Abstract
Introduction:
Alcohol use increases cancer risk, yet awareness of this association is low. Alcohol control policies have the potential to reduce alcohol-caused cancer morbidity and mortality. Research outside the U.S. has found awareness of the alcohol–cancer link to be associated with support for alcohol control policies. The purpose of this study is to estimate the prevalence of support for 3 communication-focused alcohol policies and examine how awareness of the alcohol–cancer link and drinking status are associated with policy support among U.S. residents.
Methods:
Investigators analyzed data from the 2020 Health Information National Trends Survey 5 Cycle 4. Analyses were performed in 2021. The proportion of Americans who supported banning outdoor alcohol advertising and adding warning labels and drinking guidelines to alcohol containers was estimated. Weighted multivariable logistic regression was used to examine how awareness of the alcohol–cancer link and drinking status were associated with policy support.
Results:
Most Americans supported adding warning labels (65.1%) and drinking guidelines (63.9%), whereas only 34.4% supported banning outdoor alcohol advertising. Americans reporting that alcohol had no effect/decreased cancer risk had lower odds of support for advertising ban (OR=0.56), warning labels (OR=0.43), and guidelines (OR=0.46) than Americans aware of the alcohol–cancer link. Moreover, heavier drinkers had lower odds of support for advertising ban (OR=0.41), warning labels (OR=0.59), and guidelines (OR=0.60) than nondrinkers.
Conclusions:
Awareness of the alcohol–cancer link was associated with policy support. Increasing public awareness of the alcohol–cancer link may increase support for alcohol control policies.
INTRODUCTION
An estimated 3 million deaths globally were caused by alcohol consumption in 2016, including 378,000 (12.6%) alcohol-attributable cancer deaths.1 Alcohol consumption increases the risk of 7 cancer types, including cancers of the breast, mouth, and colon.2–4 All beverage types containing ethanol increase cancer risk in a dose–response relationship, and consumption even at low levels increases the risk of some cancers.2,5 Moreover, alcohol cessation is associated with decreased cancer risk over time.6–8 Therefore, policies designed to reduce alcohol consumption may be effective cancer prevention strategies.9
In the U.S., alcohol contributed to an average of >75,000 cancer cases and almost 19,000 cancer deaths per year between 2013 and 2016.10 However, despite this significant cancer burden, awareness of alcohol’s carcinogenic risks is suboptimal.11 For instance, a 2017 national survey found that only 38% of Americans were aware that drinking too much alcohol could lead to cancer.12 Similarly, a 2017 survey by the American Institute for Cancer Research found that 39% of Americans were aware that alcohol increased cancer risk.13 In the same survey, 93% and 80% of Americans were aware that tobacco and asbestos, respectively, increased cancer risk.13 Studies monitoring alcohol consumption and the related harms also reveal increasing alcohol use, high-risk drinking, and alcohol use disorders among Americans.14–16 Interventions to mitigate these trends are needed, and raising awareness of the alcohol–cancer link may offer a promising new strategy to increase the motivation to reduce alcohol consumption.17
Improving awareness about the link between alcohol and cancer may also help to increase support for alcohol control policies, which in turn may reduce the population risk for alcohol-related morbidity and mortality, including from cancer.9 For instance, Buykx et al.18 found that awareness of the alcohol–cancer link was associated with greater odds of support for 7 alcohol control policies (e.g., increasing price, health warnings) among adults in New South Wales, Australia. Similarly, Bates and colleagues19 surveyed adults in England and found that awareness of the carcinogenic effects of alcohol was associated with support for a variety of alcohol control policies, including pricing and marketing policies. In addition, Weerasinghe et al.20 used a quasi-experimental design to examine the impacts of adding cancer warnings to alcoholic beverages in Yukon, Canada. The researchers found that knowledge of the alcohol–cancer link was associated with greater odds of support for regulating alcohol pricing, availability, and marketing. Furthermore, a media campaign in Denmark describing the alcohol–cancer link increased both awareness and support for minimum unit pricing, a ban on alcohol advertising, and mandatory nutrition labeling.21 To date, no known research has examined the relationship between awareness of the alcohol–cancer link and support for alcohol control policies in the U.S. population.
Elucidating the factors that influence the support for alcohol control policies is vital to policymakers because policy support helps to shape political will, an essential component of public health policy implementation.22–24 Although some work has been done to examine the public support for alcohol taxation and sales restrictions in the U.S., less is known about the support for communication-related policy measures such as advertising restrictions and product labeling, which have the potential to influence consumption behaviors.25 Moreover, measuring the support for communication-focused alcohol policies is timely because several public health organizations recently (October 2020) submitted a petition to the Alcohol and Tobacco Tax Trade Bureau encouraging new cancer-specific health warnings for alcoholic beverages.26 The aims of this study are to estimate population-level support for communication-related alcohol control policies and examine how awareness of the alcohol–cancer link and drinking status are associated with policy support in a U.S. sample.
METHODS
Study Sample
Investigators analyzed data from the Health Information National Trends Survey (HINTS) 5 Cycle 4 (2020), a cross-sectional, nationally representative postal survey that is administered to assess Americans’ access to, need for, and use of cancer and health information and technology. HINTS used a 2-stage sampling design. In the first stage, a stratified random sample of addresses was selected, with an oversample of addresses from a stratum containing high-minority census tracts. In the second stage, 1 adult was selected from each selected household. A detailed description of the HINTS design has been published elsewhere.27,28 The survey was administered February 27, 2020–June 15, 2020.
Measures
Support for 3 communication-related alcohol policies was measured: (1) banning outdoor alcohol advertising, (2) requiring health warnings on alcoholic beverage containers, and (3) requiring recommended drinking guidelines on alcoholic beverage containers. Policy support was measured by asking: To reduce the problems associated with excessive alcohol use, to what extent would you support or oppose... with each policy listed in a matrix grid. Response options were strongly oppose, oppose, neither support nor oppose, support, and strongly support. Responses were dichotomized to support (support/strongly support) versus all other responses.
Awareness of the alcohol–cancer link was measured separately for wine, beer, and liquor by asking: In your opinion, how much does drinking the following types of alcohol affect the risk of getting cancer? Responses were decreases risk a lot, decreases risk a little, no effect, increases risk a little, increases risk a lot, and don’t know. Responses were coded as increases risk (a little/a lot), no effect/decreases risk (no effect/decreases risk a little/a lot), and don’t know. Responses of no effect/decreases risk were combined because both responses are incorrect and because <3% reported decreased risk for liquor and beer. The 3 alcohol–cancer awareness items were highly correlated (r=0.87–0.90). Therefore, the authors created a composite alcohol–cancer awareness item to serve as the primary predictor. If respondents reported that any of the 3 beverage types increased cancer risk, they were categorized as increases risk. Among those remaining, respondents reporting no effect/decreased risk for any beverage type were categorized as no effect/decreased risk. All remaining respondents reported don’t know for all items and were categorized accordingly.
Current drinking was measured using 3 items: (1) During the past 30 days, how many days per week did you have at least one drink of any alcoholic beverage?; (2) During the past 30 days, on the days when you drank, about how many drinks did you drink on average?; and (3) During the past 30 days, how many times did you have [5 for male respondents, 4 for female respondents] or more alcoholic drinks on one occasion? A figure pictorially showing 1 drink equivalents of beer (12 fluid ounces), malt liquor (8–9 fluid ounces), wine (5 fluid ounces), and 80-proof distilled spirits (1.5 fluid ounces per shot) was displayed before these items. Using the 2020–2025 U.S. Dietary Guidelines for Americans,29 participants were categorized as heavier drinkers (consuming >1 drink per day or ≥4 drinks on a single occasion for female respondents or >2 drinks per day or ≥5 drinks on a single occasion for male respondents), drinkers (consuming ≤1 drink per day and not consuming ≥4 drinks on a single occasion for female respondents or ≤2 drinks per day and not consuming ≥5 drinks on a single occasion for male respondents), and noncurrent drinkers (no past 30 –day drinking), referred to as nondrinkers in the remaining part of this paper.
Sociodemographic measures included sex, age, race/ethnicity, education, and income perceptions (e.g., living comfortably or finding it difficult on present income). Given the high rates of drinking among cigarette smokers, this study also controlled for current smoking status. Furthermore, because previous research suggests that significant variance in policy support is explained by political ideology or party identification, political viewpoint was included.30,31 All models also adjusted for survey return time stamped before or after the coronavirus disease 2019 (COVID-19) pandemic was declared on March 11, 2020.
Statistical Analysis
All analyses were performed in 2021 using StataSE, version 16. Analyses were weighted with sample weights, and 50 jackknife replicate weights were applied to compute design-adjusted SEs. The design-corrected Pearson chi-square/(second-order correction of Rao and Scott32) was used to assess bivariate relationships. A total of 3 weighted multivariable logistic regression models were run to examine the relationship of awareness between the alcohol–cancer link, drinking status, and other covariates with policy support. Investigators looked for evidence of multicollinearity first by including all the 3 alcohol–cancer awareness items (i.e., wine, beer, liquor) with predictor variables and covariates. Valence inflation factor values were 5.25 for liquor, 6.56 for beer, and 6.14 for wine–cancer awareness items. After replacing the 3 alcohol–cancer awareness items with the composite alcohol–cancer awareness item, the variance inflation factor for the composite variable was 1.09 (variance inflation factor values for all other predictors were ≤1.17). Therefore, the authors included a composite measure of alcohol and cancer awareness in all models.
This study also examined whether the relationship between awareness of the alcohol–cancer link and policy support was moderated by drinking status. Interactions between awareness and drinking status yielded 4 interaction terms in each model (Appendix Table 1, available online). Of the total 12 interaction terms (across 3 models), 11 of 12 were nonsignificant. Therefore, all interaction terms were removed, and results are reported with no moderation.
A total of 3 sensitivity tests were conducted. The authors first reran all models using the missing indicator method, which involved creating a response option for missingness for all variables. This approach maximizes sample size by retaining observations with missing data. Missingness for all variables ranged from 0% to 9.9% and was highest for political viewpoint (9.9%), Hispanic ethnicity (9.2%), and drinking status (8.4%). Next, each model was rerun, replacing the composite awareness item with the 3 separate awareness items. This was conducted for both listwise deletion and the missing indicator approach. Finally, each model was rerun 3 times, and the composite awareness item was replaced with 1 of the 3 awareness items (e.g., wine–cancer awareness item). Similarly, this was run using both listwise deletion and the missing indicator approach.
Sensitivity analysis results are reported in Appendix Tables 2 and 3 (available online). Listwise deletion and missing indicator approaches yielded similar findings; therefore, results from listwise deletion are reported. Similarly, findings from the inclusion of a composite variable of awareness and a single measure of awareness for a single beverage type yielded comparable results. Results are reported from models using the composite awareness variable to avoid multicollinearity and because this approach more robustly captures awareness. The HINTS 5 was given a non-human subjects determination by the NIH Office of Human Subjects Research through exemption #13204 on April 25, 2016.
RESULTS
A total of 3,865 adults participated in HINTS 5 Cycle 4 (response rate=36.66% using the American Association for Public Opinion Research Response Rate 4 formula27), representing a population estimate of 253,815,197 Americans. Most Americans supported requiring health warnings (65.1%) and drinking guidelines (63.9%) on alcoholic beverage containers. Fewer Americans (34.4%) supported banning outdoor alcohol advertising. Approximately half of Americans (52.1%) were nondrinkers; 19.6% were (past 30–day) drinkers, and 28.3% were heavier drinkers.
Awareness of the alcohol–cancer link was low, with 31.8% reporting an increased risk (composite variable). By individual beverage type, 20.3%, 24.9%, and 31.2% of Americans were aware that wine, beer, and liquor, respectively, increased cancer risk. Moreover, approximately half of participants responded don’t know (wine: 51.8%, beer: 54.0%, liquor: 51.6%) to the 3 awareness items, and 48.7% were coded as don’t know using the composite variable. Table 1 includes weighted point estimates and 95% CIs for support for each policy overall and by sociodemographic characteristics.
Table 1.
Unadjusted, Weighted Proportion of Americans Supporting Alcohol Control Policies Overall and by Sociodemographic Characteristics
| Characteristics | Outdoor alcohol advertising ban |
Health warning labels |
Drinking guidelines |
|||
|---|---|---|---|---|---|---|
| Weighted estimate (95% CI) | p-value | Weighted estimate (95% CI) | p-value | Weighted estimate (95% CI) | p-value | |
|
| ||||||
| Overall | 34.4 (31.9, 37.1) | — | 65.1 (62.6, 67.5) | — | 63.9 (61.4, 66.4) | — |
| Alcohol-cancer risk beliefs | <0.001 | <0.001 | <0.001 | |||
| Increase risk | 37.8 (33.4, 42.5) | 75.4 (70.8, 79.4) | 72.6 (68.2, 76.7) | |||
| No effect/decrease risk | 23.6 (19.4, 28.3) | 56.1 (48.3, 63.6) | 53.2 (45.9, 60.4) | |||
| Don’t know | 36.5 (32.7, 40.4) | 61.9 (58.9, 64.9) | 62.5 (59.6, 65.3) | |||
| Current drinking | <0.001 | 0.003 | <0.001 | |||
| Nondrinker | 42.8 (38.3, 47.4) | 70.2 (65.4, 74.5) | 70.2 (66.0, 74.1) | |||
| Drinker | 28.6 (23.4, 34.5) | 64.8 (59.0, 70.3) | 59.6 (53.2, 65.7) | |||
| Heavier drinker | 21.9 (18.8, 25.4) | 56.9 (50.9, 62.7) | 56.7 (51.2, 62.0) | |||
| Sex | 0.002 | 0.014 | <0.001 | |||
| Female | 39.3 (36.2, 42.4) | 69.5 (66.3, 72.5) | 70.1 (67.1, 72.9) | |||
| Male | 30.0 (25.7, 34.6) | 61.3 (56.4, 66.0) | 58.1 (53.8, 62.4) | |||
| Age, years | <0.001 | 0.817 | 0.203 | |||
| 18–39 | 32.8 (27.5, 38.6) | 66.6 (60.0, 72.6) | 68.0 (61.3, 74.1) | |||
| 40–59 | 30.6 (27.1, 34.3) | 65.5 (61.7, 69.1) | 61.7 (57.4, 65.9) | |||
| ≥60 | 42.3 (39.4, 45.3) | 64.7 (61.3, 67.9) | 63.3 (60.1, 66.4) | |||
| Race | 0.007 | 0.134 | 0.085 | |||
| White | 32.2 (29.3, 35.2) | 63.9 (61.4, 66.3) | 61.7 (58.8, 64.5) | |||
| Black | 42.4 (35.3, 50.0) | 65.7 (58.4, 72.3) | 68.7 (61.0, 75.4) | |||
| Other | 42.3 (33.9, 51.2) | 72.3 (63.8, 79.4) | 68.8 (60.6, 76.0) | |||
| Ethnicity | 0.939 | 0.207 | 0.090 | |||
| Non-Hispanic | 34.5 (31.5, 37.8) | 64.7 (62.0, 67.3) | 62.8 (60.2, 65.3) | |||
| Hispanic | 34.3 (28.3, 40.8) | 69.7 (62.4, 76.1) | 69.3 (61.9, 75.8) | |||
| Education | 0.234 | 0.018 | 0.009 | |||
| ≤High school/technical school | 35.8 (30.9, 40.9) | 60.8 (55.5, 65.8) | 59.2 (53.7, 64.5) | |||
| Some college | 31.0 (26.4, 36.0) | 66.8 (62.7, 70.7) | 64.1 (59.5, 68.4) | |||
| ≥College degree | 36.2 (32.4, 40.2) | 69.4 (65.8, 72.8) | 69.9 (66.0, 73.5) | |||
| Income perceptions | 0.032 | 0.485 | 0.578 | |||
| Living comfortably | 30.3 (26.9, 33.9) | 63.1 (59.0, 67.0) | 62.3 (58.0, 66.5) | |||
| Getting by | 36.8 (32.9, 40.9) | 67.1 (62.8, 71.2) | 66.0 (61.3, 70.4) | |||
| Finding it difficult | 36.5 (31.0, 42.4) | 66.4 (58.2, 73.7) | 64.4 (57.1, 71.0) | |||
| Current smoking | 0.039 | 0.024 | 0.031 | |||
| Nonsmoker | 35.4 (32.7, 38.1) | 67.0 (64.3, 69.0) | 65.3 (62.8, 67.7) | |||
| Smoker | 27.7 (21.5, 34.9) | 56.0 (46.4, 65.2) | 55.6 (46.4, 64.4) | |||
| Political ideology | 0.556 | 0.069 | 0.004 | |||
| Liberal | 36.5 (31.4, 41.9) | 71.3 (65.2, 76.7) | 72.2 (66.9, 77.0) | |||
| Moderate | 34.0 (29.0, 39.3) | 64.4 (59.7, 68.9) | 62.2 (57.0, 67.2) | |||
| Conservative | 33.0 (29.5, 36.7) | 62.5 (57.5, 67.2) | 60.1 (55.7, 64.3) | |||
Note: Boldface indicates p<0.05.
Current drinking is defined as past 30-day drinking. p-values are from design-corrected Pearson chi-square with the second-order correction of Rao and Scott.32 Drinkers consumed ≤1 drink per day and did not consume ≥4 drinks on a single occasion for women and ≤2 drinks per day and did not consume ≥5 drinks on a single occasion for men. Heavier drinkers consumed >1 drink per day or consumed ≥4 drinks on a single occasion for women and >2 drinks per day or consumed ≥5 drinks on a single occasion for men.
Across all the 3 policies, support was higher among Americans aware that alcohol increased cancer risk than among those responding no effect/decreases risk. For example, 75.4% and 72.6% of Americans aware that alcohol increases cancer risk supported adding health warnings and drinking guidelines on beverage containers compared with 56.1% and 53.2% among those responding no effect/decreases risk, respectively. Similarly, 37.8% of those aware supported banning outdoor alcohol advertising, compared with 23.6% of those responding no effect/decreases risk. Policy support among Americans responding don’t know fell in between those responding increases risk and no effect/decreases risk, with 36.5% supporting advertising bans, 61.9% supporting warnings, and 62.5% supporting guidelines.
Policy support was highest among nondrinkers, followed by drinkers, and was lowest among heavier drinkers. For instance, 42.8% of nondrinkers supported restrictions on outdoor alcohol advertising, compared with 28.6% of drinkers and 21.9% of heavier drinkers. Similarly, 70.2% of nondrinkers, 64.8% of drinkers, and 56.9% of heavier drinkers supported adding health warning labels on alcoholic beverages. Examining unadjusted support by demographic characteristics, support was generally higher among females than among males, among non-Whites than among Whites, among those with higher educational attainment, and among nonsmokers than among smokers.
Table 2 reports the adjusted odds of supporting the 3 alcohol control policies. Across all 3 policies, Americans responding that alcohol has no effect/decreases risk had lower odds of support than those responding that alcohol increases cancer risk (advertising ban: OR=0.56, warnings: OR=0.43, guidelines: OR=0.46). Americans responding don’t know also had lower odds of supporting warnings (OR=0.54) and guidelines (OR=0.63) than those aware of the alcohol–cancer link.
Table 2.
Weighted Odds of Supporting Communication-Related Alcohol Control Policies
| Variables | Outdoor alcohol advertising ban (n=2,817) |
Health warning labels (n=2,821) |
Drinking guidelines (n=2,819) |
|||
|---|---|---|---|---|---|---|
| Weighted OR (95% CI) | p-value | Weighted OR (95% CI) | p-value | Weighted OR (95% CI) | p-value | |
|
| ||||||
| Alcohol-cancer risk beliefs | ||||||
| Increase risk | ref | ref | ref | |||
| No effect/decrease risk | 0.56 (0.38, 0.81) | 0.003 | 0.43 (0.26, 0.71) | 0.002 | 0.46 (0.29, 0.74) | 0.002 |
| Don’t know | 0.84 (0.61,1.17) | 0.290 | 0.54 (0.41, 0.72) | <0.001 | 0.63 (0.46, 0.86) | 0.004 |
| Current drinking | ||||||
| Nondrinker | ref | ref | ref | |||
| Drinker | 0.60 (0.40, 0.88) | 0.010 | 0.86 (0.55, 1.34) | 0.496 | 0.69 (0.46, 1.04) | 0.073 |
| Heavier drinker | 0.41 (0.29, 0.57) | <0.001 | 0.59 (0.37, 0.93) | 0.023 | 0.60 (0.41, 0.87) | 0.009 |
| Sex | ||||||
| Female | ref | ref | ref | |||
| Male | 0.68 (0.50, 0.93) | 0.019 | 0.75 (0.52, 1.07) | 0.113 | 0.64 (0.47, 0.88) | 0.006 |
| Age, years | ||||||
| 18–39 | ref | ref | ref | |||
| 40–59 | 0.91 (0.62, 1.34) | 0.635 | 1.10 (0.71, 1.72) | 0.652 | 0.91 (0.59, 1.42) | 0.674 |
| ≥60 | 1.62 (1.12, 2.35) | 0.011 | 1.11 (0.74, 1.68) | 0.605 | 1.01 (0.65, 1.57) | 0.958 |
| Race | ||||||
| White | ref | ref | ref | |||
| Black | 1.51 (0.96, 2.38) | 0.075 | 1.17 (0.74, 1.86) | 0.491 | 1.50 (0.88, 2.55) | 0.130 |
| Other | 1.35 (0.86, 2.13) | 0.185 | 1.30 (0.80, 2.10) | 0.280 | 1.23 (0.75, 1.99) | 0.403 |
| Ethnicity | ||||||
| Non-Hispanic | ref | ref | ref | |||
| Hispanic | 1.07 (0.67, 1.71) | 0.769 | 1.32 (0.82, 2.12) | 0.251 | 1.29 (0.90, 1.84) | 0.166 |
| Education | ||||||
| ≤High school/technical school | ref | ref | ref | |||
| Some college | 0.80 (0.53, 1.21) | 0.284 | 1.41 (0.96, 2.09) | 0.082 | 1.38 (0.92, 2.07) | 0.121 |
| ≥College degree | 1.19 (0.81, 1.76) | 0.371 | 1.54 (1.03, 2.30) | 0.035 | 1.79 (1.15, 2.79) | 0.011 |
| Income perceptions | ||||||
| Living comfortably | ref | ref | ref | |||
| Getting by | 1.33 (1.05, 1.68) | 0.017 | 1.16 (0.79, 1.71) | 0.432 | 1.13 (0.78, 1.62) | 0.514 |
| Finding it difficult | 1.39 (0.95, 2.01) | 0.085 | 1.14 (0.66, 1.96) | 0.638 | 1.00 (0.63, 1.58) | 0.999 |
| Current smoking | ||||||
| Nonsmoker | ref | ref | ref | |||
| Smoker | 0.84 (0.52,1.34) | 0.449 | 0.78 (0.50, 1.23) | 0.286 | 0.89 (0.57,1.39) | 0.589 |
| Political affiliation | ||||||
| Liberal | ref | ref | ref | |||
| Moderate | 0.89 (0.62,1.29) | 0.534 | 0.74 (0.49,1.10) | 0.130 | 0.66 (0.46, 0.94) | 0.023 |
| Conservative | 0.86 (0.59, 1.24) | 0.403 | 0.73 (0.47,1.12) | 0.144 | 0.67 (0.46, 0.96) | 0.030 |
Note: Boldface Indicates p<0.05.
Current drinking Is defined as past 30-day drinking. Drinkers consumed ≤1 drink per day and did not consume ≥4 drinks on a single occasion for women and ≤2 drinks per day and did not consume ≥5 drinks on a single occasion for men. Heavier drinkers consumed >1 drink per day or consumed ≥4 drinks on a single occasion for women and >2 drinks per day or consumed ≥5 drinks on a single occasion for men. All models adjusted for survey return time stamped before or after the COVID-19 pandemic was declared on March 11, 2020 by the WHO.
The adjusted models also revealed that heavier drinkers had lower odds of support for banning outdoor alcohol advertisements (OR=0.41), warnings (OR=0.59), and guidelines (OR=0.60) than nondrinkers. Drinkers had lower odds of supporting a ban on outdoor alcohol advertising (OR=0.60) than nondrinkers. Drinkers and nondrinkers had similar odds of support for warnings and guidelines.
Unlike awareness of the alcohol–cancer link and drinking status, no covariates were associated with support for all the 3 policies. However, several associations between covariates and policy support were found. For instance, males had lower odds of supporting an outdoor advertising ban (OR=0.68) and drinking guidelines on beverage containers (OR=0.66) than females. In addition, Americans aged ≥60 years had greater odds (than those aged 18 –39 years) of supporting an outdoor alcohol advertising ban (OR=1.62). Those with a college degree or greater had higher odds of supporting the inclusion of warning labels (OR=1.54) and drinking guidelines (OR=1.79) on beverage containers than those with a high-school degree/technical degree or less. Those getting by had higher odds (OR=1.33) of supporting an outdoor advertising ban than those describing their income as living comfortably. Finally, moderate (OR=0.66) and conservative (OR=0.67) Americans had lower odds of support for adding drinking guidelines than liberal Americans. No associations were found between policy support and race, ethnicity, or smoking status.
DISCUSSION
Most Americans supported adding health warning labels (65.1%) and recommended drinking guidelines (63.9%) to alcoholic beverage containers. By contrast, only 34.4% of Americans supported banning outdoor alcohol advertising. The odds of support for all the 3 policies were lower among Americans believing that alcohol has no effect/reduces cancer risk than Americans aware of the alcohol–cancer link. Americans responding don’t know had lower odds of support for adding warning labels and drinking guidelines to alcohol containers than those aware of the alcohol–cancer link. Moreover, the odds of support were lower among heavier drinkers than among nondrinkers.
This is the first study to examine the relationship between alcohol control policy support and awareness of the alcohol–cancer link among a national U.S. sample. Research from Australia, England, Denmark, and Canada using different designs and measures has consistently found policy support to be associated with awareness of the alcohol–cancer link.18–21 Collectively, these findings suggest that increasing awareness of the alcohol–cancer link may increase alcohol control policy support, which may ultimately expedite policy adoption and implementation. Furthermore, >50% of Americans are unaware that alcohol affects cancer risk. Efforts are clearly needed to inform the public about this important modifiable cancer risk factor.
The finding that policy support was higher among nondrinkers is also consistent with previous research. Buykx and colleagues18 found that policy support was inversely related to alcohol consumption quantity. Similarly, Bates et al.19 found drinking status to be a significant predictor of support for alcohol price/availability, marketing/information, and harm reduction policies, with nondrinkers having greater support. These findings are unsurprising because alcohol control policies may trigger greater reactance among drinkers than among nondrinkers.33
Although >60% of Americans supported adding warning labels and drinking guidelines to beverage containers, about a third supported banning outdoor alcohol advertising. The use of the term banning instead of reducing outdoor alcohol advertising may evoke reactance owing to preferences for personal control,33 which may explain why fewer respondents endorsed such bans than those who supported warning labels and drinking guidelines (which are informational and do not restrict personal choice).
Only awareness of the alcohol–cancer link and current drinking status were consistently associated with policy support. These findings underscore the potential significance of awareness of the alcohol–cancer link and drinking status as factors that may influence alcohol control policy support. There are, of course, many cultural, social, political, and commercial factors that affect alcohol consumption in the U.S. Therefore, raising awareness of the alcohol–cancer link alone may not be sufficient for increasing policy support, but these data reveal a consistent association (experimental or longitudinal data are needed to assess causality).
Limitations
Limitations of this study include the reliance on self-reported drinking status, which may have resulted in misclassification owing to under-reporting of drinking.34 However, the authors have no reason to believe that it would affect the relationship between consumption and policy support. In addition, the alcohol measures assessed past 30–day drinking status and do not capture seasonal drinking behaviors.35 Second, the measures of awareness of the alcohol–cancer link were not conditional on the quantity of alcohol consumed. However, a previous national survey assessed how drinking too much alcohol affected cancer risk and found similarly low levels of awareness.12 Third, the policy support measures were worded, To reduce the problems associated with excessive alcohol use…, which could have encouraged higher endorsement owing to social desirability, although it seems unlikely that this would affect the associations with awareness and consumption. Fourth, the alcohol control policies available in the HINTS survey focused exclusively on communication-specific themes such as advertising and labeling.
CONCLUSIONS
Alcohol is a leading modifiable risk factor for cancer, yet most Americans are unaware that alcohol increases cancer risk. Misperceptions about the alcohol–cancer link were associated with lower odds of support for 3 alcohol control policies. Moreover, heavier drinkers had lower odds of policy support than nondrinkers. Because public opinion is one aspect of political will, which has been described as an essential component to the implementation of public health policies, increasing awareness and subsequently policy support may help increase the adoption of preventive alcohol policies.22–24 Increasing awareness of the alcohol–cancer link, such as through multimedia campaigns and patient–provider communication, may be an important new strategy for health advocates working to implement preventive alcohol policies.
Supplementary Material
ACKNOWLEDGMENTS
No financial disclosures were reported by the authors of this paper.
Footnotes
CREDIT AUTHOR STATEMENT
Andrew B. Seidenberg: Conceptualization; Formal analysis; Methodology; Writing - original draft; Writing - review & editing. Kara P. Wiseman: Conceptualization; Methodology; Writing - review & editing. Raimee H. Eck: Writing - review & editing. Kelly D. Blake: Writing - review & editing. Heather N. Platter: Writing review & editing. William M.P. Klein: Conceptualization; Methodology; Writing - review & editing.
SUPPLEMENTAL MATERIAL
Supplemental materials associated with this article can be found in the online version at https://doi.org/10.1016/j.amepre.2021.07.005.
REFERENCES
- 1.WHO. Global status report on alcohol and health 2018. Geneva, Switzerland: WHO, 2018. https://www.who.int/publications/i/item/9789241565639. Published September 27, Accessed October 5, 2021. [Google Scholar]
- 2.Bagnardi V, Rota M, Botteri E, et al. Alcohol consumption and site-specific cancer risk: a comprehensive dose–response meta-analysis. Br J Cancer. 2015;112(3):580–593. 10.1038/bjc.2014.579. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Stewart BW, Wild CP. World cancer report 2014. Lyon, France: International Agency for Research on Cancer; 2014. https://publications.iarc.fr/Non-Series-Publications/World-Cancer-Reports/World-Cancer-Report-2014. Published 2014. Accessed March 11, 2021. [Google Scholar]
- 4.World Cancer Research Fund/American Institute for Cancer Research. Continuous update project expert report 2018. Alcoholic drinks and the risk of cancer. London, United Kingdom: World Cancer Research Fund/American Institute for Cancer Research; 2021. https://www.wcrf.org/wp-content/uploads/2021/02/Alcoholic-Drinks.pdf. Published 2021. Accessed August 13, 2021. [Google Scholar]
- 5.Bagnardi V, Rota M, Botteri E, et al. Light alcohol drinking and cancer: a meta-analysis. Ann Oncol. 2013;24(2):301–308. 10.1093/annonc/mds337. [DOI] [PubMed] [Google Scholar]
- 6.Rehm J, Patra J, Popova S. Alcohol drinking cessation and its effect on esophageal and head and neck cancers: a pooled analysis. Int J Cancer. 2007;121(5):1132–1137. 10.1002/ijc.22798. [DOI] [PubMed] [Google Scholar]
- 7.Ahmad Kiadaliri A, Jarl J, Gavriilidis G, Gerdtham UG. Alcohol drinking cessation and the risk of laryngeal and pharyngeal cancers: a systematic review and meta-analysis. PLoS One. 2013;8(3):e58158. 10.1371/journal.pone.0058158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Marron M, Boffetta P, Zhang ZF, et al. Cessation of alcohol drinking, tobacco smoking and the reversal of head and neck cancer risk. Int J Epidemiol. 2010;39(1):182–196. 10.1093/ije/dyp291. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Alattas M, Ross CS, Henehan ER, Naimi TS. Alcohol policies and alcohol-attributable cancer mortality in U.S. states. Chem Biol Interact. 2020;315:108885. 10.1016/j.cbi.2019.108885. [DOI] [PubMed] [Google Scholar]
- 10.Goding Sauer A, Fedewa SA, Bandi P, et al. Proportion of cancer cases and deaths attributable to alcohol consumption by U.S. state, 2013–2016. Cancer Epidemiol. 2021;71(Pt A):101893. 10.1016/j.canep.2021.101893. [DOI] [PubMed] [Google Scholar]
- 11.Scheideler JK, Klein WMP. Awareness of the link between alcohol consumption and cancer across the world: a review. Cancer Epidemiol Biomarkers Prev. 2018;27(4):429–437. 10.1158/1055-9965.EPI-17-0645. [DOI] [PubMed] [Google Scholar]
- 12.Wiseman KP, Klein WMP. Evaluating correlates of awareness of the association between drinking too much alcohol and cancer risk in the United States. Cancer Epidemiol Biomarkers Prev. 2019;28(7):1195–1201. 10.1158/1055-9965.EPI-18-1010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.American Institute for Cancer Research. 2017 AICR cancer risk awareness survey report. Washington, DC: American Institute for Cancer Research; 2017. https://www.aicr.org/assets/docs/pdf/reports/AICR%20Cancer%20Awareness%20Report%202017_jan17%202017.pdf. Published 2017. Accessed February 12, 2021. [Google Scholar]
- 14.Breslow RA, Castle IP, Chen CM, Graubard BI. Trends in alcohol consumption among older Americans: National Health Interview Surveys, 1997 to 2014. Alcohol Clin Exp Res. 2017;41(5):976–986. 10.1111/acer.13365. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Grant BF, Chou SP, Saha TD, et al. Prevalence of 12-month alcohol use, high-risk drinking, and DSM-IV alcohol use disorder in the United States, 2001–2002 to 2012–2013: results from the National Epidemiologic Survey on Alcohol and Related Conditions. JAMA Psychiatry. 2017;74(9):911–923. 10.1001/jamapsychiatry.2017.2161. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Grucza RA, Sher KJ, Kerr WC, et al. Trends in adult alcohol use and binge drinking in the early 21st-century United States: a meta-analysis of 6 national survey series. Alcohol Clin Exp Res. 2018;42(10):1939–1950. 10.1111/acer.13859. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.LoConte NK, Brewster AM, Kaur JS, Merrill JK, Alberg AJ. Alcohol and cancer: a statement of the American Society of Clinical Oncology. J Clin Oncol. 2018;36(1):83–93. 10.1200/JCO.2017.76.1155. [DOI] [PubMed] [Google Scholar]
- 18.Buykx P, Gilligan C, Ward B, Kippen R, Chapman K. Public support for alcohol policies associated with knowledge of cancer risk. Int J Drug Policy. 2015;26(4):371–379. 10.1016/j.drugpo.2014.08.006. [DOI] [PubMed] [Google Scholar]
- 19.Bates S, Holmes J, Gavens L, 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):688. 10.1186/s12889-018-5581-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Weerasinghe A, Schoueri-Mychasiw N, Vallance K, 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(2):398. 10.3390/ijerph17020398. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Christensen ASP, Meyer MKH, Dalum P, Krarup AF. 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. 10.1016/j.ypmed.2019.05.010. [DOI] [PubMed] [Google Scholar]
- 22.Richmond JB, Kotelchuck M. Co-ordination and development of strategies and policy: the United States example. In: Holland WW, Detels R, Knox G, eds. Oxford Textbook of Public Health. New York, NY: Oxford University Press, 1991. [Google Scholar]
- 23.Asher H Polling and the Public: What Every Citizen Should Know. 9th ed. Washington, DC: CQ Press, 2004. [Google Scholar]
- 24.Kingdon JW. Agendas, Alternatives, and Public Policies. 2nd ed. Boston, MA: Addison-Wesley Educational Publishers Inc., 1995. [Google Scholar]
- 25.Diepeveen S, Ling T, Suhrcke M, Roland M, Marteau TM. Public acceptability of government intervention to change health-related behaviours: a systematic review and narrative synthesis. BMC Public Health. 2013;13(1):756. 10.1186/1471-2458-13-756. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Department of the Treasury Alcohol and Tobacco Tax and Trade Bureau. Petition for a report to congress supporting a label on alcoholic beverages warning the public that consumption can cause cancer. Washington, DC: Department of the Treasury Alcohol and Tobacco Tax and Trade Bureau; 2020. https://consumerfed.org/wp-content/uploads/2020/10/Citizen-petition-re-alcohol-cancer-warning.pdf. Published 2020. Accessed March 1, 2021. [Google Scholar]
- 27.National Cancer Institute. Health Information National Trends Survey 5 (HINTS 5) Cycle 4 methodology report. Bethesda, MD: National Cancer Institute; 2020. https://hints.cancer.gov/docs/methodologyreports/HINTS5_Cycle4_MethodologyReport.pdf. Published 2020. Accessed May 20, 2021. [Google Scholar]
- 28.Nelson DE, Kreps GL, Hesse BW, et al. The Health Information National Trends Survey (HINTS): development, design, and dissemination. J Health Commun. 2004;9(5):443–484. 10.1080/10810730490504233. [DOI] [PubMed] [Google Scholar]
- 29.Department of Agriculture US, HHS. Dietary Guidelines for Americans, 2020–2025. 9th Edition Washington, DC: U.S. Department of Agriculture, HHS; 2020. www.DietaryGuidelines.gov. Published 2020. Accessed August 13, 2021. [Google Scholar]
- 30.Gollust SE, Attanasio L, Dempsey A, Benson AM, Fowler EF. Political and news media factors shaping public awareness of the HPV vaccine. Womens Health Issues. 2013;23(3):e143–e151. 10.1016/j.whi.2013.02.001. [DOI] [PubMed] [Google Scholar]
- 31.Lau RR, Heldman C. Self-interest, symbolic attitudes, and support for public policy: a multilevel analysis. Polit Psychol. 2009;30(4):513–537. 10.1111/j.1467-9221.2009.00713.x. [DOI] [Google Scholar]
- 32.Rao JN, Scott AJ. On chi-squared tests for multiway contingency tables with cell proportions estimated from survey data. Ann Statist. 1984;12(1):46–60. 10.1214/aos/1176346391. [DOI] [Google Scholar]
- 33.Bensley LS, Wu R. The role of psychological reactance in drinking following alcohol prevention messages1. J Appl Soc Psychol. 1991;21(13):1111–1124. 10.1111/j.1559-1816.1991.tb00461.x. [DOI] [Google Scholar]
- 34.Stockwell T, Zhao J, Greenfield T, Li J, Livingston M, Meng Y. Estimating under- and over-reporting of drinking in national surveys of alcohol consumption: identification of consistent biases across four English-speaking countries. Addiction. 2016;111(7):1203–1213. 10.1111/add.13373. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Uitenbroek DG. Seasonal variation in alcohol use. J Stud Alcohol. 1996;57(1):47–52. 10.15288/jsa.1996.57.47. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.