| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
LETTER TO THE EDITOR |
The Research Institute of Public Health
School of Public Health and Clinical Nutrition
University of Kuopio
PO Box 1627
70211 Kuopio
Finland
E-mail: tomipekka.tuomainen{at}uku.fi
Dear Sir:
We read with great interest the recent meta-analysis, by Barclay et al (1), of glycemic index (GI), glycemic load (GL), and the risk of chronic disease. The authors made a respectable effort in covering a multitude of outcomes, which—while sparing a good deal of time for a busy reader—resulted in a lack of depth in the dissection.
In our understanding, the very purpose of a meta-analysis is to arrive at a judgment of an association from analyzing a large base of evidence, which usually means several original studies—surely more than a single study, and preferably >2 studies. However, the analysis by Barclay et al of the association of GI and GL with stroke risk, for example, is based on a single study by Oh et al (2). A quick look at Figure 3 (on page 635 of the meta-analysis) suggests that several studies have been summed up, but a closer look reveals that the first point estimate is for lean subjects and the second is for overweight subjects from the study by Oh et al, and that the last 2 figures give fixed- and random-effects estimates based on the first 2 figures.
Heart disease (also in Figure 3) does not fare much better: only 2 studies are referenced (3, 4). Furthermore, although the label "heart disease" is a good approximation of coronary heart disease (3, 4), the label "eye disease" is far too wide a term, as compared with the disease actually studied by Schaumberg et al (5), which was age-related cataract, also evaluated from a single study.
Because there is vigorous discussion, also in the general public, about the possible health effects of GI- and GL-modified diets and about the possible risk effects of a past high-GI or high-GL diet (or both), one should be careful in drawing firm conclusions. This is especially true if the evidence is insufficient and if—as in this case—the risk factor itself is still vague.
Although it is reasonably tempting to draw the big picture, diseases are not equal, and not all of them necessarily have a common background. Therefore, a message such as that given on page 634—"This meta-analysis provides high-level evidence that diets with a high GI, high GL, or both ... increase the risk of chronic lifestyle-related diseases."—is perhaps an oversimplification.
ACKNOWLEDGMENTS
None of the authors had a personal or financial conflict of interest.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
