ASA Adjudication on Department of Health

Ad

A TV ad for a National Health Service (NHS) alcohol awareness campaign featured two women sharing a bottle of wine. One woman was shown as semi-translucent so her internal organs and skeleton were visible. The voice-over stated “If you’re a woman drinking two large glasses of wine or more a day, you could be putting your health at risk … you’re three times more likely to get mouth cancer.” On-screen text stated “Women drinking 40g+ of alcohol/day. Source: Department of Health analysis of ‘Corrao, 1999’”.

Issue

Two complainants challenged whether the claim "youre three times more likely to get mouth cancer" could be substantiated.

BCAP TV Code

Response

The Department of Health (DH) explained that mouth cancer was a term that included cancer of the oral cavity and pharynx. They sent us the Corrao meta-analysis on the risk of alcohol consumption referenced in the ad and argued that it showed that alcohol increased the risk of developing mouth cancer. DH said that a meta-analysis combined and analysed the results from a large number of different studies and was a highly regarded type of study as it could give an accurate picture of a large body of evidence.

DH explained that the Corrao analysis considered 16 studies relating to the oral cavity and pharynx that showed that drinking 50 g of alcohol per day increased the risk of mouth cancer by three times. They said they also confirmed this figure with Cancer Research. They said the ad referenced two large glasses of wine per day which equated to roughly 48 g or 50 g of alcohol. DH argued that the risks would be higher for women drinking in excess of two large glasses of wine per day and believed the ad made a conservative estimate of the overall risk from drinking. They said that the ad gave one reference for all three conditions mentioned in the ad combined rather than specifying each one separately. DH therefore referred to datasets that included relative risk from 40 g or more a day, but the specific relative risk for mouth cancer was over three times for 50 g per day. They explained that they were clear to state "40g+ of alcohol/ day" in the ad to reflect the fact that they were referring to consumption of in excess of 40 g per day.

DH said the Corrao analysis included two study types: case-controlled studies, which compared people with cancer with healthy controls and assessed how much they drank in the past; and cohort studies that took a large group of healthy individuals, assessed their lifestyle and monitored them over a long period of time to ascertain which subjects developed cancer.

DH said they were aware that other lifestyle factors, such as diet, smoking and oral hygiene, could affect the relative risk of developing mouth cancer. However, they argued that the Corrao meta-analysis took this into account because it only picked up studies that adjusted for major risk factors such as smoking. DH explained that this meant the statistic quoted in the ad had already accounted for lifestyle variations and related solely to the effect of alcohol consumption.

DH explained that although they had relied on the Corrao study as substantiation for the statistic relating to the risk of developing cancer, they did not rely on it alone to establish a causal link between alcohol and cancer. DH said there was consistent evidence, from over 100 studies, that showed a clear dose-response relationship between the consumption of alcohol and the likelihood of developing mouth cancer. They said that the European Code Against Cancer and the International Agency for Research into Cancer (IARC), which was part of the World Health Organisation, had described the evidence linking alcohol to mouth cancer as "convincing" and "definitely established".

DH went on to add that alcohol had been classified as a category 1 carcinogen which was the highest risk category for determining if a substance caused cancer. Furthermore, DoH believed there was also a biological argument linking alcohol to cancer because alcohol metabolised into the carcinogen acetaldehyde, which was also classified as category 1 by the IRAC because it caused DNA damage. DH therefore believed there was a clear consensus that there was convincing evidence that alcohol was a carcinogenic in humans, and that alcohol consumption increased the likelihood of developing mouth cancer.

Clearcast said they considered the claims carefully at the pre-clearance stage to ensure they did not exaggerate the risks associated with drinking alcohol. They highlighted the fact that the ad specified the basis of the claim and also included on-screen text that made clear that the claim was based upon drinking a specific amount of alcohol per day. Clearcast said they also sent the substantiation to their medical consultant who confirmed that the claims were not false or misleading.

Assessment

Not upheld

The ASA acknowledged the complainants concern that the Corrao study used to support the advertised claims was a meta-analysis. However, we understood that meta-analysis was highly regarded within the medical profession as a means of assessing a large body of evidence, accounting for differences between trials, to form conclusions based on a large sample size. We therefore did not object to the use of a meta-analysis to support a claim, so long as the review itself was robust and relevant and its conclusions were reflected in the ad.

We noted that the Corrao study showed that at 50 g/day the relative risk of developing mouth cancer increased by three times, and that this figure had a 95% confidence level. We also noted the study stated "direct trends in risk were observed for cancer of the oral cavity and pharynx". We acknowledged the references made by DH to the categorisation of alcohol as a carcinogen by IRAC and considered that this further supported the risks associated with alcohol consumption. We noted that the National Institute of Clinical Excellence (NICE) stated within their guidance on head and neck cancers that "alcohol consumption is a particularly important risk factor for cancers of the mouth ... after adjustment for tobacco; the more alcohol consumed, the greater the risk". We considered, therefore, that there was medical consensus on the causal link between high alcohol consumption and the risk of developing cancer of the mouth.

We noted that the ad was phrased conditionally and stated "If youre a woman drinking two large glasses of wine or more a day, you could be putting your health at risk ... youre three times more likely to get mouth cancer", which we considered was appropriate within the context of assessing relative risk of developing cancer. We considered that the ad made clear that the claim "youre three times more likely to get mouth cancer" was based on women consuming in excess of 40 g of alcohol per day by drinking two large glasses of wine, which was supported by the Corrao study. We therefore concluded that the claim was unlikely to mislead.

The ad was investigated under CAP (Broadcast) TV Advertising Standards Code rules 5.1.1 (Misleading advertising) and 5.2.1 (Evidence) but we did not find it in breach.

Action

No further action necessary.

Adjudication of the ASA Council (Broadcast)