A review questioned sugar limits using evidence standards built for drug trials, but long-term diet guidance also relies on other study designs.
The debate over sugar guidance turns on a basic question: what counts as strong evidence when researchers can’t assign people to follow different diets for decades? A 2016 review argued that evidence supporting limits on added sugar was low quality. Its use of a grading method that favors randomized trials raises a broader issue: standards suited to testing medicines may not capture the evidence available for long-term nutrition questions.
The review was published by the International Life Sciences Institute, a Washington-based organization funded by multinational food and agricultural companies, including Coca-Cola and Monsanto. Its paper, titled “The Scientific Basis of Guideline Recommendations on Sugar Intake,” questioned the evidence behind recommendations to limit sugar intake. Applying the GRADE framework, the authors rated the overall evidence as low to very low. The organization and its funding have drawn criticism over possible conflicts of interest; coverage of the review and its funders appeared in FoodNavigator and The New York Times.
Why trial standards can be difficult to apply to diet
GRADE, short for Grading of Recommendations Assessment, Development and Evaluation, was developed to make recommendations more transparent and evidence-based. Its approach gives particular weight to randomized controlled trials. That emphasis makes sense for many treatments: researchers can assign participants to receive a medicine or a comparison treatment, then measure what happens. Yet the design becomes harder to apply when the question concerns ordinary eating over many years.
Diet studies can’t usually conceal which eating pattern someone is following, and there’s no simple placebo diet. Participants may also find it difficult to maintain an assigned diet for long periods. Randomized trials are costly, so they often run for weeks or months. That timescale can be too short to assess outcomes that may take years to emerge, such as some cancers. These limits don’t make trials unhelpful; they mean that trials alone may not answer every question about long-term eating patterns.
Long-term observational research can follow groups with different habits and compare their health over time. Such studies can’t establish cause and effect as cleanly as a well-run trial, because other differences between groups may help explain the results. But they can address questions that researchers can’t realistically test by assigning people to lifelong diets. The source article compares this challenge with research on smoking: people can be followed over time, while assigning them to smoke for decades would be neither practical nor ethical.
The same general limitation applies to some lifestyle guidance. Research on sudden infant death syndrome, or SIDS, found an association between stomach sleeping and risk. The article describes how advice to place infants on their backs was followed by a decline in SIDS rates. Testing that question by randomly assigning large numbers of babies to sleep positions and tracking deaths would raise serious ethical concerns. The example illustrates why evidence from different study designs can matter when trials aren’t suitable.
What the sugar review does and doesn’t establish
A low rating under a particular evidence framework is not, by itself, proof that a recommendation is wrong. It reflects how the available research fits the framework’s criteria. For nutrition questions, that assessment needs to account for the limits of randomized trials as well as the strengths and weaknesses of observational evidence. A lifetime dietary pattern is difficult to study with the same methods used to test a short-term drug treatment.
That distinction matters when reading claims that nutrition recommendations lack evidence because they aren’t supported by long-term randomized trials. The source article argues that such criticism can set an unrealistic standard if it dismisses other relevant research by design. It also draws a comparison with tobacco companies’ efforts to challenge evidence about health harms by promoting a narrow definition of acceptable proof. Historical accounts describe Philip Morris working with public relations firms and lawyers on a “sound science” effort that sought to shape standards of scientific evidence. The article points to research on that effort here and here.
Funding and institutional ties are relevant context to examine, but they don’t settle whether a study’s methods are sound. Readers and researchers need to assess the question asked, the evidence included, the grading criteria and the authors’ potential conflicts together.
Randomized trials remain valuable, especially when researchers can ethically and practically assign treatments. But they’re one part of the evidence base. As Sir Austin Bradford Hill, a pioneer of randomized trials, cautioned, treating controlled trials as the only acceptable route to knowledge would take the idea too far. His point remains useful here: the best way to assess a health question depends on what can be studied, over what period, and with what ethical limits.
What to watch in nutrition guidance
For people evaluating advice about added sugar, the practical takeaway is to look beyond a single evidence grade. Ask what types of studies were reviewed, whether the research addresses long-term outcomes, and how the authors handled uncertainty. Recommendations are strongest when they reflect the full body of relevant evidence and explain where that evidence remains limited.
This account is part of an eight-part series on industry influence over dietary and health guidance. The earlier installment, “The Corporate Playbook behind Undermining Dietary Guidelines,” discusses the wider context. Related material includes “How Big Sugar Manipulated the Science for Dietary Guidelines”, a discussion of hormone therapy trials in “How Did Doctors Not Know About the Risks of Hormone Therapy?”, and further accounts of industry influence in “Sugar Industry Attempts to Manipulate the Science” and “Big Food Using the Tobacco Industry Playbook.” The central question for future reviews is whether their methods fit the long timescales and practical limits of nutrition research.
