On Nutrition

Odds are you’ve caught wind of a new Harvard study suggesting that the more sugar-sweetened beverages (SSBs) you drink, the greater your risk of dying prematurely from a chronic disease, especially cardiovascular disease. Certainly, SSBs are the single largest source of added sugar in the American diet, and unlike the natural sugars found in plant and dairy foods, added sugars tend to come in a “package” that’s high in calories and low in nutrients. But the headline-grabbing Harvard study doesn’t shed as much light as you might think.

SSBs include soda, energy drinks, sports drinks, and other beverages that are sweetened with sugar (including “natural cane sugar”), high fructose corn syrup, or other calorie-containing sweeteners. This also includes “fruit drinks,” but not 100% fruit juice. While there are good reasons to avoid these beverages — or at least limit how much you drink — the reasons can’t be summed up by a few statistics.

The problem with observational studies

The sugar sweetened-beverages study, published in March in the journal Circulation, is one of many on associations between diet and lifestyle habits and the onset of various disease in participants of the long-running Nurses Health Study and the Health Professionals Follow-Up Study.

The results suggest that if you drink one or fewer SSB per week, your risk of dying from cardiovascular disease increases by 6%; it goes up by 31% if you drink two or more. This means if you have a 20% risk without drinking any SSBs, your risk would increase to 21.2% with a one-SSB-per-week habit, or 26.2% if you were in the habit of drinking two or more SSBs per day.

“Do these beverages increase the risk of mortality and cardiovascular disease? That’s a good question,” said Mario Kratz, Ph.D., a faculty member and researcher at Fred Hutchinson Cancer Research Center and the University of Washington. However, he said because the Harvard study is observational, it can’t actually answer the question. “The overarching problem is that there’s overwhelming evidence that most of the observations we see in purely observational studies based on food frequency questionnaires are not reproducible by randomized controlled trials.” He said that the scientific community is becoming increasingly critical of observational studies being used as anything other than a generator of questions to be explored more fully through randomized trials.

“In observational studies we have the issue of confounding, or confusion of effect. Thus, if you find an association between something like sugar-sweetened beverages and cancer, you can’t be sure that this association actually reflects a cause-effect relationship,” Kratz said. What else do high SSB drinkers eat? What lifestyle habits do they have? This is something that the Harvard researchers tried to account for, in order to isolate the specific role of SSBs. Trouble is, Kratz said that many foods aren’t measured accurately when people are asked to remember how much and how often they eat them, and while major lifestyle habits like exercise and smoking might be measured reasonably well, it’s not possible to measure all variables that might affect something like risk of premature death. “It would seem very likely to me that in the U.S. population, those who drink the most sugar-sweetened beverages have very different dietary and lifestyle habits than those who drink one only occasionally,” he said.

Advertising

What do randomized controlled trials tell us?

For a more accurate picture of the connection between SSBs and health, it’s better to look at well-controlled diet-intervention studies. Kratz said there’s conclusive research showing that drinking sugar-sweetened beverages doesn’t trigger an adequate satiety response. In other words, calories from a soda or sports drink won’t fill you up as much as an equivalent number of calories from food would. “Calorie intake tends to go up substantially when people consume sugar-sweetened beverages.”

In one 2015 study from Kratz’s lab, participants were given three or four SSBs per day — equal to about 25% of their estimated calorie needs — to drink for eight days, and they only partially compensated for the added sugar by eating less food. “If you ask individuals to consume 400 calories in the form of sugar-sweetened beverages, they may reduce food calories by 100 calories, but not by 400. Sugar-sweetened beverages contribute to excessive calories,” says Kratz.

Kratz also points to evidence that sugar in beverage form has unique metabolic effects, contributing to fatty liver, dyslipidemia — aka high cholesterol and/or triglycerides — and insulin resistance. He said that because part of the sugar in SSBs is fructose, it tends to increase fat in the liver, especially in liquid form. “If you have more fat in the liver, you make more VLDL. That’s a very clear path through which sugar in liquid form could contribute to dyslipidemia — that’s even worse if you consume excess calories.” VLDL, or very low-density lipoproteins, is a precursor to LDL cholesterol, which is commonly seen as a major risk factor for cardiovascular disease. If high SSB consumption increases VLDL, this could certainly lead to cardiovascular disease and other health problems that might cause premature death, but that’s not something you can peg a specific number on.

What about artificially sweetened beverages?

Some former artificially sweetened beverage (ASB) drinkers have switched back to SSBs because of the belief that sugar is more “natural” and the fear that artificial sweeteners pose a risk to health. Kratz said randomized controlled trials do suggest that ASBs may have some negative effects on the gut microbiota, but “whether this has negative health consequences is not clear at this point. For sugar-sweetened beverages, we can be quite confident that they will have negative health effects, particularly when consumed regularly and in large quantities.” So while water is the best beverage for staying hydrated, if you are a devoted drinker of soda or other sweetened beverages, ASBs may be the more healthful choice in the end.