"Three cups of coffee a day could help keep diabetes away," The Independent reports. A US study has found a link between increased coffee intake and a small – but significant – reduction in type 2 diabetes risk, but not proof that drinking coffee prevents diabetes.
The study found that adults who increased their coffee consumption by more than one cup a day reduced their risk of developing type 2 diabetes by around 11% in the following four years. The reduction in relative risk held for a range of initial coffee consumption levels and appeared to remain up to 16 years later.
Coffee consumption was assessed every four years. It appeared that increasing coffee consumption during this period made a difference to diabetes risk in the subsequent four years and longer.
Coffee has many biologically active compounds, making it plausible that it could influence diabetes disease risk. However, this study falls short of proving that it does. The researchers did adjust the results for possible underlying factors that could explain the association, but, as with any study of this kind, some factors may have been overlooked.
Even if there was a direct cause and effect relationship, the reduction of risk was modest. It could be dangerous to assume that upping your coffee intake will protect you against diabetes risk if you continue to ignore established diabetes risk factors, such as lack of exercise and obesity.
Where did the story come from?
The study was carried out by researchers from Harvard School of Public Health and was funded by research grants from the US National Institutes of Health. One of the authors also received support from the American Heart Association.
One of the authors declared a potential conflict of interest in that he received grant funding from Nestec (the company behind the Nestlé brand), a multinational that produces many goods, including coffee. He said that this is an ongoing arrangement and Nestec had no influence on the design of the study or the decision to publish the results.
The study was published in the peer-reviewed medical journal, Diabetologia.
The article in The Independent was balanced, discussing the strengths and limitations of the study. But the Daily Mirror's and the Daily Express' coverage did not meet the same standard. Both papers made claims along the lines of "drinking more coffee immediately cuts your risk of diabetes". This wasn't backed up by the study's findings.
Most media reports included the important fact that even if people who drank more coffee did tend to have a lower risk of type 2 diabetes, it does not necessarily follow that coffee consumption was directly responsible.
What kind of research was this?
A cohort study is one of the best methods of assessing a change in behaviour, such as increased coffee consumption, because it tracks behaviour over time. However, it cannot prove that changes in coffee consumption directly affect disease risk – it can only suggest whether this is likely or not.
A randomised controlled trial (RCT) would be required for proof of cause and effect, but these also have practical drawbacks, such as being complex to organise, with many people potentially dropping out, and being very expensive to perform.
What did the research involve?
The researchers used information from diet questionnaires and new cases of type 2 diabetes from three existing long-term cohort studies. The information was used to study whether changes in coffee consumption influenced the risk of developing type 2 diabetes.
Diet was assessed every four years using a food frequency questionnaire. Any change in coffee consumption was therefore pieced together in four-year gaps.
New cases of type 2 diabetes were self-reported by participants using questionnaires sent every two years and subsequently validated with a follow-up survey on symptoms to confirm the diagnosis.
The existing cohort studies were:
- 48,464 women in the Nurses' Health Study (1986-2006)
- 47,510 women in the Nurses' Health Study II (1991-2007)
- 27,759 men in the Health Professionals Follow-up Study (1986-2006)
The main analysis compared the risk of developing type 2 diabetes in people who changed their coffee consumption in the four-year period, comparing them with those who did not change their consumption habits. This was carried out for each cohort separately to assess the consistency of findings, and the results were then combined.
The analysis took many other factors affecting the risk of type 2 diabetes (confounders) into account, including:
- family history of diabetes
- menopausal status and postmenopausal hormone use
- oral contraceptive use
- history of hypertension and hypercholesterolaemia
- change in smoking status
- changes in physical activity
- intake of alcohol and other beverages
- changes in coffee and tea consumption
- history of physical examination
- baseline body mass index (BMI)
- weight change
What were the basic results?
During the follow-up periods, which spanned up to 20 years, there were 7,269 new cases of type 2 diabetes recorded. The main statistically significant findings from the pooled and adjusted analysis were:
- Participants who increased their coffee consumption by more than one cup per day (median change = 1.69 cups per day) over a four-year period had an 11% (95% confidence interval [CI] 3-18%) lower relative risk of developing type 2 diabetes in the subsequent four years, compared with those who made no changes in consumption.
- Participants who increased their coffee consumption by less than one cup per day did not significantly change their disease risk compared with those who made no change at all.
- Participants who decreased their coffee intake by more than one cup per day (median change = −2 cups per day) had a 17% (95% CI 8% to 26%) higher relative risk of developing type 2 diabetes. These people tended to be older and have greater weight gain than their comparison group over four years.
- Participants who decreased their coffee consumption by less than one cup per day did not significantly change their disease risk compared with those who made no change at all.
- Changes in tea consumption were not associated with type 2 diabetes risk.
After assessing disease risk in the subsequent four years, the researchers assessed disease risk over longer time periods: 12 years in one cohort and 16 in the other two. In these pooled multivariable analyses:
- Participants who increased their total coffee consumption by more than one cup per day had a 13% lower risk (95% CI 5% to 21%) of type 2 diabetes compared with those who had relatively stable intakes.
- However, decreasing coffee consumption by more than one cup per day was not associated with an increased risk for type 2 diabetes (HR 1.09; 95% CI 0.92 to 1.30).
How did the researchers interpret the results?
The researchers concluded that, "Increasing coffee consumption over a four-year period is associated with a lower risk of type 2 diabetes, while decreasing coffee consumption is associated with a higher risk of type 2 diabetes in subsequent years."
This study indicates that people who increase their coffee consumption by more than a cup a day may reduce their relative risk of developing type 2 diabetes in the subsequent four years compared with those who make no change.
This decrease in relative risk was still apparent using longer time periods of as much as 16 years. No such association was found for tea at any time, and the findings were independent of initial coffee consumption.
This study's strengths were that it recruited a large amount of people, adjusted for many potential confounders, and was long term. However, because of the cohort design we cannot be sure that coffee is causing the difference in disease risk for sure, as it could be down to other factors.
An RCT would be needed to prove a causal link. However, as the study authors mention, these studies are difficult to conduct given the long follow-up time needed to detect the development of type 2 diabetes.
There are also problems with high costs, uncertainty regarding the ideal intervention period, and the possibility of a large number of participants failing to adhere to the assigned beverage. So, this cohort approach is a practical next-best approach.
The finding that there was a higher risk of type 2 diabetes linked with decreasing coffee intake may represent a real change in risk related to coffee consumption, or it may be reverse causation. That is, people with medical conditions associated with a risk of developing type 2 diabetes may reduce their coffee consumption after diagnosis.
The study authors were aware of this, so they removed medical conditions from their pooled and adjusted analysis. However, there was little difference between the initial results and the adjusted results, suggesting that reverse causation was not responsible.
Coffee has many biologically active compounds, making it plausible that it could influence disease risk. But this study falls short of proving that this is the case.
Proven methods of reducing your diabetes risk include losing weight if you are overweight, eating a healthy varied diet, and taking regular exercise. Read more about reducing your diabetes risk.
Analysis by Bazian
Edited by NHS Website
Links to the headlines
The Independent, 25 April 2014
Daily Mail, 25 April 2014
The Daily Telegraph, 25 April 2014
Daily Mirror, 24 April 2014
Daily Express, 25 April 2014
Links to the science
(zip file, 673kb). Diabetologia. Published online April 25 2014