Citations 0
Editor's Correspondence
January 22, 2007

The Potential Benefit of Coffee Consumption—Reply

Author Affiliations

Copyright 2007 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2007

Arch Intern Med. 2007;167(2):205. doi:10.1001/archinte.167.2.205-b

In reply

We appreciate the thoughtful comments from several readers on our recent article.1 Mascitelli and colleagues point to the possibility that phenolic compounds in coffee, particularly chlorogenic acid, may protect against the possible increased risk of diabetes due to high iron intake through its effect on reducing the absorption of dietary iron. We concur with this possibility. A previous analysis from our study cohort demonstrated a direct association between dietary heme-iron intake and/or supplemental iron with risk of type 2 diabetes mellitus, especially among alcohol drinkers.2 Future analyses carefully examining coffee and heme-iron intake in association with diabetes risk appear warranted. Peyrin-Biroulet and Roblin speculate that the homocysteine-raising effect of caffeine may explain the weaker association between regular coffee and diabetes than between decaffeinated coffee and diabetes. We are unable to directly evaluate this pathway because we did not measure homocysteine. However, a couple of lines of evidence dampen our enthusiasm. Our colleagues have recently published an article on coffee intake and a number of inflammatory diseases in this same cohort,3 finding an independent inverse association between coffee intake and total mortality as well as mortality from cardiovascular disease and a number of inflammatory-mediated diseases. These associations were not weaker for regular coffee than for decaffeinated coffee, thus providing no support for the caffeine-homocysteine pathway. In addition, findings from the Nurses Health Study and Health Professionals Follow-up Study are in conflict with our results, suggesting that caffeine may indeed protect against diabetes.4 Clearly, more studies are needed on this topic, including randomized trials. Finally, we agree with Cheng that it is premature to make public health recommendations to increase coffee consumption.

First Page Preview View Large
First page PDF preview
First page PDF preview