Reading Beyond the Headlines

Posted by The Evidence Blog on April 17, 2012


by Jill Shuman, Managing Editor, News Service , Hayes, Inc. (follow on Twitter)

The British Medical Journal (BMJ) recently published a meta-analysis suggesting a link between increased white rice consumption and the risk of developing type 2 diabetes (T2D).

To examine the link between the consumption of white rice and the risk of T2D, the meta-analysis reviewed four observational studies describing seven cohorts of patients in China, Japan, the United States, and Australia. The studies, which ranged in length from 4 to 22 years, included a total of approximately 352,000 participants; none had diabetes at the start. Over the course of the studies, 13,284 people developed diabetes. An association between increased rice consumption and diabetes risk was found in both the Asian and the western countries, although the link was stronger among people in Asian countries. Given that the Chinese cohorts ate 4 servings of rice per day compared with less than 5 servings per week for the U.S. and Australian cohorts, this is not a surprising finding.

The meta-analysis had methodological issues, as well. Of the seven cohorts, two didn’t control for differences in dietary consumption patterns, one didn’t control for a family history of diabetes, six didn’t control for income/socioeconomic status, and none controlled for consumption of other refined grains or sugars—all key factors in confirming that any single dietary component causes a medical condition. In reality, many factors other than rice could have affected participants’ risk of developing T2D, including physical activity, alcohol intake, and obesity. In addition, the statistics cited as confirming the association between white rice and T2D in the western countries were not at all robust—with a confidence interval ranging from 0.94 to 1.33.

Sensing popular interest in the findings, BMJ issued a press release touting that “White Rice Increases the Risk of Diabetes.” International media outlets picked up the press release, with headlines such as “Eating White Rice Daily Ups Diabetes Risk,” “White Rice Could Cause Diabetes,” and “Think Twice About Rice Is New Study's Advice.” However, most of them neglected to include the (last!) paragraph in the press release citing the gist of the study’s accompanying editorial: “More, bigger studies are needed to substantiate the research hypothesis that white rice increases the chances of getting T2D.”

We should be uneasy about the BMJ press release and of the news sources that cavalierly picked up the press release without checking the facts. The study and the editorial were available to the media as the study broke—but that’s assuming that many of the media outlets have health and science reporters who understand study design and statistics, which we know is not the case.

What I found most troubling is that the journal, itself, questioned the validity of the meta-analysis. According to its accompanying editorial, “Although the findings of the current study are interesting they have few immediate implications for doctors, patients, or public health services and cannot support large scale action.” Yet, the journal still chose to issue a press release that indicted a relatively harmless food in the development of a chronic disease. While we all know that whole grains are a better choice than white rice, much of the world’s population survives on white rice without much choice in the matter.

Our Hayes News Service tries to cover journal articles that use evidence to truly inform clinical practice. We chose not to cover the BMJ story, for the inconsistencies noted above. However, as I’ve watched the poor media coverage of this story increase over the past 27 days, I’m thinking that perhaps we should rethink our strategy. Maybe it’s as important to report on the hype as it is to report on the evidence.

What do you think?

Topics: Hayes Blog

Sign up to receive updates from our blog

Our latest articles

New Call-to-action