; 19: e6780.

76.

Bonar

Bauermeister

Blow

, et al. A randomized controlled trial of social media interventions for risky drinking among adolescents and emerging adults. Drug Alcohol Depend 2022; 237: 109532.

77.

Boyle

LaBrie

Baez

, et al. Integrating social media inspired features into a personalized normative feedback intervention combats social media-based alcohol influence. Drug Alcohol Depend 2021; 228: 109007.

78.

Mukamal

Chen

Rao

, et al. Alcohol consumption and cardiovascular mortality among US adults, 1987 to 2002. J Am Coll Cardiol 2010; 55: 1328–1335.

79.

Veeranki

Zhao

, et al. Relationship of alcohol consumption to all-cause, cardiovascular, and cancer-related mortality in US adults. J Am Coll Cardiol 2017; 70: 913–922.

80.

Kontos

Blake

Chou

, et al. Predictors of eHealth usage: insights on the digital divide from the Health Information National Trends Survey 2012. J Med Internet Res 2014; 16: e172.

81.

Chen

Feeley

. Numeracy, information seeking, and self-efficacy in managing health: an analysis using the 2007 Health Information National Trends Survey (HINTS). Health Commun 2014; 29: 843–853.

82.

Xie

Zhang

Deng

. Self-efficacy and work motivation among inclusive education teachers in China. Int J Disabil Dev Educ 2024; 71: 236–250.

83.

Hayes

. Partial, conditional, and moderated mediation: quantification, inference, and interpretation. Commun Monogr 2018; 85: 4–40.

84.

Slater

Lawrence

Comello

MLG

. Media influence on alcohol-control policy support in the US adult population: the intervening role of issue concern and risk judgments. J Health Commun 2009; 14: 262–275.

85.

Oosterveen

Tzelepis

Ashton

, et al. A systematic review of eHealth behavioral interventions targeting smoking, nutrition, alcohol, physical activity and/or obesity for young adults. Prev Med 2017; 99: 197–206.

86.

Wernhart

Gahbauer

Haluza

. Ehealth and telemedicine: practices and beliefs among healthcare professionals and medical students at a medical university. PLoS One 2019; 14: e0213067.

87.

Wadsworth

Archibald

Payne

, et al. Shared medical appointments and patient-centered experience: a mixed-methods systematic review. BMC Fam Pract 2019; 20: 1–13.

88.

Street

Haidet

. How well do doctors know their patients? Factors affecting physician understanding of patients’ health beliefs. J Gen Intern Med 2011; 26: 21–27.

89.

Matulis

McCoy

. Patient-centered appointment scheduling: a call for autonomy, continuity, and creativity. J Gen Intern Med 2021; 36: 511–514.

90.

Claggett

Kitchens

Paino

. Identifying the peripheral cues in the credibility assessment of online health information. Inf Manag 2024; 61: 104037.

91.

Rimal

Real

. Perceived risk and efficacy beliefs as motivators of change: use of the risk perception attitude (RPA) framework to understand health behaviors. Hum Commun Res 2003; 29: 370–399.

92.

Wang

Shi

Kong

. Online health information seeking: a review and meta-analysis. Health Commun 2021; 36: 1163–1175.

93.

Finney Rutten

Hesse

St. Sauver

, et al. Health self-efficacy among populations with multiple chronic conditions: the value of patient-centered communication. Adv Ther 2016; 33: 1440–1451.

94.

Chang

Zhang

Gwizdka

. The effects of information source and eHealth literacy on consumer health information credibility evaluation behavior. Comput Human Behav 2021; 115: 106629.

95.

Sbaffi

Rowley

. Trust and credibility in web-based health information: a review and agenda for future research. J Med Internet Res 2017; 19: e218.

96.

Bandura

. Human agency in social cognitive theory. Am Psychol 1989; 44: 1175.

97.

Wei

. Self-efficacy, information-processing strategies, and acquisition of health knowledge. Asian J Commun 2013; 23: 54–67.

98.

Swar

Hameed

Reychav

. Information overload, psychological ill-being, and behavioral intention to continue online healthcare information search. Comput Human Behav 2017; 70: 416–425.

99.

National Cancer Institute . Alcohol and cancer risk, www.cancer.gov/about-cancer/causes-prevention/risk/alcohol/alcohol-fact-sheet#:∼:text=Alcohol%20consumption%20has%20also%20been,the%20harms%20of%20alcohol%20consumption (2021, accessed 13 August 2024).

100.

Rehm

Shield

. Alcohol and mortality: global alcohol-attributable deaths from cancer, liver cirrhosis, and injury in 2010. Alcohol Res 2014; 35: 174.

101.

Petticrew

Maani

Knai

, et al. How alcohol industry organisations mislead the public about alcohol and cancer. Drug Alcohol Rev 2018; 37: 293–303.

102.

Usher-Smith

Silarova

Ward

, et al. Incorporating cancer risk information into general practice: a qualitative study using focus groups with health professionals. Br J Gen Pract 2017; 67: e218–e226.

103.

Zhao

Lynch

Jr Chen

. Reconsidering baron and kenny: myths and truths about mediation analysis. J Consum Res 2010; 37: 197–206.

104.

Jiao

Schulz

Chang

. Addressing the role of eHealth literacy in shaping popular attitudes towards post-COVID-19 vaccination among Chinese adults. Hum Soc Sci Commun 2024; 11: 1–11.

105.

Bareham

Kaner

Hanratty

. Managing older people’s perceptions of alcohol-related risk: a qualitative exploration in northern English primary care. Br J Gen Pract 2020; 70: e916–e926.

106.

World Health Organization . Alcohol, www.who.int/news-room/fact-sheets/detail/alcohol (2024, accessed 13 August 2024).

107.

Sahu

Subba

Giri

. Cancer awareness and attitude towards cancer screening in India: a narrative review. J Family Med Prim Care 2020; 9: 2214–2218.

108.

Welfordsson

Danielsson

Bjorck

, et al. Mixed messages? Exposure to reports about alcohol’s suggested cardiovascular effects and hazardous alcohol use: a cross-sectional study of patients in cardiology care. BMC Public Health 2024; 24: 1302.

109.

Whitman

Pletcher

Vittinghoff

, et al. Perceptions, information sources, and behavior regarding alcohol and heart health. Am J Cardiol 2015; 116: 642–646.

110.

Klatsky

Friedman

Siegelaub

. Alcohol consumption before myocardial infarction: results from the Kaiser-Permanente epidemiologic study of myocardial infarction. Ann Intern Med 1974; 81: 294–301.

111.

Ellison

. The French paradox: 20 years later. J Wine Res 2011; 22: 105–108.

112.

Willett

. Diet and breast cancer. J Med Internet Res 2001; 249: 395–411.

113.

Mudry

. The poison is in the dose: the French paradox, the healthy drinker, and the medicalization of virtue. Food Cult Soc 2010; 13: 91–114.

114.

BrandTrack.Pro . Top brands and product placements in movies & tv – April 2025 report, https://brandtrack.pro/top-brands-and-product-placements-in-movies-tv-april-2025-report/ (2025, accessed 2 May 2025).

115.

Hines

. Alcohol brand placement in movies has doubled over the last decade. VinePair, 1 May 2019.

116.

Grube

. Alcohol in the media: drinking portrayals, alcohol advertising, and alcohol consumption among youth. In: Bonnie

Schwarzer

(eds) Reducing underage drinking: a collective responsibility. Washington: National Academies Press, 2004, pp.597–624

117.

Medling

Gobeil

Sawalha

, et al. Relation of patient’s opinion of alcohol’s health effects and drinking habits among hospitalized patients with cardiovascular disease. Am J Cardiol 2022; 179: 31–38.

118.

Rutherford

Leung

Stjepanović

, et al. Through the looking glass: an alcohol advertisement every 3 min. Drug Alcohol Rev 2024; 43: 1426–1434.

119.

Krnel

Levičnik

van Dalen

, et al. Effectiveness of regulatory policies on online/digital/internet-mediated alcohol marketing: a systematic review. J Epidemiol Glob Health 2023; 13: 115–128.

120.

Hobin

Shokar

Vallance

, et al. Communicating risks to drinkers: testing alcohol labels with a cancer warning and national drinking guidelines in Canada. Can J Public Health 2020; 111: 716–725.

121.

Lam

Tracz

Lucey

. Age, gender, and ethnicity of counsellor trainees and corresponding counselling self-efficacy: research findings and implications for counsellor educators. Int J Adv Couns 2013; 35: 172–187.

122.

Anker

Reinhart

Feeley

. Health information seeking: a review of measures and methods. Patient Educ Couns 2011; 82: 346–354.

123.

Baumann

Gaertner

Schnuerer

, et al. Belief incongruence and the intention–behavior gap in persons with at-risk alcohol use. Addict Behav 2015; 48: 5–11.

124.

DiBello

Miller

Carey

. Self-efficacy to limit drinking mediates the association between attitudes and alcohol-related outcomes. Subst Use Misuse 2019; 54: 2400–2408.

125.

Choi

Myung

Lee

. Light alcohol drinking and risk of cancer: a meta-analysis of cohort studies. Cancer Res Treat 2018; 50: 474–487.

126.

Waller

Wagner

. The sensitivity to threat and affiliative reward (STAR) model and the development of callous-unemotional traits. Neurosci Biobehav Rev 2019; 107: 656–671.

127.

Gause

Elliott

Delker

, et al. Association between change in self-efficacy to resist drinking and drinking behaviors among an HIV-infected sample: results from a large randomized controlled trial. J Health Psychol 2018; 23: 829–839.

128.

Ponizovskiy

Gofman

. Cognitive status and addiction denial in the early stages of alcohol addiction. Eur Psychiatry 2017; 41: s874–s874.

129.

Winstock

Holmes

Ferris

, et al. Perceptions of alcohol health warning labels in a large international cross-sectional survey of people who drink alcohol. Alcohol Alcohol 2020; 55: 315–322.