‘Green tea may lower the risk of colon, stomach and throat cancers in women’, says the Daily Mail, perhaps causing readers to rush to put the kettle on.
The Mail reports on a large, long-term observational study of people’s tea drinking habits and their risk of developing cancers of the digestive system.
However, the findings of the study require careful interpretation. The only statistically significant association found was a 14% decreased risk in any type of digestive system cancer (such as bowel cancer or stomach cancer) in people who regularly drank any type of tea compared to people who were not regular tea drinkers.
The media can be excused for its focus on green tea, as 88% of the tea drinkers in this Chinese cohort were drinking green tea. However, when the researchers restricted their analysis to the women who only drank green tea, the association between tea and cancer became only borderline statistically significant – at the upper end of the margin of error, the preventative effect could be zero.
Also, despite the headlines, no significant association was found between any tea consumption – or green tea alone – and risk of any specific digestive system cancer.
Overall, this is an interesting and well conducted study, but it does not provide conclusive evidence that drinking green tea - or any other tea – will influence your risk of cancers of the digestive system.
Where did the story come from?
The study was carried out by researchers from Vanderbilt University School of Medicine, Nashville, the National Cancer Institute, Rockville, US, and the Shanghai Cancer Institute, China. Funding was provided by the National Cancer Institute.
The study was published in the peer-reviewed open-access, American Journal of Clinical Nutrition.
The media reports of this study were fair, and while there is some slightly misleading interpretation about a ‘green tea effect’, this does not really affect the overall reporting of the conclusions.
What kind of research was this?
The researchers highlighted the fact that animal and laboratory studies have suggested that certain antioxidants found in tea may have a protective role against cancers of the digestive system. Antioxidants are molecules that are thought to protect against cell damage. However, previous observational studies in humans have given inconclusive results.
The current prospective cohort study aimed to see whether tea consumption affected the risk of digestive system cancers in middle-aged Chinese women.
The researchers made the choice to study Chinese women, as previous studies have focused on Japanese people, who have different drinking habits.
They decided to investigate women taking part in the Shanghai Women’s Health Study (SWHS) because the women in this cohort reported low levels of smoking and alcohol consumption. These other behavioural factors could have the potential to confound the relationship (for example, the amount of tea consumed could have a relationship to how much a person smokes tobacco and drinks alcohol – and both are known risk factors for cancers of the digestive system).
However, despite the researchers trying to reduce the possibility of confounding from these factors, there is still the possibility of confounding from other lifestyle or environmental factors which may influence tea drinking behaviour and cancer risk.
What did the research involve?
Between December 1996 and May 2000 the SWHS study recruited 74,941 women aged 40-70 years from seven urban areas in Shanghai, China.
At enrolment they were interviewed and completed a self-report questionnaire collecting information including:
- body measurements
- physical activity
- diet (including tea consumption)
- menstrual and reproductive history
- medical history
- occupational history
- information from each participant’s spouse (such as medical history and smoking and alcohol habits)
Diet and physical activity were assessed through previously well-established questionnaires for these types of factors.
The researchers excluded participants who had ever smoked or who drank alcohol regularly.
They also excluded those with missing data on the variables of interest, reported drinking implausibly high amounts of tea (more than 700 grams a month – the average tea consumption in the UK is around 150 grams a month), or reported any history of cancer.
Questions on tea included:
- the age they started drinking tea (or stopped if they no longer drank tea)
- if they drank tea regularly (defined as three or more times per week, continuously for longer than six months)
- the type of tea they drank and the amount of dry leaves used
Updated information on tea consumption was collected at follow-up at an average 2.6 years. Further two to three yearly surveys gained information on cancer diagnoses, which were confirmed via home visits and reviews of medical records. The cancer registry was also reviewed to confirm the site of cancer. Participation rates were above 95% at all follow-up points.
When conducting statistical analyses between tea consumption and risk of cancers of the digestive system the researchers took into account:
- marital status
- body mass index (BMI)
- waist-hip ratio
- physical activity
- meat, fruit and vegetable intake
- spousal smoking habits
- family history of cancers of the digestive system or diabetes
Tea drinkers were compared to women who never drank tea regularly (meaning they did not meet the above criteria for regular drinking).
Regular tea drinkers were further split into the following catogories:
- tea drinking for less than 15 years and less than 100 grams a month
- less than 15 years and 100 grams or more a month
- 15 years or more and less than 100 grams a month
- 15 years or more and 100 grams or more a month
What were the basic results?
A total of 69,310 women were followed for an average 11 years, during which 1,255 digestive system cancers occurred, including cancers of the stomach, oesophagus, colorectum (large bowel), liver, pancreas and gallbladder, or bile duct.
Just under a third of the women (28%) reported being regular tea drinkers. Most tea drinkers (88%) reported drinking green tea only or green tea in combination with black or scented tea (5%). Only a small proportion of women only drank other tea types:
- 3.54% only drank scented tea – that is, jasmine tea (white or green tea plus jasmine flowers) or green, black, or oolong tea in combination with herbs, other flowers, or fruit
- 1.1% drank black tea alone or in combination with scented tea
- 0.7% drank only oolong tea
- 1.4% drank other types of tea
On average, people drank 100g of tea per month, and the average duration of tea consumption was 15 years.
The researchers’ main finding was that, compared with women who never drank tea regularly, regular intake of any type of tea was associated with a 14% reduced risk of any cancer of the digestive system (hazard ratio 0.86, 95% confidence interval [CI] 0.74 and 0.98). There were also significant trends for the digestive system cancer risk to decrease as the amount of tea consumed and the duration of tea drinking increased.
However, when looking at the individual cancers, tea drinking had no significant effect on the risk of any specific digestive system cancer.
Also, despite the fact that most tea consumed was green, when they restricted their analyses to only those women who reported drinking green tea (either alone or in combination with other teas) the reduced risk of any type of digestive system cancers became only of borderline significance (hazard ratio 0.86, 95% CI 0.75 and 1.00).
Again, no significant association was found between only green tea consumption and any specific type of cancer (though in all analyses there was the same general direction of the effect – that is towards reducing risk).
How did the researchers interpret the results?
The media interpretation of this story appears to come from the researchers’ main conclusion: ‘In this large prospective cohort study, tea consumption was associated with reduced risk of colorectal and stomach/oesophageal cancers in Chinese women’.
However, though there was a general trend towards reduced risk for individual cancers, none of the analyses for these specific cancer types were statistically significant.
This study has strengths:
- it included a large sample of almost 70,000 Chinese women
- it followed them up for 11 years
- it reliably collected data on lifestyle factors and cancer outcomes
- it had high participation rates at all follow-up points
Importantly, the researchers have also adjusted their analyses for sociodemographic, lifestyle and medical factors which could have a potential confounding influence on cancer risk.
Further to this, the study benefits from studying only a non-smoking, non-drinking population of women: tea drinking may be related to smoking and alcohol intake, which are well established risk factors for digestive system cancers. As such, smoking and alcohol could be confounding any association between tea drinking and cancer, so it was useful that these two factors not been ruled out from the start.
Therefore, this is a well conducted study, but when interpreting the findings it is important to consider the following points:
- The only significant association found was a 14% decreased risk in any type of digestive system cancer with regular drinking (defined as three or more times per week, continuously for longer than six months) of any type of tea compared to non-regular drinking. The media focus on green tea is understandable as 88% of the tea drinkers in this cohort were drinking green tea. However, when the researchers restricted their analysis to the women who only drank green tea, the association between green tea and cancer became only borderline statistically significant.
- Also, despite media headlines and a general trend towards reduced risk, no significant association was found between any tea consumption – or green tea alone – and risk of any specific digestive system cancer.
- This study only included Chinese women, and therefore the results may not be applicable to men or women of different cultures, who may have very different tea drinking habits and other lifestyle habits or environmental exposures that may alter their risk of digestive system cancers.
- Related to this, it is not known whether the green tea described here would be exactly the same as the green tea sold in the UK, or whether the 88% of women in this cohort could just be describing that their tea leaves were green in colour, as opposed to black.
Overall, this is an interesting and well conducted study, but it does not provide conclusive evidence that drinking green tea – or any other tea – will influence your risk of digestive system cancers.
Analysis by Bazian
Edited by NHS Website
Links to the headlines
Daily Mail, 1 November 2012
Links to the science
The American Journal of Clinical Nutrition. Published online October 10 2012