In recent months, the question has sparked harsh words, flared tempers, op-eds, demonization, and even talk of creating policy that would ban or limit consumption of sugar-sweetened beverages like sodas. But what does the scientific evidence really show once emotion and exaggeration are removed from the equation?
With the intent of setting the scientific record straight, respected academics and nutrition researchers came together to discuss the issues in Boston on Tuesday, April 23, at Experimental Biology 2013. The symposium, organized by the American Society for Nutrition, would be the second of its kind over two years at the conference to evaluate sugar and how it relates to health. The event was sponsored by the Corn Refiner’s Association (CRA) and endorsed by the Medical Nutrition Council.
Exaggerated and “emotion-raising” statements
Biostatistician David Allison, MD, a professor at University of Alabama at Birmingham, addressed how misinformation in the peer-reviewed literature and the media about sugar and sugary drinks has spread confusing messages to the public. He called out several examples of how studies with an epidemiological design have reported their findings in exaggerated ways. Many of these trials have found only ecological correlations linking sugary drink consumption to rising obesity numbers. Correlations are the weakest form of evidence and don’t establish causation. There are also correlations linking the rise in bottled water consumption and obesity, for example, he said.
However, despite discovering only weak associations in their data along with inconsistencies, many of these studies are titled in ways that inflate their findings and they use language that suggests causality. Examples he gave include “Sugary beverages represent a threat to global health” (Popkin),”Public health: The toxic truth about sugar (Lustig), and “Fructose: pure, white, and deadly. Fructose, by any other name, is a health hazard.” (Bray). These examples have “emotion-raising” language that leads to further exaggerations and inaccuracies in press releases and in articles by journalists, Dr. Allison said. The most recent example of an article with exaggerated statements, Dr. Allison told me after the event, is one from Bloomberg entitled “Daily Soft Drinks Raise Diabetes Risk by 22%, Study Shows”.
“There’s no surprise that people are confused,” Dr. Allison said. “Imagine you’re a newspaper reporter with no scientific training reading these titles trying to write an article that will inform regulators and the public about sugar-sweetened beverages. You think cause and effect has been demonstrated when in fact it’s only a correlation.”
In addition, as there is a tendency for studies sponsored by industry to show bias, Dr. Allison reminded that there also exists “white hat bias”—a systemic bias among academic scientists that leads to exaggerating findings from their data. “When you look at the peer-reviewed literature, there are multiple demonstrable biases and inappropriate statements that serve to exaggerate,” he said.
The standard for acting or creating policies is subjective and depends on the situation and one’s values and judgments, he said. In contrast, the standard for drawing a conclusion about causation requires randomized controlled trials. Are there any? Yes, Dr. Allison said there do exist several randomized controlled trials that evaluate sugar-sweetened beverages and changes in body weight. Some do show, he concedes, that within the context of additional calories sugary drinks may lead to weight gain.
Dr. Allison makes clear he’s not “advocating” sugary drinks. He agrees that it makes common sense for individuals wishing to lose weight to limit their intake of sugary drinks and other nonessential kilocalories. However, he maintains that the current evidence testing whether reducing sugary drink consumption reduces body weight in consumers has shown only equivocal effects. “As scientists and scholars, we can and should hold ourselves to higher standards,” he said.
Solids versus liquids
Purdue University professor of nutrition science Richard Mattes, Ph.D., believes that sugary drinks do lend to increased risk of weight gain in ways beyond their contribution of calories. He’s not so much concerned with the sugar content of sugar-sweetened beverages; the problem, he said, is the difference in the way those calories are consumed—as liquids versus solids. “When we consume beverages, our bodies are incredibly quick at processing and excreting them. A calorie is not a calorie anymore.”
Citing research from his own lab and that of others, Mattes explained that the evidence suggests that the body’s sensations in response to clear beverages are different in comparison to solid food. For example, when hungry our bodies sense that they want to eat food until full, but our bodies tend to be relatively thirsty all day long. There is a strong “stop” signal and a weak “go” signal when eating solid foods, he suggested. In the case of clear beverages, the opposite is true: there is a weak “stop” signal, and a strong “go” signal.
These sensory signal differences of clear beverages also don’t appear to apply with viscous energy-dense beverages such as soups or liquid meal-replacements. The reason may due to the pattern in which viscous energy-dense beverages are consumed. They are usually consumed in a regimented way, such as at a specific time of day. The dietary pattern that people use to consume clear beverages, on the other hand, is often as frequent as snacking. It’s not because people are eating more food, it’s the change in the way they’re eating their food, Mattes explained.
Frequent sugary drink consumption and the rise of obesity and diabetes in the U.S. has nutritional biologist Kimber Stanhope, Ph.D., an associate project scientist at UC Davis, alarmed despite lack of sufficient evidence establishing causation. She said that while adequately powered clinical trials and longer interventions are needed, she questions why it would be necessary to avoid acting now. “Do we need to wait for these results before we revise the dietary guidelines and start educating the public accordingly?” she asked.
In randomized controlled trials that Stanhope’s lab has produced, sugars—whether as table sugar, high-fructose corn syrup, glucose, and fructose—all comparably increase body fat in subjects when consumed in excess. She also presented data showing that pure glucose versus pure fructose consumed in excess may behave differently in the body. Pure glucose, for example, led to increases of subcutaneous adipose tissue (fat just under the skin) while fructose led to increases of visceral adipose tissue (fat around the organs), which may contribute to increased insulin resistance and cardiovascular risk. Notably, her research also shows that when comparing high-fructose corn syrup and table sugar, these don’t show changes related to body composition suggesting that the metabolism of glucose and fructose behaves differently when consumed together.
Sugar on the brain and addiction
Stanhope also pointed out that glucose and fructose may have different affects on the brain. For example, she cited a recent study published in JAMA (Page et al) that showed glucose consumption promoted appetite suppression in the brain whereas fructose did not. The study has been a source of much speculation regarding whether or not the fructose moiety of sugar and high-fructose corn syrup could be addictive versus other types of carbohydrates.
Nutritional neurologist Miguel Alonso, MD, believes that functional MRIs (fMRIs) as used in the JAMA study and others can indeed be useful for studying brain regions that are involved in decision making and reward aspects of both drugs and food. He also finds interest in the clinical overlap between obesity and drug addiction. For example, there is evidence of a shared vulnerability for the two because of genetic predisposition.
However, Dr. Alonso said that the notion that sugar or any other particular food component is addictive is highly speculative and based on weak and limited evidence. Addiction is highly complex and generally defined by several criteria in an individual that involve tolerance, withdrawal symptoms, lack of control, and interference with daily activities, he said. He also questions what findings can be drawn from fMRI studies that compare fructose versus glucose.
“There is limited and insufficient data in humans on the link between food, obesity and addiction. Brain overlaps do exist, but that does not equal addiction,” Dr. Alonso said. “Recent fMRI data suggest neuroadaptation following repeated intake of palatable food, including sweetened beverages. There is a need to clarify the link, particularly experimental interventions.
Experimental studies are needed to clarify associations found in cross-sectional studies related to the brain, he said. A study by Burger and Stice, for example, found an association between frequent consumption of ice cream and a reduction in sensitivity in regions of the brain related to food reward. These studies lead to questions about how long it takes for neuroadaptations to occur, how specific these changes are, and how they contribute to eating patterns under normal conditions. Dr. Alonso said that while some recent fMRI data suggest different acute brain responses in response to fructose versus glucose, there is still need for research that includes normal conditions of intake and behavioral correlates to better interpret the findings.
Cardiologist Jame Rippe, M.D., founder and director of the Rippe Lifestyle Institute and professor of biomedical sciences at the University of Central Florida, whose organization has participated in funding Dr. Alonso’s pilot research, said that new studies using tools such as functional MRI to study the brain were lending interesting results. But he said that it’s not surprising that areas of “the epicenter of where decisions come to pass” would “light up,” or become active, in response to sugar or any other carbohydrate-containing foods. “It makes all the sense in the world, since the brain relies on glucose,” he said.
However, he said that the speculation about how different sugars—high-fructose corn syrup, sucrose, glucose or fructose—are handled by the brain and talk of addiction feels like, as Yogi Berra once said, “déjà vu all over again.” He reminded of similar arguments he’s made before in previous years as a CRA consultant. It has been a challenge, he said, to educate consumers that the way high-fructose corn syrup and table sugar (sucrose) are treated in the body is the same metabolically, as supported by several randomized controlled trials. And, it’s also been a challenge to help people understand that fructose and glucose are “always consumed together” in a normal diet. One cannot make too much of studies using high amounts of pure glucose or pure fructose whether it be in animals or humans. When consumed together, their absorption and metabolism are different.
What about dose?
Dr. Rippe added that the way sugars are treated in the body has been well studied and regulators and health organizations should take note of that. And he praised the scientists at the event and ASN for coming together to speak out against inaccuracies reported in the peer-reviewed literature and the media.
In addition, Dr. Rippe took issue with the American Heart Association (AHA) for coming out with a statement recommending that sugar be limited to only 150 Kcal per day for men and 100 Kcal for women. The recommendation is only about a third of what has been set by the Institute of Medicine in current Dietary Guidelines for Americans. The IOM dictates, yet does not recommend, that up to 25 percent of kilocalories (Kcal) can be consumed safely from added sugars.
“The AHA has acknowledged that their recommendation wasn’t based on any definitive proof, just their thoughts,” Dr. Rippe said. He added that the IOM recommendations were based on actual evidence from the current literature.
But he said his lab “put the AHA recommendations to the test” anyway. Rippe’s lab conducted a trial that randomized subjects to diets consisting of either 8, 18, or 30 percent of total Kcals from added sugars. Despite consuming roughly four times more than AHA recommendations, the subjects in the group consuming 30 percent added sugars did not show any difference in cardiovascular risk factors as compared to the other groups (including blood lipids, leptin, ghrelin, or fasting insulin). The evidence suggests there are no dangers in consuming sugars up to 25 percent of Kcals within the context of an energy-balanced diet. Some will point out that the study was funded by the CRA and should not be trusted, but Dr. Rippe’s results are corroborated by other randomized controlled trials in the literature.
After the symposium, I spoke to Friedman School professor of nutrition Edward Saltzman, who chaired the event, who said to me that his reaction to the meeting could be summarized as, “I believed every speaker.” That poses a problem, he added, because of from a distance it’s easy to understand why the general public remains more confused due to conflicting evidence on sugar-sweetened beverages. “The issue is far from resolved,” Saltzman said.
Making the medicine to go down
The day after the event, I spoke to professor of nutrition and metabolism David Jenkins, M.D., who told me that he thought more attention should be given to the work of nutrition researcher John Sievenpiper, M.D., of St. Michael’s Hospital, University of Toronto. He said Dr. Sievenpiper “has an army of researchers combing the literature” on sugar and its effects on health.
Dr. Sievenpiper is the lead author of several meta-analyses (of which Dr. Jenkins is often listed as a co-author) comparing sugars on different parameters related to metabolism and body weight. As discussed in my prior interview with Dr. Sievenpiper, these meta-analyses found no differences in the effects of fructose, glucose, or other carbohydrates on several parameters on body weight, blood lipids, and blood pressure. In addition, Dr. Sievenpiper’s research suggests a marked benefit from fructose for glycemic control when consumed in amounts normally found in fruit.
Dr. Jenkins said that he did agree with the thrust of some researchers—such as Walter Willet, M.D., and David Ludwig, M.D.—to recommend that the public reduce consumption of sugary drinks in general when they are identified as providing excessive calories in people’s diets. However, the Atwater Lectureship recipient disagrees with attempts to force reduced consumption using warning statements or regulations.
Most well recognized for developing the low-glycemic index concept, Dr. Jenkins also suggested that sugars be used as put by Mary Poppins, “to make the medicine go down.” More to the point, he said people should use sugar strategically in combination with nutrients that would otherwise “taste like cardboard,” like viscous fiber. Present in several vegetables and whole-grain foods, viscous fiber is not often thought of as palatable, however, it is shown in studies to slow down the rate in which sugar and other carbohydrates are absorbed, reduce total and LDL cholesterol, as well as increase satiety (the feeling of fullness). “If I could get people to eat ‘cardboard’ by itself I would, but put a little sugar on it, a little salt, or a little fat, and people will eat it,” he said.
Whether Dr. Jenkins would approve of washing down steel cut oats, asparagus, or Brussels sprouts with a bottle of Mountain Dew or any other brand of soft drink, he didn’t exactly say. But it may be worth a try for those who have a sweet tooth and who wouldn’t include these highly beneficial foods in their diets otherwise.
Update – 5-17-13 – The videos from the ASN session have just become available, so here they all are for your enjoyment (by order of appearance):