"Even one glass of wine a day raises the risk of cancer: Alarming study reveals booze is linked to at least seven forms of the disease," reports the Mail Online.
The main finding was that existing evidence supports the link between alcohol consumption and cancer at seven sites, including the throat, gullet, liver, colon, rectum and female breast.
The links were said to be strongest for heavy drinking, but this study suggested that even low or moderate drinking may contribute to a significant proportion of cancer cases because of how common this level of drinking is. The study also suggests there's no evidence of a "safe" level of drinking with respect to cancer.
However, it's important to be aware that this review doesn't state how the author identified and assessed the research they've drawn upon. We don't know whether all relevant research has been considered and the conclusions must be considered largely the opinion of this single author.
Nevertheless, the main finding of the link between alcohol and these seven cancers is already well recognised. Recently updated government recommendations state there's no safe level of alcohol consumption, and men and women are advised not to regularly drink more than 14 units a week. This review further supports this advice.
Where did the story come from?
The study was carried out by one researcher from the University of Otago, New Zealand. No external funding was reported.
Generally the media coverage of this topic was accurate, although the tone of the reporting tended to suggest this was a new discovery, when the link between alcohol and certain types of cancer is well established.
What kind of research was this?
This was a review which aimed to summarise data from published biological and epidemiological research, and meta-analyses that have pooled data, to evaluate the strength of evidence that alcohol causes cancer.
Alcoholic drinks have been considered potentially carcinogenic (cancer causing) for a while, but there are still concerns about the validity of some observational studies finding links with cancer, and uncertainty about precisely how alcohol causes cancer.
A systematic review is the best way of gathering and summarising the available research around a particular topic area. But in this case the exact methods are not described in the paper and it's not possible to say whether they were systematic.
There's a possibility that some relevant research may have been missed and that this review is giving an incomplete picture of the issue.
What did the research involve?
The author of this review reports drawing upon biological and epidemiological research as well as meta-analyses conducted in the last 10 years by a number of institutions, including the World Cancer Research Fund and American Institute for Cancer Research, the International Agency for Research on Cancer and the Global Burden of Disease Alcohol Group.
The majority of epidemiological research seemed to come from cohort and observational studies.
The research was reviewed and summarised in a narrative format which explored the evidence that alcohol causes cancer, while contrasting this with the notion that alcohol consumption may offer some form of protection from cardiovascular disease.
No methods are provided and the author does not describe how they identified the research, as you would expect from a systematic review. For example, they do not give the literature databases searched, the search dates, search terms, study inclusion or exclusion criteria, or descriptions of how studies were quality assessed.
What were the basic results?
There were several findings from this study, the main one being that existing evidence supports the link between alcohol consumption and cancer at seven sites: oropharynx (mouth and throat), larynx (voice box), oesophagus (gullet), liver, colon (bowel), rectum and female breast.
The strength of the association differed by the site of the cancer. It was strongest for the mouth, throat and oesophagus, with the review suggesting that someone who drinks more than 50g of alcohol a day is four to seven times more likely to develop these types of cancer compared to someone who doesn't drink. As the author says, the interaction of smoking with alcohol is also believed to contribute to the risk of these cancers.
The link was comparatively weaker for colorectal, liver and breast cancer. The review suggests someone who drinks more than 50g of alcohol a day is 1.5 times more likely to develop these types of cancer compared to someone who doesn't drink.
For all of these associations there was a dose-response relationship, where increased consumption was linked with an increase in cancer risk. This applied to all types of alcoholic drinks. The highest risks were associated with heavier drinking. There was also some suggestion that the level of risk goes down over time when alcohol consumption stops.
Recent large studies have found uncertain evidence whether low to moderate consumption has a significant effect on total cancer risk. But given that this level of consumption is common in the general population, the author considers that it could still contribute to a significant number of cases.
Furthermore, they say there is no clear threshold of what constitutes a harmful level of alcohol consumption, and therefore no safe level of drinking with respect to cancer.
The author also suggests that confounding factors may be responsible for the protective effect between alcohol consumption and cardiovascular disease that has been found in previous studies. For example, this may be due to the potential bias caused by misclassification of former drinkers as abstainers.
The research went on to report that alcohol is estimated to be responsible for approximately half a million deaths from cancer in 2012 and 5.8% of cancer deaths worldwide, deeming it to be a significant public health burden.
How did the researchers interpret the results?
The author concluded: "There is strong evidence that alcohol causes cancer at seven sites, and probably others. The measured associations exhibit gradients of effect that are biologically plausible, and there is some evidence of reversibility of risk in laryngeal, pharyngeal and liver cancers when consumption ceases."
"The highest risks are associated with the heaviest drinking, but a considerable burden is experienced by drinkers with low to moderate consumption, due to the distribution of drinking in the population."
This narrative review aimed to summarise data from published biological and epidemiological research to discuss the strength of evidence that alcohol causes cancer.
The author gives their main finding as a link between alcohol consumption and cancer at seven sites, and also that the highest risks seem to be associated with heavier drinking. However, they state there's no "safe" drinking threshold and that low to moderate consumption still contributes to a significant number of cancer cases.
The biggest limitation of this review is that it doesn't appear to be systematic. The author provided no methods for how they identified and appraised the research they drew on. Despite referencing a number of large studies and reviews, this study and its conclusions have to be considered largely the opinion of the author following their appraisal of the evidence.
We don't know whether the review has considered all research relevant to the topic and is able to reliably quantify the risks of cancer – overall or at specific sites – associated with alcohol consumption.
An additional limitation to keep in mind is that this data mainly appeared to be from observational studies. These cannot prove cause and effect. The individual studies will likely have varied considerably in the additional health and lifestyle factors they took account of when looking at the links with alcohol. For example, smoking, diet and physical activity are all factors likely to be associated both with level of alcohol consumption and cancer risk.
As the author notes in particular, confounding factors may be responsible for the observed protective effect between alcohol consumption and cardiovascular disease.
Another limitation is that alcohol consumption is likely to be self-reported in the studies analysed, which may be inaccurate and lead to misclassification. For example, a potential bias that the author notes is classifying former drinkers as abstainers.
The author does consider the limitations of these observational findings, saying: "The limitations of cohort studies mean that the true effects may be somewhat weaker or stronger than estimated currently, but are unlikely to be qualitatively different."
But despite the methodological limitations of this review, it does support current understanding around this topic. Cancer Research UK also reports that alcohol can increase risk of these seven cancers and that there is no "safe" alcohol limit.
While we can't give a safe limit to drink when it comes to cancer, people are advised to follow current alcohol recommendations, which are to drink no more than 14 units per week and to spread your drinking over three days or more if you drink as much as 14 units a week.
Analysis by Bazian
Edited by NHS Website
Links to the headlines
Mail Online, 22 July 2016
The Guardian, 22 July 2016
The Telegraph, 22 July 2016
Daily Mirror, 22 July 2016
The Independent, 22 July 2016
Links to the science
Addiction. Published online July 21 2016