This Simple Chart Explains the Devastating Influence Industry has on Scientific Integrity

Does Consumption of Sweetened Soft Drinks Lead to Population Weight Gain?

The scientific method has advanced humanity by leaps and bounds over the last 500 years. At its base is a pure and objective quest to better understand the world we live in and how it influences us. Scientists pose a hypothesis, then design experiments to test its validity. Data is collected and analyzed, and conclusions are reached. The information is published only after being scrutinized by peers. This is true in any field, be it physics, chemistry, or nutrition.

Based on scientific evidence accumulated from painstaking research, health professionals and individuals alike are able to reach conclusions that help them make decisions. In the world of nutrition, scientific advances have led, for example, to the understanding of micro-nutrients such as vitamins. We know what happens when we don’t get enough vitamin C (scurvy). We know what foods are a good source of vitamin C (bell peppers, kiwi, oranges…).

There is an implicit trust that people place in the work of scientists. It is based on hundreds of years of credibility built up through the scientific method. When new research is published that recommends consuming more of nutrient A, or less of nutrient K, we tend to accept it as a universal truth.

What happens when the science is tainted? Can we still rely on conclusions from a non-objective source?

This is exactly what a recent review, published in PLOS, wanted to find out. The name of this fascinating article is: Financial Conflicts of Interest and Reporting Bias Regarding the Association between Sugar-Sweetened Beverages and Weight Gain: A Systematic Review of Systematic Reviews

A systematic review is not a new experiment. It’s simply a lot of legwork sifting through many past experiments that have a common hypothesis, in order to identify a consensus, or lack thereof.

17 systematic reviews have been conducted in the past, to test the connection between caloric soft drinks (regular Coke, Pepsi, etc…) to weight gain. The current systematic review is a review of their conclusions. Ten of the previous reviews showed that sugary soft drink consumption may lead to weight gain, and 7 did not find such evidence.

The researchers divided these 17 reviews into two groups: those that were independent, and those that had some conflict of interest (usually funding by industry). The chart above illustrates that division. What you can clearly see is that industry funding led to a more favorable result to industry. The statistical significance is enormous, as is the practical significance to anyone who relies on “evidence based” advice from nutrition professionals.

What can be done?

In an ideal world, industry would never be allowed to directly fund research. But this would severely limit scientists, who have come to rely on donations and grants from corporations to push the boundaries of knowledge.

The authors of this study suggest that

clear guidelines and principles (for example, sponsors should sign contracts that state that they will not be involved in the interpretation of results) need to be established to avoid dangerous conflicts of interest.

Would that be enough? What do you think? How can science be funded if not through corporations?

  • EVIL food scientist

    It depends quite a bit on the questions you are asking in your studies and how you ask them. “Does adding liquid sugar to your diet without any other changes contribute to weight gain?” Sure. I don’t even have to do a study to figure out that one. Add calories in the form of easily metabolized sugar (driving your insulin levels all over the charts) to a diet that’s already at a maintenance level for calories and you are going to gain weight.
    The way you ask the question so you get a Coke/Pepsi acceptable answer is “Does adding suggested serving size (8 oz or so) amounts of liquid sugar in small quantities (2-3 servings a day) while removing the corresponding amount of calories from a maintenance level diet from a fit population with no insulin resistance contribute to the weight gain of that particular group over a short to moderate amount of time?” Probably not going to see a statistically significant amount of weight gain.

    • Danielle Robertson

      Right on the nose. Plus, as a food scientist I know how hard it is to get funding for your research. Clinical trials are VERY expensive and even with the clearest instructions there will always be a small percentage of non-compliance. Diet Logs and Food Frequency Questionnaires are great tools but they do have their limitations. It’s difficult to get enough funding to do a quality study that controls for enough variables and mitigates the non-compliance. Obviously people who have an interest in the results are more willing to fork over that money, but will the guidelines and principals above be enough? Maybe. Maybe we need to take a more critical look at how studies are performed to determine if the difference in results is due to study design or truly different data.

      • Jason Harrison

        I thought that diet logs have been shown to be highly inaccurate. Even if you only eat calibrated food (with known “Nutrition Facts” values) you’re relying on participants to accurately measure their consumption of these foods, accurately report exercise and thermal load, etc.

        • Danielle Robertson

          Sorry, I meant that diet logs ARE a good tool, as the shown by the paper “Validity and reliability of a new food frequency questionnaire
          compared to 24 h recalls and biochemical measurements: pilot phase of
          Golestan cohort study of esophageal cancer.” [European Journal of Clinical Nutrition]. I just meant that they are a great tool but even those have some limitations that we should always be mindful of when reviewing research. Here’s an example of what I mean:

  • Chef Mike in Burlington ON

    …and next, further industry funded scientific research that proves that A: water is wet B: ice is cold, C: Bill Clinton did not have sex with that woman…

    • Persnickety Pete

      Is water wet? Or does it make things wet?

  • Rob Grant

    Agree with the gist of the article and that it makes a very important point. The significance of the difference shown by the first chart is a bit better than 1% on a simple test of proportions though, with n=17, which maybe isn’t everyone’s idea of ‘enormous’ statistical significance?

  • Howard Lee Harkness

    I discovered this phenomenon about 12 years ago after reading over 100 “studies” in nutrition (see I pretty much quit reading studies after I saw a study published in a “peer-reviewed” journal which removed the sugar from the test group diet, and concluded from the numerous health improvements that “animal protein harmful for T2 diabetics.” I kid thee not.