Alcohol and breast cancer risk

Globally, alcohol is identified as causing significant disability, disease and mortality [1] and is considered a risk factor for a range of soft tissue cancers [2, 3] including breast cancer [4, 5]. International research suggests that consumption at low levels previously considered ‘safe’ significantly elevate cancer risk [4–7]. Cancer Council Australia recommends that people limit their alcohol, or avoid alcohol all together [8]. However, the regular consumption of alcohol is widespread in Australia—roughly 17% of Australians use alcohol at levels that put them at risk of harm over their lifetime [9]. These statistics are similar to high risk consumption rates seen in other developed countries, reinforcing the global nature of this public health problem. For example, in 2016 19.0% of Canadians aged 12+ reported alcohol consumption that classified them as heavy drinkers [10]. Similarly, data from 2015 suggests that 26.9% of Americans aged 18+ reported that they engaged in binge drinking (~4 drinks for women and 5 drinks for men in 2 hours) [11].

The consumption of alcohol—how, what, when, where and in what ways it is consumed—is shaped by a complex system of social, psychological, behavioural, economic, legal and physical environmental factors that interoperate at individual, community and population levels [12]. Although there has been some success in reducing rates of consumption in Australia (a decrease from 20% in 2010 to 17% 2014), reducing consumption levels to meet national guidelines remains a challenge [13] as alcohol consumption in Australia has been described as ‘culturally ingrained’ [14]. Notably, alcohol consumption levels among ‘middle-aged’ (45–64 years) Australian women remain higher than among other age groups [15–17]. In this same group we have also witnessed the trebling of cancer incidence over the past 50 years [6]. Thus, alcohol remains a notable public health problem [18] with the attributable risk of alcohol for all cancer being ~6% [8].

In order to minimise breast cancer risk, it is critical that women understand the risks associated with alcohol so they can make informed decisions regarding consumption [19, 20]. According to an Australian Institute of Health and Welfare (2014) report, the awareness of breast cancer is high in Australia, relative to other cancers [21]. However, public knowledge regarding the association between alcohol consumption specifically and cancer risk remains poor [13, 22, 23]. Just under half (47%) of Australians identify drinking alcohol as a risk factor for cancer [24]. Furthermore, knowledge about alcohol as a risk factor for cancer may be poorest among the heaviest drinkers in Australia [24]. This is concerning, given new evidence suggesting a dose-response-relationship between alcohol and breast cancer. That is, the greater the quantity of alcohol consumed, the greater the risk of breast cancer. The excess risk for cancer among women who drink heavily is estimated to be 60% compared to non-drinkers [25]. Clear and meaningful communication of the potential role of alcohol in cancer causation, and the promotion of behaviour change strategies for modifiable risk factors, are central to prevention [26] and to decreasing cancer incidence [6].

Current interventions to reduce alcohol consumption in Australia

To date, extensive research in Australia and abroad has looked at interventions to reduce harmful drinking. In particular, much of the effort to minimise risky alcohol consumption has focused on developing and implementing preventive interventions in primary health care settings [27–30]. Despite considerable evidence for the efficacy and cost-effectiveness of such primary health care-based approaches [31–36], widespread incorporation of alcohol prevention in routine care has not occurred in Australia or elsewhere [31, 36]. Furthermore, widespread incorporation of interventions in relation to alcohol-related cancer prevention specifically is unlikely given the evidence that only 37% of general practitioners attending health prevention seminars around Australia agreed that the regular intake of alcohol, even at light levels, can lead to cancer [37]. Specific to breast cancer risk, much of the intervention literature focuses on diet (not specific to alcohol), physical inactivity [38–40], and improving mammography screening uptake [41].

Study rationale and objective

Existing literature suggests that there is a need for greater training for primary care providers regarding alcohol and cancer risk, communication strategies for use in primary care settings, and the development of interventions outside of the primary care setting [42]. However, to date the research necessary to inform population- and policy-level interventions is limited. In response, the present study sought to identify women’s perceptions of the role of alcohol in breast cancer risk and investigate the complexity of their logic with regards to alcohol consumption and reduction. Critical reflection of women’s logic around consumption provides foundational information to begin to establish participant-driven population- and policy-level interventions that may be used to addresses alcohol consumption, and either directly or indirectly reduce breast cancer risk. Our work also furthers conversations regarding the complexity of interventions required to reduce alcohol consumption in this population.

Existing knowledge about breast cancer risk

Recommendations for how health organisations and government can decrease consumption

What can be done at the policy/legislative level?

When asked about what the government or health organisations should do to reduce consumption, women also referred to what we identify as policy or legislative changes. For example, labelling alcohol products was provided as a strategy for reducing consumption.

“I know this because of being in the industry but the education on the bottle saying how many standard drinks are in there. I know depending on the alcohol volume the standard drinks change so therefore I look at it and go ‘oh, okay, this one’s a bit stronger so I probably shouldn’t have quite so much’.”

(Paulette)

The existing legislation regarding alcohol consumption and driving was often mentioned as a factor that influenced consumption patterns. Driving was identified as a socially acceptable rationale for not drinking, and gave women an ‘out’, avoiding peer pressure.

“…like Saturday night when I’m going out with friends I won’t have two drinks because I’ll have to drive home. I think that’s a lot more acceptable these days too. People are a lot more wary of not being over the limit when you’re driving and I think that’s education as well.”

(Nicole)

“No. If I’m driving I wouldn’t drink anyway because I know that for me a lot less than the legal limit will affect my reactions a lot so I’m not confident with any kind of alcohol if I’m doing something like that.”

(Lia)

Taxation was identified by some participants as a proposed strategy for decreasing consumption.

“For me, for it to be effective you’d just crank up the price. I think it was earlier this week somebody was suggesting putting a higher tax on sugar drinks, like soft drinks and stuff, and apparently Malcolm Turnbull’s said ‘no, we pay enough taxes anyway’. I think that would be an incentive—it would for me—but, you know, unfortunately I think the people that can least afford it are the ones that would still pay for it on the whole, across the population.”

(Nicole)

However, others felt that taxation as an approach to reduction would be ineffective. Explanations for the failure of such an approach included people continuing to find a way to purchase alcohol because it is something they enjoy. The pleasure, both physically and socially, of drinking was expressed as follows:

“…even if there’s a government incentive or government tax about that I’m still going to probably be indulging because I enjoy it or I like it or I like the taste of it or it’s just a habit, or all of them.”

(Toni)

“Perhaps like taxing—no, that’s not going to work because it doesn’t work for people that smoke so it’s not going to work for people that really like to drink.”

(Angela)

Cynthia suggested that strategies targeting different sub-populations are required. She indicated that her income and her education make her unique, and that certain strategies might work for some population groups, but not others.

“…You know, giving people information is not going to change their behaviour. I’m well educated. I have all the information. I know the risks…. We’re middle class people; we know stuff….You’ve got all that sort of methodology over here and you’ve got your behaviours and your information, so making, you know, settings, putting tax on alcohol, you know, that wouldn’t affect me because I have the means to overcome the tax whereas tobacco, you know, now an average cost of …a packet [of cigarettes] has gone up to $35, a huge incentive to stop smoking, whereas the way we structure our taxes on alcohol, that isn’t going to affect somebody like me. It might affect a pensioner. It might affect a woman who’s stopped working but it’s not going to affect me and it’s not going to affect most of my friends.”

(Cynthia)

Josephine suggested that there have been changes over time in terms of the availability of alcohol that may play a role in increased consumption.

“…well, yeah, so you drive past them all the time whereas years ago you had to go out of your way to get it.”

(Josephine)

She noted the number of locations where you can purchase alcohol and how this is likely an influence on consumption. She suggested that accessibility is an issue the government might address to reduce consumption.

Decreasing consumption is the responsibility of the individual

Alcohol consumption was identified as a personal issue by some participants, leading them to conclude that reducing consumption is not the responsibility of health organisations or government. Behaviour change for these women related to willpower, resilience, initiative and taking control. This was said to require self-reflection to first identify the problem and then take action.

“Well, you know, I was—I just thought ‘this is not sensible behaviour. This is unhealthy behaviour. You don’t need to be doing this. There are probably some risks here for your health’”

(Cynthia)

Despite seeing it as an individual responsibility, the strategies used by women, discussed below, may be used to inform messaging targeting behaviour change at a population level. Women talked about methods for cutting back or cutting out alcohol all together, and the strategies used to do so. Self-monitoring, rule setting, writing down their consumption, or using tools to measure how much they are drinking were methods identified as helpful in controlling alcohol consumption. For example:

“That’s one way and the three alcohol free days just to say ‘no, not going to drink on these days; that’s it’.”

(Sophie)

“… I started marking on the calendar. I had a [name of] changing my drinking behaviour and I decided that I had to have more non-drinking days in a month than drinking days so I had to have at least 15 non-drinking days and that should bring me down. I tried to think about not having, you know, more than this and more than that, but what I in the end did was count drinks over a week …”

(Cynthia)

“I would like to buy a set of glasses that have got the line because it would be more of a mental trigger to go ‘okay, no, that’—you know, from time to time I think about exercising willpower, measuring out 125ml but if the line’s not there, well, you know, you can…”

(Carrie)

Changing consumption patterns was another method identified for reducing alcohol intake—for example, purchasing smaller quantities, switching to lower alcohol content beverages and removing stocks of alcohol in the home.

Alcohol and breast cancer: Changing risk perception

The level of awareness of alcohol as a risk factor for breast cancer was low in our group of participants. Participants initially questioned the association based on their knowledge of people who drink and do not have cancer, or who do not drink and do have cancer. The confusion related to alcohol as a risk factor for cancer, but not always causing cancer. These data are in line with research identifying that individuals engage with multiple and often conflicting frameworks when making risk-based decisions [26]. Individuals’ perceptions of risk, or whether they or others can be regarded as a ‘disease candidate’ [48, 49] are reliant on the input of personal, social and cultural sources as well as expert-endorsed or celebrity-endorsed information. Our participants drew on their personal experiences to build an evidence-base that was seen as incongruent or conflicting with the information we provided regarding alcohol as a risk factor. While this demonstrates the challenges that might be faced in communicating the risks of alcohol at a population level, our data also suggest that the association might not be such a hard sell. The women reported widespread knowledge regarding alcohol having a negative impact on health more generally. As such, it ‘made sense’ or was seen as ‘logical’ that it would be associated with breast cancer. These data suggest that changing risk perception and the notion of what a breast cancer ‘candidate’ looks like, may be obtainable.

Reducing consumption requires more than education and awareness

The lack of awareness of alcohol as a risk factor for breast cancer may suggest that the first step to reducing consumption in this group of women is education and awareness. Inconsistent with previous Australian research [50], our participants supported the notion that providing women with more information about the association might help reduce consumption. However, while effective information about breast cancer risk has been argued to help women to make more accurate calculations of their personal risk [19], data are not convincing regarding the role of messaging in reducing consumption. Australian campaigns targeting alcohol behaviour via mass media have had limited success; likely a consequence of the overshadowing of persistent product marketing, social norms, and behaviours driven by addiction or habit [51]. Indeed, Dixon, Pratt (13) argue that while campaigns can raise awareness regarding the links between alcohol and cancer, as well as knowledge of drinking guidelines, “a single campaign may be insufficient to measurably curb drinking behaviour in a culture where pro-alcohol social norms and product marketing are pervasive” (p. 1).

Participants spoke of the success of smoking and skin cancer campaigns, and suggested that the government and health organisations should consider following similar lines of messaging to reduce alcohol consumption. However, the success of campaigns in Australia targeting smoking and skin cancer required extensive investment—both time and monetary. Alcohol is significantly ‘ingrained’ [14] in the Australian culture, and so it is likely to take decades before we see similar effects to what have been witnessed with campaigns on sun safety and smoking reduction.

The importance of taking into account social norms and habits

Exploring women’s logic regarding consumption provided us with insight regarding the social norms and habits that shape their behaviours. Their social norms and habits were particularly salient when women spoke of how and why they have made changes in their lives to reduce their alcohol consumption. Women spoke of choosing lower alcohol content beverages, switching to non-alcoholic beverages, choosing to be the designated driver, and self-monitoring their intake until lesser quantities became habit. Their rationale for changing habits provides a window into the values of these women, and what they may prioritise over drinking, providing opportunity for their voices to be acknowledged in campaign messaging. They identified the negative effects of alcohol in their lives; namely the impact on their weight, overall mood, social relationships, sleep, and productivity. It is possible that providing targeted messaging regarding how a reduction in alcohol consumption may improve their social, physical and mental health will be a way to indirectly lower breast cancer risk. However, we acknowledge that providing messages of this nature, while appealing to the cognitive evaluative processes, do not address the affective processes that shape risk perception [52]. As noted, individuals engage and consult with multiple, often conflicting frameworks when making risk-based decisions [26]. The data do however suggest that perhaps education about the health risks of alcohol will not be as persuasive as an emphasis on the more salient and immediate effects of alcohol.

The role of industry in alcohol consumption

In addition to the social norms and habits shaping consumption, we acknowledge that the alcohol industry (product marketing) and competing messages [53] likely play a larger role in shaping consumption than is identified in our data. Research demonstrates that Australian adults are regularly exposed to advertisements that depict alcohol consumption as fun, social and inexpensive, reinforcing existing cultural norms that prevent them from meeting current consumption guidelines [54]. In fact, the alcohol industry have been identified as actively working against health messages and purposefully working to embed alcohol into Australian culture [55]. Therefore, to achieve reductions in alcohol consumption, it has been recommended that continual education regarding the harms of alcohol needs to occur within a range of national, state and local policy controls. These controls may include managing the physical availability (access) and economic availability (price) of alcohol, and addressing the cultural place of alcohol in Australia through targeted and research-based social marketing and public education platforms, which have been recommended elsewhere [13, 42, 56]. However, Australia is a climate arguably weighted towards the promotion of alcohol, playing a large force in what can and will be done at a policy-level to reduce alcohol consumption. Industry, as a form of revenue, play a large role in what happens at a policy-level. Indeed Bates, Holmes (22) argue that changes to policy controls need to be considered within the context of the revenue generated by the alcohol industry in Australia, as well as citizen concerns for individual freedoms.

While our participants did not refer to the role of industry or government in the promotion of alcohol, nor the impact this may have on their alcohol consumption, there was an overall sense of support for government interventions targeting alcohol consumption (e.g. labelling and taxation). Public support for alcohol-related policies has been found to be associated with the extent to which policy are deemed intrusive or restrictive, with people tending to prefer policies they perceive to impact other people and not themselves [22]. The women in our study, for the most part, were supportive of reducing their alcohol intake by their own accord, suggesting that such policies would not have a significant impact on their lives. This likely explains their support for government interventions.

Risk perceptions surrounding breast cancer

Our data offer novel insight regarding the role that risk perception, as related to a breast cancer diagnosis, may play in women’s alcohol consumption. Within our discussions, women spoke of breast cancer as a ‘survivable cancer’. This may influence the extent to which they regard messaging about the reduction or elimination of behaviours they enjoy, such as consuming alcohol. In a similar vein, some women identified breast cancer as being extremely prevalent—the notion that ‘everyone gets breast cancer’. We question whether efforts to reduce risk may be impacted by this perception, with women assuming that behaviour has little to do with whether they will be diagnosed. The notion of optimistic bias, or overly optimistic judgements about breast cancer risk [57] may be helpful in explaining some women’s engagement in alcohol consumption despite known risks. Empirical research suggests that women often perceive their personal risk of breast cancer to be exceedingly less than that of other women. The low perception of risk has been found to have negative implications for engagement in preventative measures [58], and therefore may influence willingness to make lifestyle changes, including reducing alcohol consumption.

Strengths and limitations

This research provides an in-depth understanding of how women in a high-risk age group for breast cancer view the barriers/facilitators to reducing alcohol consumption as a modifiable risk factor for breast cancer, as well as factors supporting/inhibiting messages about breast cancer risk. The semi-structured interview method enabled the direction of inquiry to be guided by the participant as well as the key areas of interest. This participant perspective will support policy makers and social scientists in understanding more about the processes of risk communication. The research was undertaken by a multidisciplinary team comprising epidemiology, sociology, oncology, anthropology and occupational therapy. This supported trustworthiness during analysis and interpretation of the data as team members were able to interrogate the findings from a variety of professional dispositions. Further, the framework analysis method [46] provided a systematic approach with which to explore information central to our research objective.

Another strength of this study was the diversity of women we interviewed according to self-perceived alcohol consumption, breast cancer risk and socioeconomic status. A limitation in this respect, however, was that we sampled by self-reported diversity. This was undertaken in this way to help examine how women’s perceptions of breast cancer risk and alcohol consumption levels influenced their attitudes about alcohol consumption. This strategy, however, means we do not have an empirical picture of the differences in socio-demographic, risk status for breast cancer and level of alcohol consumption. Similarly, the self-selecting nature of our sample may have been a limitation to the research. Because the research was undertaken only in South Australia the results may not be generalisable to other contexts. It is also important to acknowledge that our data analysis does not take into account diversity across middle-aged women, and the extent to which individual characteristics and experiences will shape intervention effectiveness, or lack thereof. Further, we do not explore the role of gender in shaping consumption patterns, and how findings may differ in men of this age group. However, our data and critical reflection allow us to propose considerations in the development of risk communication that can be tested through other research methodologies.