Chocolate And Cholesterol - At Some Point It's Okay To Call B.S.
    By Hank Campbell | April 21st 2008 12:19 PM | 5 comments | Print | E-mail | Track Comments
    About Hank

    I'm the founder of Science 2.0®.

    A wise man once said Darwin had the greatest idea anyone ever had. Others may prefer Newton or Archimedes...

    View Hank's Profile
    John Erdman, a University of Illinois professor of food science and human nutrition who also chairs the Mars, Inc. Scientific Advisory Council and has received millions in funding from Mars, Inc., recognizes that taking money from a candy bar company (Mars Inc.) to do a study of their (Mars Inc.) candy bar proving it is healthy will have skeptics.

    Not here. Hey, if Philip-Morris wants to highlight a study saying cigarettes cure cancer or Exxon-Mobil needs to promote a study saying automobile carbon monoxide improves asthma, we won't ridicule them just because of the funding. We'll ridicule them because of the methodology.

    “Eating two CocoaVia dark chocolate bars a day not only lowered cholesterol, it had the unexpected effect of also lowering systolic blood pressure,” said Erdman on the results of a peer-reviewed study in The Journal of Nutrition.

    Except the participants were also put on the American Heart Association’s “Eating Plan for Healthy Americans” (the Step 1 diet) two weeks before the study started.

    “After the participants started the AHA diet, a lot of them began to lose weight, so we had to keep fussing at them to eat more. We didn’t want a weight change because that also lowers cholesterol,” said Ellen Evans, a University of Illinois professor of kinesiology and community health and co-author of the study.

    Then why put them on a diet at all? They found 49 people with slightly elevated cholesterol and normal blood pressure, chose some of them for the double-blind, placebo-controlled, cross-over study, divided the participants into two matched groups and gave them one of two types of CocoaVia (© Mars, Inc.), one with plant sterols and one without, and declared it a victory for candy when cholesterol dropped.

    How did they control for the diet? Well, they didn't.

    “I know that it was a double-blinded trial that wasn’t skewed toward a particular result,” said Erdman, who chairs the Mars Scientific Advisory Council. “Moreover, the paper was peer-reviewed and published in the Journal of Nutrition, which ranks in the top 10 percent of all the biological science journals.”

    The trial didn't have to be skewed toward a particular result in order to cause a predictable one. Taking a group of people with high cholesterol and giving them a candy bar did not lower cholesterol. Taking those people and putting them on a healthy diet did.

    Would their cholesterol have dropped with just the diet? Absolutely. That's why the AHA recommends it. The drop is in spite of the chocolate, not a result.

    Erdman attributes the drop in cholesterol numbers (total cholesterol by 2 percent and LDL or “bad” cholesterol by 5.3 percent) to the plant sterols that have been added to the bar and the drop in blood pressure to the flavanols found in dark chocolate. Again, why give credit to these new plant sterols added into a chocolate bar Mars, Inc. wants to sell rather than saying something more clear like, 'if you go on a healthy diet, you can still enjoy some chocolate'?

    I have no problem with funding sources. Corporate funding does a world of good and I have never dismissed studies that dispute global warming, for example, just because of their funding, but it gets to a point when you have the same people saying the same thing over and over that you have to be a little skeptical. Mars has spent millions of dollars studying the beneficial aspects of flavanols found in cocoa beans and learning how to retain their benefits during the refining process. Kudos to them for that because at least our tax dollars are not paying for it.

    And to the credit of their candy bar, people who ate the formulation with the sterol-containing products had lower cholesterol than those who ate the chocolate without it, but that wouldn't get headlines and generate sales of a new product like 'chocolate lowers cholesterol' will.

    Heck, Kuna Indians in Panama eat chocolate every day and are terrifically healthy. Why not give chocolate the credit and ignore their terrifically active lifestyles and say chocolate can replace going to the gym?

    Healthier chocolate is still chocolate. It's not inherently good for you.

    I would never say that funding makes people dishonest but I will say that companies are not funding research they think will make them look bad.

    Merrill Goozner, director of the integrity in science project at the Center for Science in the Public Interest, co-authored a paper called Relationship between Funding Source and Conclusion among Nutrition-Related Scientific Articles and found that studies funded by industry were four to eight times as likely to reach conclusions in the financial interests of sponsors.

    One of the other co-authors of this Journal of Nutrition study, Catherine Kwik-Uribe, happens to be research manager of Mars, Inc. That's pretty incestuous.

    What about the peer reviewed aspects? What does it really mean?

    In last year's "Neurobiology Of Chocolate" panel at the AAAS meeting, only one of the researchers was not funded by Mars - and this was an AAAS meeting. One of the panel's organizers was Harold Schmitz, a visiting professor at UC Davis and the Chief Science Officer of Mars, Inc.

    Carl Keen, chairman of the UC Davis nutrition department, has seen his group be the recipient of $10 million in research from Mars just since 1997 and the chair he holds was created by Mars.

    So when Erdman says the study was peer-reviewed, a lot of our calibrating the results and the review depends on just who the reviewers were and their relationship to Mars.

    Again, Mars is not buying scientists off. I would never allege such a thing. But it should not be a surprise that they prefer to fund research they like.

    So eat some chocolate in moderation. It won't kill you. Maybe their new dark chocolate bar with additives will help with cholesterol, but will it help as well as a healthy diet alone? No, it won't.

    If you need an excuse to eat more chocolate or to buy a new candy bar from Mars, Inc., this study is perfect for you. But if someone in your family (you know, people you care about) wants to use these results to justify eating junk food, it's okay to call B.S.

    Study: Robin R. Allen, LeaAnn Carson, Catherine Kwik-Uribe, Ellen M. Evans and John W. Erdman, Jr, Daily Consumption of a Dark Chocolate Containing Flavanols and Added Sterol Esters Affects Cardiovascular Risk Factors in a Normotensive Population with Elevated Cholesterol, American Society for Nutrition J. Nutr. 138:725-731, April 2008


    Seth Roberts

    "Would their cholesterol have dropped with just the diet? Absolutely." You're probably wrong. There were two groups: One group got chocolate with plant sterols; the other group got chocolate without plant sterols. The group that got the chocolate without plant sterols did not have lower cholesterol at the end of the experiment than at the beginning of the experiment. Both groups got the "healthy diet".

    "Why give credit to these new plant sterols?" Because the 2% reduction in total cholesterol was a between-group difference-- between a group that got chocolate with plant sterols and a group that got chocolate without plant sterols.

    If what you are arguing is true - basically that the candy bar with the additive made the difference - then why put them on a diet at all? It's not like these were all people of the same gender, age and body type where the other aspects of the diet were an additional control.

    I did state that the people who ate the chocolate with the additive had lower cholesterol than those who ate the chocolate without but it will be a tough sell defending (1) that all of these people would not have been healthier without any chocolate at all on a healthy diet.

    (1) Based on this study, anyway. But I was only talking about this study. Mars Inc. and Warren Buffett are also now buying Wrigley's. Look for 'chewing gum lowers cholesterol' studies in 2010 I can ridicule. :)

    T Ryan Gregory
    I haven't read the study, but based on your description it appears the controls were incomplete. If you are going to introduce a third variable (diet) then you need a third group (diet with no candy bars). Maybe chocolate makes it *worse*, but less so with the sterols present. The diet alone may have been the most significant contributor such that chocolate only impedes this effect to varying degrees depending on the type of bar (i.e., chocolate is either bad or very bad).
    Seth Roberts
    "Why put them on a diet at all?" Several possible reasons: 1. To make the study more attractive to possible subjects. 2. To make the study more attractive to the ethics board that must approve it. 3. Because it is more impressive to show benefit on top of a healthy diet than on top of an average diet. 4. To reduce dietary variation between subjects and groups.
    T Ryan Gregory
    Because it is more impressive to show benefit on top of a healthy diet than on top of an average diet.
    They showed no such thing, as I said above.