'Drinking several cups of coffee every day can significantly reduce the risk of bowel cancer' advises the Daily Mail. The story is based on US research that found that drinking four or more cups a day leads to a small but significant reduction in risk…
‘Six cups of coffee a day could cut the risk of bowel cancer by 40 per cent’ the Daily Mail explains, while The Daily Telegraph points out that just ‘four cups a day’ leads to a 15% rate of reduction.
The news follows a long-term study that tracked the behaviour and associated health outcomes (cohort study) of just under half a million Americans over the course of 10 years.
At the start of the study participants completed questionnaires on their diet and lifestyle and during the follow-up period the researchers looked at the number of colorectal cancers (commonly known as bowel cancer) that developed.
Those who drank four to five cups of coffee (caffeinated or decaffeinated) a day had a 15% reduced risk of developing cancer, while those who drank six or more had a 26% reduced risk, compared with non-drinkers.
Sadly for tea drinkers, there was no reduction in bowel cancer risk, but being a US study, the number of tea drinkers was relatively small.
This was a well-conducted study that does suggest a link between coffee consumption and reduced bowel cancer risk. However, the researchers did recommend that further investigation into the link is needed, including study of the specific chemicals in coffee that could be having an effect.
It should be noted that caffeine is a stimulant, and drinking excessive amounts of it can lead to unwanted side effects, such as restlessness, insomnia and dehydration.
There are more well established methods of reducing your risk of bowel cancer such as taking regular exercise, eating a healthy diet and quitting smoking.
Pregnancy and breastfeeding
If you are pregnant then it is recommended that you drink no more than 200mg of caffeine a day (roughly two mugs of instant coffee), as higher levels than this have been linked to complications such as miscarriage and low birth weight.
You should also avoid drinking caffeine if breastfeeding as it can pass into your breast milk which can make your baby irritable and cause sleeping problems.
Where did the story come from?
The study was carried out by researchers from the US National Cancer Institute and Imperial College, London, and was funded by the Intramural Research Program of the US National Cancer Institute.
The study was published in the peer-reviewed American Journal of Clinical Nutrition.
The media is generally representative of this research, though they do not spell out some of the inherent limitations of a cohort study. This is important as it means that it is difficult to say whether a direct link exists between coffee consumption and bowel cancer risk. There may be other unidentified lifestyle factors that contributed to bowel cancer risk.
Some news stories also report the study’s finding that tea had no effect on cancer risk, without mentioning that few heavy tea drinkers were included in the study.
The advice from the Daily Express – supposedly from an ‘unnamed’ bowel cancer charity – that you should ‘consult your GP about coffee drinking first’ seems a little bizarre. In the absence of any serious health conditions you probably don’t need to check with your doctor if it is okay to drink four cups of coffee a day.
What kind of research was this?
This was a cohort study – the National Institutes of Health-American Association of Retired Persons (NIH-AARP) Diet and Health Study – which assessed various aspects of diet and lifestyle and has followed participants for a number of years looking at various health outcomes. This particular study examined information on coffee consumption which was collected at the start of the study, and looked to see how this related to the development of colorectal cancer.
Such a cohort study can demonstrate associations, but it cannot prove cause and effect as there may be other unmeasured factors that are associated with both coffee drinking and bowel cancer risk and so underlie the observed link. Also, self-reported coffee consumption at one point in time may not be a reliable measure of longer-term patterns. The ideal way to investigate the effects that dietary factors have upon a specific health outcomes would be a randomised controlled trial (RCT). However, an RCT of this type is arguably not feasible due to a number of factors, such as:
- adherence (making sure people are not drinking too much or too little coffee)
What did the research involve?
Between 1995 and 1996 the NIH-AARP Diet and Health Study recruited AARP members, aged between 50 and 71 years old, from eight areas in the US. This study included 489,706 participants.
At the start of the study, participants completed a questionnaire which included information on sociodemographics (factors such as occupation and income), as well as:
- body measures
- lifestyle (such as physical activity and smoking)
This questionnaire included a 124-item food frequency questionnaire, which included information on intake and portion sizes over the past year. The researchers used the results of the questionnaire to classify people according to their coffee and tea intake. Coffee intake was split into six categories as follows:
- less than one cup a week
- one cup a day
- two to three cups a day
- four to five cups a day
- six or more cups a day
Tea was assessed in five slightly different categories:
- less than one cup a month
- one to three cups a month
- one to six cups a week
- at least one cup a day
Tea and coffee intake were also classified depending on whether the coffee or tea was caffeinated or decaffeinated more than half of the time. The researchers further validated their assessments by carrying out two, 24-hour dietary assessments of a sample of 1,953 participants. This assessment included information on whether the coffee was ground, instant or espresso.
Cancer cases were identified through linking to state cancer registries, which included codes for the specific type of cancer. Participants were followed for an average of 10.5 years to the end of December 2006.
The researchers analysed the association between tea and coffee intake and the development of cancer. In their analyses they adjusted for various potential confounding factors, including:
- physical activity
- body mass index (BMI)
- red meat consumption (a diet rich in red meat is a known risk factor for bowel cancer)
- fruit and vegetable consumption (a diet rich in fresh fruit and vegetables can reduce bowel cancer risk)
- alcohol intake
- the use of hormone replacement therapy (HRT) in women
What were the basic results?
There were 6,946 new cases of colorectal cancer during the 10.5 years of follow-up. In the baseline assessments, approximately 90% of the cohort drank coffee, and 16% reported drinking four or more cups a day.
- Compared with non-drinkers of coffee, those who drank four to five cups a day had a 15% reduced risk of developing colorectal cancer (hazard ratio 0.85; 95% confidence interval (CI) 0.75 to 0.96).
- Compared with non-drinkers of coffee, those who drank six or more cups a day had a 26% reduced risk (hazard ratio 0.74, 95% confidence interval 0.61 to 0.89).
- There was no significant association between coffee and cancer for those who drank three or fewer cups a day (neither increased nor decreased risk).
- When they split into caffeinated and decaffeinated coffee, the only significant associations were a 17% decreased risk for those who drank six or more cups of caffeinated a day, and a 21% decreased risk for those who drank four to five cups of decaffeinated a day.
- No significant associations were observed according to categories of tea intake. However, the majority of participants did not drink tea, with 62% of participants reporting no tea consumption at baseline.
How did the researchers interpret the results?
The researchers conclude that in their large US cohort, coffee consumption was inversely associated with colorectal cancer (that is, as intake goes up, risk goes down).
This study finds an association between coffee drinking and bowel cancer risk, and benefits from its large sample size and thorough follow-up of outcomes. It does, however, contain important limitations:
Reduction of risk was only seen at high levels of coffee consumption
The results don’t give a clear message about the effects of caffeinated or decaffeinated coffee. The only significant associations observed were for six or more cups of caffeinated coffee a day (but not for four to five); and for four to five cups of decaffeinated coffee a day (but not for those who drank six or more cups). All other risk associations were non-significant. Such seemingly random significant results makes it possible that these could be chance associations.
When looking at coffee overall (both caffeinated and decaffeinated) the only risk reductions observed were for those who drank either four to five, or six or more cups a day.
These groups of very high levels of coffee consumers actually represented a small minority of the people actually taking part in the study. And the smaller the numbers detected by researchers (the sample size), the more likely it is that any effect detected by the researchers could be the result of chance.
Possible inaccuracies in reported coffee consumption
Coffee consumption was self-reported at one point in time only. This may not be a reliable measure, and may not be representative of longer-term patterns. Also, specific information on the type of coffee consumed was only gained from the small subset of participants who completed the 24-hour recall.
As the researchers confirm, no information was taken on factors on such as:
- the type of coffee bean
- how the coffee was brewed
- the amount of caffeine found in caffeinated coffee
It is also worth noting that although they found no association between tea drinking and bowel cancer, there were few heavy tea drinkers in the cohort. So it is difficult to give a true assessment of the effects of tea drinking (again with a small sample size the potential protective effects of heavy tea drinking could be ‘scrambled’ by the effects of chance).
In addition, the researchers collected no information on the type of tea drunk, and therefore this category included a range of tea types, including ‘normal’, herbal and green teas.
Possibility that other factors underlie the association
The research demonstrates an association between coffee drinking and bowel cancer risk, but it cannot prove causation. Though the researchers have made careful attempts to adjust for many other risk factors for bowel cancer that may be involved in the relationship, it is still possible that there are other, unmeasured factors that are associated with both coffee drinking and bowel cancer risk and so underlie the observed link. As a purely theoretical example, it could be the case that drinking decaffeinated coffee has no direct preventative effect. However, people who do drink ‘de-caff’ tend to be more health conscious, eat a healthier diet and exercise more – and this could lead to the reduction in risk.
It is always difficult to specifically take into account all the different aspects of a person’s diet and lifestyle.
May not apply to other population groups
This was a study of retired US citizens, who were predominantly white and well-educated. The findings may not apply to people of younger age groups, or in other countries where different society and environmental factors may influence both their coffee consumption and cancer risk.
Further research is needed
As the researchers acknowledge, further investigation into the link is needed, in particular looking at the specific chemical constituents of coffee that could be having an effect.
For now, it is important to consider that it is alright to consume most things in moderation. Caffeine is a stimulant, and drinking excessive amounts can give other unwanted side effects. Anyone wanting to reduce their risk of cancer should ensure that they consume a healthy diet, take exercise and stop smoking.
Analysis by Bazian. Edited by NHS Choices. Follow Behind the Headlines on twitter.