Diabetics have been urged to cut out coffee, according to a news article in the Daily Mail. The newspaper reports that an American study has shown that “a daily dose of caffeine raises blood sugar by 8 per cent”. They go on to say that drinking caffeine may undermine the effects of medication and that simply giving up drinks containing caffeine may be a way of lowering blood sugar.
The researchers used a sensor implanted under the skin of 10 people with type 2 diabetes to measure changes in glucose when the participants took caffeine capsules containing the equivalent of about four cups of brewed coffee a day. The design of this study, the small number of participants and the short timescale all indicate that it is unwise to issue advice based on this research alone. Confirmatory research using randomised designs and larger numbers of patients is needed.
Where did the story come from?
Dr James Lane and colleagues from the Duke University Medical Centre, Durham, USA carried out the research. It is not clear from the online version, which was published ahead of print, who funded the study. The study was published online in the peer reviewed medical journal: Diabetes Care.
What kind of scientific study was this?
The researchers picked five men and five women (average age 63) from their clinic who were habitual drinkers of brewed coffee. All the participants had been diagnosed with type 2 diabetes at least six months before the study began, and their condition was being managed by a stable regimen of diet, exercise, and oral tablets, but not insulin injections.
Other than diabetes, they were all free of major medical disorders, were non-smokers, and had not been prescribed other medications known to affect glucose metabolism. They were also mostly overweight or obese with an average body mass index (BMI) of about 32. Blood tests indicated that they managed their diabetes well. The measure of HbA1c was low, an average of 6.4%, suggesting that sugar levels had been well controlled over the past 12 weeks.
A questionnaire asked the participants to report their usual drink consumption and this was used to calculate their average daily caffeine intake as 520mg per day, with a wide variation in the amount individuals took.
Glucose concentration was monitored using the continuous glucose monitoring system (CGMS), a device inserted just under the skin of the abdomen. The device gives the average glucose concentration every five minutes throughout the day.
Following insertion of the device, the participants took 250mg of caffeine in capsules at breakfast and the same again at lunch. The participants all had the same liquid breakfast of 720 cal, and their usual diet at lunch and dinner. They recorded what they ate, their medication and avoided strenuous exercise over the study. Each participant took caffeine capsules for 24 hours and placebo (dummy capsules) for 24 hours. The sensor was removed on the third day.
What were the results of the study?
The average 24-hour glucose concentration curves show that caffeine increased average levels of daytime glucose (6am to 10pm) compared to placebo.
The 0.6mmol/L increase in average glucose level over 24 hours compared to placebo was statistically significant. The average level was 7.4 mmol/l on the placebo days compared to 8.0 mmol/L on the caffeine days.
The researchers say that average glucose concentrations were also elevated in the three hours following the standardised breakfast (8.7 compared to 8.0 mmol/l), lunch (7.8 compared to 6.8 mmol/l), and dinner (8.6 compared to 6.8 mmol/l) on the days that caffeine was consumed.
What interpretations did the researchers draw from these results?
The researchers say that “Caffeine had adverse effects on glucose metabolism, producing higher average daytime glucose concentrations”, and exaggerated glucose responses after meals.
What does the NHS Knowledge Service make of this study?
This small study in people living with diabetes, adds to laboratory data that has suggested a similar effect. However, the authors acknowledge that it is uncertain how caffeine affects diabetics. There are also other limitations to the interpretation of this study:
- Apart from the standardised breakfast drink, the researchers did not control the calorie intake of the volunteers, who may have compensated for a lack of caffeine by eating more.
- The small number of participants means that differences between the groups may have arisen by chance
- The authors base their conclusions on changes in glucose concentration after meals. However, their statistical analysis was performed on average readings over 24 hours, or for three hours after meals. It would have been possible to perform further statistical analysis, which took into account the similarities and differences between the groups at the start of the study, but this was not done.
- There was no “wash-out period” between the active and placebo groups, so it is possible that the lower glucose concentrations on the placebo day could have occurred as a response to caffeine withdrawal. The report does not analyse the data according to whether caffeine or placebo was taken first. Though the groups were balanced in some unexplained way.
- The authors themselves point out that the reverse pattern is seen overnight, where caffeine appears to reduce blood sugar levels.
Longer studies with randomised controlled designs, in which participants are followed-up for longer, and ideally with periods between measurements that allow the short-term effects of caffeine to wear off, will help to reveal whether the effect shown here is real and if it applies to all people with type 2 diabetes.
Sir Muir Gray adds...
An overdose of almost anything carries risks and the numbers of people in this study are too small to make firm recommendations for all people with diabetes. If in doubt, people with diabetes could try to build an extra 30 minutes of walking a day into their life style; that would compensate for the coffee effect - if it is real, and have its own benefits - if not.
Analysis by Bazian
Edited by NHS Website
Links to the headlines
The Daily Telegraph, 28 January 2008
Daily Mail, 28 January 2008
The Guardian, 28 January 2008
Links to the science
Diabetes Care 2007; Oct 31 [Epub ahead of print]