This is, as far as we are aware, the first study on health information about cardiovascular health and alcohol use outside of a US context. Our findings show that about a third of patients in cardiology services – particularly those with a higher level of education – described being exposed to information sources that reported healthy heart effects with moderate alcohol consumption. Most participants reported accessing health information about alcohol from media sources, rather than from health professionals. There was a strong tendency for health information sources to convey mixed or conflicting messages about the cardiovascular effects of alcohol – irrespective of information source. Exposure to messages that moderate drinking has healthy heart effects was associated with increased odds of hazardous alcohol use.
We found that many patients with cardiovascular disease report being exposed to suggestions of a healthy heart effect, an observation consistent with those of a US-based survey (7). Approximately half of participants in our study had heard that alcohol can affect the heart, of whom approximately two thirds indicated being exposed to reports of a healthy heart effect. Our finding that older patients were more likely to have heard that moderate alcohol consumption can affect the heart than younger patients is in line with results from the Health eHeart Study, a US-based survey of 5,582 people with CVDs (16). This may suggest either that such reports tend to target older people, that older people are more perceptive to this type of message, or that information about alcohol and the heart has been less widely reported in recent years.
Overall, our finding that less than half of participants had heard that moderate alcohol consumption can affect the heart is consistent with limited public awareness of the wider risks of alcohol use. It has long been established, for example, that alcohol consumption increases the risk of cancer, yet public awareness of this link is low, particularly among men and those with lower education (22, 23). In our study, there was no association between gender or educational level and awareness that alcohol can affect the heart. However, among participants who had heard of cardiovascular effects, those with lower education were less likely to have heard that alcohol is good for the heart, which corresponds to the findings in the Health eHeart Study (16). In our study, we did not ask participants whether they had heard that specific types of alcoholic beverages can affect the heart – rather we asked about alcohol consumption in general. Yet previous findings indicate that wine is more frequently perceived as heart healthy than spirits or beer (24). Given that that regular wine consumption may be associated with higher socioeconomic position (25), it is possible that beverage preferences may have contributed to the differences we observed in exposure to reports of healthy heart effects according to educational level. Indeed, Whitman et al reported that participants believing alcohol to be heart healthy had higher weekly wine consumption than those who did not believe alcohol to be good for the heart, as well as higher overall weekly alcohol consumption (16). We further note that knowledge of alcohol’s cardiovascular effects may also be linked to wider awareness of the risks of alcohol. In a recent study, those aware that alcohol use increases the risk of heart disease were more likely to also be aware of the link between alcohol and cancer (24), suggesting that risk perceptions surrounding alcohol use and CVDs may generalize to other negative health consequences.
We also found that many patients with cardiovascular disease access health information about alcohol from lay media sources. About a quarter of reports described as suggesting a healthy heart effect were from television, while a third were in newspapers. Again, these findings are consistent with those of Medling et al (7). In their study, 61% of participants had heard of healthy heart effects via television and 21% from newspapers. Further evidence regarding the importance of media sources in conveying public health information about alcohol can be found in the Alcohol Toolkit study – a survey of the general UK population (21). The Alcohol Toolkit study assessed respondents’ awareness and knowledge of national alcohol guidelines. On average, across seven repeated surveys, television and radio were the most commonly accessed sources of information about the alcohol guidelines, followed by newspapers and magazines.
In similarity to the findings of Medling et al (7), only one in eight participant in our study accessed information about alcohol directly from health professionals. Given the global consensus that any amount of alcohol consumption is unhealthy (26), it is troubling that health professionals were perceived to convey mixed messages about the cardiovascular effects of alcohol. While we did not assess the extent to which participants believed in a healthy heart effect, trust of clinical staff with regards to conversations around alcohol is reported to be high in Sweden (27). Participants may therefore have considered information from clinicians to be reliable. Furthermore, our finding that information from clinicians was ambiguous regarding the health effects of alcohol suggests that opportunities for alcohol prevention – such as brief interventions (28) – within cardiology care are currently being missed. A recent Swedish study suggests that cardiology clinicians’ knowledge of key concepts in alcohol prevention, such as the definition of hazardous alcohol use, remains limited, which could contribute to staff communicating mixed messages to patients (29).
Our observation that exposure to suggested healthy heart effects was associated with increased odds of hazardous drinking is in keeping with findings from the US (16). However, studies on awareness of the link between alcohol and cancer in the general population have failed to demonstrate similar associations (15, 24). While it is not possible to directly compare these findings due to differences in the definitions and measures used for hazardous alcohol use and study populations, existing studies suggest that the association we observed with hazardous alcohol use may be specific to CVD. We suggest future studies examine this further.
The theory of lay epidemiology proposes that media and informal information sources are influential in shaping individuals’ health beliefs (14, 15). Our findings suggest that reports of healthy heart effects are encountered in social contexts with friends and family, although somewhat less frequently than via television reports and newspaper articles. Lay epidemiology theory suggests that these conversations may alter how individuals understand and respond to the perceived risks and benefits associated with regular, low-level alcohol consumption. The HBM further proposes that any changes to individuals’ beliefs and expectations about the health effects of alcohol may subsequently lead to changes in behaviour (12). While our study does not provide evidence of causality, lay epidemiology and the HBM offer possible theoretical explanations for our finding of increased odds of hazardous alcohol use among participants exposed to reports of healthy heart effects.
While controversy, or a lack of scientific consensus, may have contributed to mixed messages in earlier reports about alcohol’s cardiovascular effects, the influence of vested commercial interests on ambiguous reporting should also be considered (30). Globally, the activity of the alcoholic beverage industry in promoting mixed messages about the health effects of alcohol is emphasized in the WHO’s 2023 guide to journalists for reporting on alcohol (26), with commercial activities increasingly recognized as key social determinants of health (31, 32). Our findings relate specifically to the Swedish context, in which alcohol trade is subject to state regulation; Swedish legislation prohibits the marketing of any alcoholic beverages on public television and radio and a government monopoly is responsible for the retail sale of alcoholic beverages (33). It is thus possible that the impact of private sector activities on how alcohol’s health effects are reported may be greater in countries with less restrictive alcohol policies (34–36).
Finally, our finding that participants perceived mixed messages in reports of alcohol’s health effects may have implications for policy implementation. While our findings are from Sweden, similar results have been reported elsewhere (7, 16, 24). Evidence suggests that public support for evidence-based alcohol harm-reduction policies is associated with knowledge of health risks, particularly the link between alcohol and cancer (37). Ambiguous reporting of the health effects of alcohol use may thus present a barrier to effective public health policy. Overall, evidence supports a comprehensive approach to reducing population-level alcohol harm, including interventions to increase public awareness around the risks of alcohol use, such as mass media campaigns (38, 39), health warning labels on alcoholic beverages (40), regulating alcohol marketing (41), and providing brief interventions during routine healthcare interactions (42).
Strengths and limitations
Study strengths included recruitment of consecutive patients where possible and an acceptable response rate and sample size, permitting analysis of sociodemographic covariates such as educational level. While our sample may not be nationally representative, an additional strength was the collection of data from three heterogenous regions in Sweden. We identified the following limitations: This study was cross-sectional in design and is thus unable to determine the direction of the associations observed, or to establish causality. We did not ask participants when information sources had been accessed – it is therefore possible that many of the healthcare interactions and media sources reported were historical. Nor did we assess whether the participants stated that they believed in the messages around heart health. It is also possible that survey responses regarding the study exposure were subject to recall bias. Additionally, despite countermeasures such as explaining that survey completion was anonymous, we acknowledge that self-reported alcohol screening methods tend to underestimate hazardous alcohol use as a result of social desirability bias (43) and other factors.