A poster displayed on buses, seen on 7 and 8 September 2017, in Bristol for Viva!, a veganism advocacy charity, featured an image of a cow’s udder and included the claims “Some dairy industry facts we bet you don’t know … Most cows are pregnant when milking. That’s why milk contains 35 hormones, including oestrogen … some of these are linked to cancer. Milk is for babies, so let Viva! wean you off the teat!”.
Two complainants, who believed the claim “milk contains 35 hormones, including oestrogen … some of these are linked to cancer” implied that drinking cow’s milk could cause cancer, challenged whether the implied claim was misleading and could be substantiated.
Viva! said the claim referred to hormones that were naturally occurring in cow’s milk, which increased during pregnancy and were essential for calf development. The claim did not refer to hormone treatments or artificial hormones.
They highlighted that the ad stated the hormones were “linked to” cancer, rather than that they “caused” cancer. They said “linked to” was wording commonly used to express an association between two factors when there was a potential or likely relationship but not an absolute causative relationship, and consumers were used to seeing the phrase “linked to” as opposed to “causes” in that context. They provided examples of health organisations, including the NHS and World Cancer Research Fund (WCRF) and a science news website referring to “links” between consumption of certain types of substances or foods and certain cancers. Viva! said they also believed it was appropriate to use the phrasing “linked to” because dairy consumption and the hormones present in milk had been implicated in increasing the risk of some cancers.
In support of the claim they referred to a range of papers, which they believed demonstrated the presence of over 35 hormones in cow’s milk, including oestrogen hormones and insulin-like growth factor (IGF-1), and a further seven papers which they believed provided sufficient evidence to support the claim that some of the hormones in cow’s milk were linked to cancer. They said that in particular IGF-1 and the oestrogen hormone oestradiol had been shown to increase the risk of cancer. They referred to two of the papers which they had provided and a table from the website of a breast cancer charity which listed studies relating to the risk of breast cancer in women with higher oestradiol levels compared to those with lower levels, and summarised their findings.
They said that all human studies that focused on diet were observational and that as a result there would always be confounding factors. They argued that in epidemiological studies the best that could be done was to control for as many confounding factors as possible. Viva! said that such studies usually called for more research, which was basic good scientific practice.
While the claim stated that some hormones in cow’s milk were “linked” to cancer rather than definitively stating that they caused cancer, the ASA considered that consumers would nonetheless interpret it to mean that because of the hormones that were present in cow’s milk, drinking cow’s milk could increase a person’s risk of developing cancer. The claim did not itself make clear whether it related to naturally occurring hormones and/or to artificial hormones, but we considered that consumers were likely to understand from the context of the ad as a whole, which focused on calves and pregnant cows, that the claim related to the naturally occurring hormones that were present in cow’s milk. We considered that such a claim must be substantiated by robust evidence that demonstrated a link between the naturally occurring hormones in cow’s milk with incidences of cancer in human populations.
Three of the papers provided by Viva! were discussions of how hormones and other substances present in milk (or dairy products more generally) were thought to impact on physiological processes which might have the effect of increasing the risk of cancer and other diseases, although no definitive evidence was cited that demonstrated that the hormones present in cow’s milk increased the risk of cancer. Two of the papers referred specifically to studies relating to the consumption of cow’s milk and cancer risk, but both reported that there was conflicting evidence in that area, and in any case none of the three papers were systematic reviews or meta-analyses of relevant studies and therefore did not subject the few studies to which they referred to rigorous analysis. We had not seen the studies referred to in the table on the breast cancer charity’s website and we were therefore unable to assess them. Furthermore, we understood those studies examined links between higher and lower levels of the hormone oestradiol in women and breast cancer risk, rather than relating to links between drinking cow’s milk and cancer.
The remaining four papers consisted of two prospective cohort studies, a retrospective study and a meta-analysis.
The first cohort study, based in Sweden, identified 22,788 individuals with lactose intolerance (who therefore were assumed to have low dairy intake) and their risks of lung, breast and ovarian cancer. The study found their risks of developing those cancers were significantly decreased compared to the general population. The authors noted that they were able to take account of some potential confounding factors because they also found that the family members of the lactose intolerant individuals had similar incidences of those cancers as the general population, thereby suggesting that it was low dairy consumption that reduced the risk of cancer. The authors concluded that the decreased risks of lung, breast and ovarian cancers in lactose intolerant individuals may be related to their specific dietary patterns but that the contributions from other potential confounding factors for which they were unable to take account should not be neglected.
The second cohort study of 21,660 male physicians in 1982 in the USA aged between 40 and 84 examined the association between intake of dairy products and the incidence and survival of prostate cancer. Baseline information on dairy intake related to consumption of whole milk, skim/low-fat milk, cold breakfast cereal, hard cheese and ice-cream, sorted into categories based on the number and regularity of servings consumed. The study found that higher consumption of skim/low-fat milk was associated with higher risk of non-aggressive prostate cancer, and higher consumption of whole milk was associated with increased risk of fatal prostate cancer. The authors identified some limitations to the study and noted that their findings needed to be confirmed by cohorts with more detailed dietary information.
The retrospective study used cancer incidence and mortality data between 1993 and 1997 collated by the World Health Organisation’s International Agency for Research on Cancer and the International Association of Cancer Registries, and food consumption data between 1961 and 1997 from the Food and Agriculture Organisation of the United Nations. It analysed data from 40 countries to explore the correlation between consumption of milk (and other foods) with cancer incidence and mortality in each country. The paper reported that milk consumption was closely correlated with breast cancer incidence, ovarian cancer and corpus uteri cancer. The authors noted shortcomings with the study were that: rates of cancer incidence and mortality were affected by regional differences in diagnosis, registration and certification, and fatality rates; data from developing countries was less complete than from developed countries because of problems including under-diagnosis and under-certification; and that consumption data was not based on recorded dietary practice but was roughly estimated based on data relating to production, imports, stock changes, food exported or used for purposes other than consumption by humans, and population.
The meta-analysis included studies that examined associations between dairy consumption and incidence of non-Hodgkin lymphoma, with a sub-group analysis relating specifically to milk consumption. It included 16 studies relating to milk consumption (14 of which were case-control studies, and two of which were cohort studies), involving 7,109 cases of non-Hodgkin lymphoma, and found that those with the highest levels of milk consumption versus those with the lowest milk consumption had a higher relative risk of developing the disease. The authors highlighted that the results were primarily based on case-control studies and that there was therefore a need for more cohort studies focusing on specific types of dairy consumption to confirm their conclusions.
We considered that the various sources provided by Viva! constituted adequate evidence that over 35 hormones were present in cow’s milk. With regard to whether the papers were sufficient evidence for the claim that some of those hormones were linked to cancer, the two cohort studies and the meta-analysis were of good methodology, were based on large groups of subjects, and had statistically significant findings. However, we were concerned that the cohort studies were unable to account for confounding factors which could affect the results, that the dietary information collected in the cohort study in the USA was self-reported and not validated, and that the meta-analysis relied primarily on the results of case-control studies. We also noted that the papers referred to other conflicting evidence and all noted the need for additional studies to confirm their findings. Further, the findings of one of the cohort studies related to dairy consumption as a whole, and it was also not clear that those which related to milk related specifically to cow’s milk. The retrospective study paper identified significant limitations with utilising the chosen data sets for such a study, and its methodology meant that it was not able to take account of confounding factors which could have affected the results. We also noted that it was not peer-reviewed and not published in a mainstream scientific journal. The studies and meta-analysis did not support Viva’s assertion that the findings of increased risk of cancer were specifically a result of the hormones present in cow’s milk rather than to other factors.
We therefore concluded the claim “milk contains 35 hormones, including oestrogen … some of these are linked to cancer”, as it would be understood by consumers to mean that due to the presence of hormones, drinking cow’s milk could increase a person’s risk of developing cancer, had not been substantiated and was therefore misleading.
The ad breached CAP Code (Edition 12) rules 3.1 3.1 Marketing communications must not materially mislead or be likely to do so. (Misleading advertising) and 3.7 3.7 Before distributing or submitting a marketing communication for publication, marketers must hold documentary evidence to prove claims that consumers are likely to regard as objective and that are capable of objective substantiation. The ASA may regard claims as misleading in the absence of adequate substantiation. (Substantiation).
The ad must not appear again in the form complained about. We told Viva! not to make claims which stated or implied that due to the presence of hormones, drinking cow’s milk could increase a person’s risk of developing cancer.