I read an interesting study this week that examined the relationship between changes in coffee and tea consumption and risk of developing type 2 diabetes over a 4-year period [1]. The authors found no association between tea consumption and diabetes risk, but it was a different story for coffee: increasing coffee consumption was associated with a reduced risk of type 2 diabetes over the 4-year study period. Also, overall, in individuals who reduced their coffee consumption, risk of diabetes went up.
The authors suggest that a possible explanation for the latter finding is ‘reverse causality’. Essentially, the idea here is that as people get sick, they then reduce their coffee consumption. In other words, illness leads to reduced coffee consumption, not the other way round. However, even then the analysis was performed to adjust for this (by excluding those with other illnesses), the results were pretty much the same.
The authors also found that the benefits associated with coffee consumption only existed for caffeine coffee (not decaffeinated). However, as they explain, decaffeinated consumption was generally low and the numbers of relevant people in the study may not have been enough to pick up any effect.
Other evidence ha founds a link between caffeinated coffee and reduced risk of type 2 diabetes (but no such link for decaffeinated coffee) [2].
Before we get too far ahead of ourselves, I feel it’s worth reminding ourselves here that these studies (so-called ‘epidemiological’ studies) cannot be used to conclude that drinking caffeinated coffee reduces the risk of type 2 diabetes (only that there is an association).
However, we do have some ‘clinical’ evidence that suggests that coffee may indeed be protective. In a study published in the American Journal of Clinical Nutrition, the effect of coffee-drinking on a variety of biochemical markers of disease was assessed [3].
A group of coffee drinkers were asked to abstain from drinking coffee for a month. The following month they were asked to drink four cups of coffee a day (a total of 600 mls of coffee a day). The month following this they were instructed to drink 8 cups of coffee a day.
Compared to drinking no coffee, drinking 8 cups a day was associated with significantly reduced levels of inflammatory markers (interleukin-18 and 8-isoprostane), as well as significantly raised levels of adiponectin. This hormone is secreted by fat cells, and has been shown to have generally beneficial effects on the body’s physiology including an ability to help the functioning of the hormone insulin. Broadly, these effects from drinking coffee would be expected to translate intorelative protection from type 2 diabetes.
While the evidence is not definitive, there is I think significant evidence that supports the idea that drinking coffee may help to reducerisk of succumbing to type 2 diabetes. There is some type 2 diabetes in my family, and I do drink coffee. I imagine I’ll continue to drink it (and enjoy it) for some time yet.
References:
1. Bhupathiraju SN, et al. Changes in coffee intake and subsequent risk of type 2 diabetes:
three large cohorts of US men and women. Diabetologia DOI 10.1007/s00125-014-3235-7
2. Floegel A, et al. Coffee consumption and risk of chronic disease in the European Prospective Investigation into Cancer and Nutrition (EPIC)-Germany study. Am J Clin Nutr 2012;95:901-908
3. Kempf K, et al. Effects of coffee consumption on subclinical inflammation and other risk factors for type 2 diabetes: a clinical trial. Am J Clin Nutr 2010;91:950-957
Hi John
An interesting article re coffee consumptions and protection from type 2 diabetes. It also appears to confer protection form various cancers.
I can;t stand the taste or even the smell of coffee. Is there anther natural alternative that would afford the same protection that you are aware of please?
Best wishes
Sonia
If there was a proper study into the effects of tea and coffee on T2 diabetes, (Caffeine?) how would anyone allow for people who take sugar with their tea and coffee and those who don’t?
Hi Dr Briffa,
What about the link between coffee and protection against liver cancer?
Yes but what kind of coffee – instant or “proper” coffee. Surely that is what has to be stated.
For the liver cancer/cirrhosis protective effect, it makes little difference whether coffee is instant or espresso. But decaf is as useless as it tastes.
Even people who take sugar in their coffee don’t usually take as much sugar that way, compared to people who drink soft drinks. I like 1 gram of sugar per cup myself.
Yes, and worth saying too that even if this is not proven, there are established well being (feel good) benefits from buying fair trade coffee.
If we consider that fatty liver is the precursor of T2D, we are on firmer ground, as the hepatoprotective effect of coffee that appears in epidemiology has been confirmed by animal testing and even some human experimentation. And the correlation between drinking coffee (around 4 cups) and avoiding cirrhosis or liver cancer is a strong one – around 0.5 OR.
I think you will appreaciate the Dietary Guidelines submission in the PDF attached to this blog
http://profgrant.com/2014/04/23/the-real-food-guidelines/
It’s a detailed argument, mainly from epidemiology, around the neutral role of SFA in the diet, the benefits of full-fat diet, the over-sold benefits of fibre and risks of salt, and other matters you’ve discussed on this blog.
In the absence of having a measurable amount of caffeine per ‘cup’ (http://en.wikipedia.org/wiki/Cup_(unit)) I’ll drink my coffee because it tastes good and provides a good incentive to get out of bed in the morning.
Over the years I have been doing my own little study of N = 1. I drink at least a dozen cups of coffee a day and have done do for years. I work as a Hospitalist and I just finished a two week stretch of 12 hour days in a high stress environment so caffeine is a bit mandatory. My blood sugar is normal and I show no signs of diabetes. Of course I also eat a Paleo style diet.
The point is, in my opinion type II diabetes is 100% preventable. We discuss diabetes like it’s something people get as they get older. No–it’s a self-induced condition based on a diet of processed food and lack of exercise. In other words you give yourself diabetes.
With the benefit of hind sight I tend to agree. I got the news in 2000 followed the official line and things gradually declined as predicted. Only in 2007 with the prospect of insulin treatment at which I went Lo Carb did any improvement take place. Result lost 4 stone and now probably leaner and healthier than before.
I have said before that getting diabetes was actually a good thing ( depending of course what you do about it !)
I disagree that it is a self induced illness.
I do however note that people who live long lives, tend to credit themselves for it. From your remarks about Paleo etc. I think perhaps you are crediting yourself for your good health.
Not EVERYONE with bad health gets it from an unhealthy lifestyle.
Sorry Dr Wilson but I strongly disagree with your statement that Type 2 is 100% preventable. A couple of years ago I was diagnosed with Type 2. My A1c at that time was a whopping 14.5. I was shocked to learn of the diagnosis as were my friends & relatives because I was such a healthy eater and always active and lean. Also, there was no history of diabetes in my family. Why I came down with it is beyond me.
Paul–You likely developed “type 1.5 diabetes”. This type of diabetes occurs in adults yet it is likely mostly an autoimmune disease like type I diabetes. No, this type of diabetes is likely not preventable.
Help! Now I am confused. I have read repeatedly that caffeine stimulates insulin, and should be avoided by people who are insulin resistant or wanting to lis weight. So is caffeine in coffee not insulinogenic?
I am getting tired of these reports on observational studies. The best they can do is to suggest an area for additional research. They do not even warrant an article! Correlation or association cannot determine causation. In addition, the quality of the data from these studies is almost always weak. All these articles do is to add to the confusion as to what people should do to prevent X, benefit the author, publisher, and the industry associated with the study.
As for the subject at hand, for my own N=1 study, I drank 4-6 cups per day for the last forty years and then last year I was diagnosed with type II diabetes. So the hypothesis does not appear true from my perspective.
What I do know now is that I should have laid off the carbs a long time ago. I wish that I had known. Instead I was following the so called ‘heart healthy’ low fat (high carb) diet on statins which I now think contributed to my diabetes.
This finding surprises me, though I am prepared to be open-minded about the epidemiology. My first thought was to wonder about the physiology, such as effects upon hormones, and the way caffeine is metabolised.
In high doses I’d expect caffeine to be endocrinology disruptive in either a primary way, ie., acting directly upon hormones and resulting in altered sates of balance, or via secondary effect through alteration to time spent resting and sleeping. (Sleep plays a big part in ‘normalising’ hormones and hormonal cascades.)
Insufficient sleep can be disruptive to cortisol and to insulin and in nature long hours of daylight in summer are a cue to raise cortisol and insulin to keep many spepcies active and preparing for winter (eating and gaining weight under the influence of seasonally raised levels of insulin).
Elevated levels of cortisol and other glucocorticoids can raise the requirement for methylation, and methylation is rhe means by which these exito-hormones – which can be exito-toxins under conditions of chronic excess – are down regulated. So caffeine in excess might place increased demand for methylation and methyl donors which are supplied by methyl donating antioxidants. If supply of methyl donating antioxidants is not in keeping with demand some reactive oxygen species can persist, and an indication might be rising levels of homocysteine. Homocysteine both conveys and indicates generally raised levels of oxidative stress.
However a possibility surprised me. Caffeine itself may be a potential donor of methyl groups. I only took a cursory look but it might be worth interested parties considering the potentiality of methyl donation as an accompaniment caffeine metabolism as a possible explanation for how drinking coffee with caffeine (not decaf) may impact positively upon markers for inflammation.