Added sugar and ultra-processed foods: what the evidence shows

Key points#

• WHO and the US Dietary Guidelines both put free or added sugars under 10% of daily calories, about 50 g on a 2,000-calorie day, and WHO adds a conditional benefit below 5%.
• The clearest sugar-specific signal is for sugar-sweetened beverages: a 2010 meta-analysis linked the highest intake to roughly 26% higher type 2 diabetes risk, plus higher metabolic syndrome risk.
• A tightly controlled 2019 feeding trial found adults ate about 500 more calories a day on an ultra-processed diet than on a matched unprocessed one, even with sugar, fat, sodium, and fiber matched.
• A 2024 umbrella review links higher ultra-processed intake to several cardiometabolic, mental-health, and mortality outcomes, but the underlying studies are observational and often low certainty, so they cannot prove cause.
• For most people the highest-yield change is swapping sugar-sweetened drinks for water or unsweetened options, and choosing whole fruit over juice.
• "Ultra-processed" is a broad Nova category and a prompt to look at the whole pattern, not a verdict on every packaged food.

What the evidence shows about added sugar and ultra-processed foods#

On added sugar and ultra-processed foods, the strongest evidence points in a simple direction. Cutting back on sugar-sweetened drinks has the most consistent link to weight gain and type 2 diabetes, so it is the single change most worth making first. The wider case against ultra-processed foods is real but softer: higher intake tracks with worse health in large observational studies, and one tightly controlled trial found people ate about 500 more calories a day on an ultra-processed diet, yet most of that evidence cannot prove cause on its own.

None of this means sugar is poison or that every packaged food is off-limits. WHO and the US Dietary Guidelines both suggest keeping free or added sugars under 10% of daily calories, about 50 grams (roughly 12 teaspoons) on a 2,000-calorie day. It helps to treat "ultra-processed" as a prompt to look at your overall eating pattern rather than a verdict on any one product. The rest of this post covers what the guidelines say, where the evidence is strong, and the swaps with the best return on effort.

What we mean by added sugar, free sugars, and ultra-processed foods#

Total sugar on a label counts everything: the lactose in milk, the fructose in an apple, and the cane sugar stirred into a soda. Added sugar is the narrower slice a manufacturer or cook puts in, meaning syrups, table sugar, and the sugars in concentrated juice. The World Health Organization uses a slightly wider term, free sugars, which folds in added sugars plus the sugars naturally present in honey, syrups, and fruit juice. The distinction matters because a glass of orange juice and a whole orange are not the same metabolic event.

Naturally occurring sugars in whole fruit, vegetables, and plain dairy sit in a different bucket. They arrive with fiber, water, and micronutrients, which slow how fast the sugar reaches your bloodstream and how much you tend to eat. That is why the guidelines aim at free or added sugars, not at the sugar in an apple.

Ultra-processed foods are a separate idea, drawn from the Nova classification, which sorts foods by how much industrial processing they have had rather than by their nutrient content. Nova's fourth group covers formulations built largely from substances extracted from foods (sugars, oils, protein isolates) plus additives you would not keep in a home kitchen. Packaged snacks, soft drinks, mass-produced bread, and reconstituted meats often land here. It is a broad, imperfect category: a shelf-stable bean soup and a candy bar can share the label, which is one reason to read it as a starting question rather than a final answer.

What the guidelines actually recommend#

Two well-known bodies land in the same place. WHO's 2015 guideline makes a strong recommendation to keep free sugars under 10% of total energy across the life course, and adds a conditional recommendation to go under 5% for further benefit, framed mostly around dental health. The Dietary Guidelines for Americans, 2020 to 2025, recommend keeping added sugars under 10% of daily calories.

On a 2,000-calorie day, 10% is about 200 calories from sugar, which works out to roughly 50 grams, or around 12 teaspoons. A single 12-ounce can of regular soda can hold 35 to 40 grams on its own, so that budget is easier to spend than most people expect. Worth saying plainly: these are thresholds to move toward, not a demand for zero. Nobody needs a perfect record, and an occasional dessert does not undo a reasonable week.

The sugar evidence: where it is strongest#

The sturdiest sugar-specific finding is about what you drink. A 2010 meta-analysis in *Diabetes Care* (Malik and colleagues) pooled prospective cohorts and found that people in the highest category of sugar-sweetened beverage intake had about a 26% higher risk of developing type 2 diabetes than those in the lowest, along with a higher risk of metabolic syndrome.

Liquid sugar seems to behave differently from the same grams eaten in solid food. You can drink 40 grams in under a minute, it does little to blunt appetite, and the calories tend to stack on top of everything else rather than replace it. Whole fruit is not the concern here. The fiber and the act of chewing that come with an apple change both the pace and the total.

Ultra-processed foods: the controlled and observational evidence#

The most direct evidence on processing itself comes from a small inpatient trial. In 2019, Hall and colleagues (*Cell Metabolism*) admitted 20 adults to a metabolic ward and rotated each person through two weeks of an ultra-processed diet and two weeks of an unprocessed one, in random order. The two menus were matched for calories offered, sugar, fat, sodium, and fiber, and participants could eat as much or as little as they wanted. On the ultra-processed arm they ate about 500 more calories per day and gained roughly a kilogram; on the unprocessed arm they lost about the same. The design is a real strength: same people, controlled kitchen, matched macronutrients. The limits are just as real: 20 participants, four weeks, one research setting. That makes it a signal to take seriously, not a final verdict.

Zoom out to the population data and the picture gets broader but softer. A 2024 umbrella review in the *BMJ* (Lane and colleagues) gathered dozens of meta-analyses and reported associations between higher ultra-processed intake and a range of outcomes, including cardiovascular disease, type 2 diabetes, depression, and all-cause mortality. The honest caveat sits inside the same paper: most of the pooled evidence is observational, and much of it was graded low or very low certainty. Association is not causation, and people who eat the most ultra-processed food tend to differ in many other ways that are hard to account for fully.

Why processing may matter beyond sugar alone#

Why might processing matter on top of sugar? Several mechanisms are plausible, and all are still being studied. Energy density is one, since many ultra-processed foods pack more calories into each bite. Texture is another: soft, easy-to-chew formulations tend to be eaten faster, and a quicker eating rate is linked to taking in more before fullness registers. There are hyper-palatable combinations of fat, salt, sugar, and flavor designed to keep you reaching for more. Lower fiber and altered satiety signaling may contribute too. These are working hypotheses rather than settled mechanism, and sorting out which ones actually drive the calorie gap is an active area of research.

Practical swaps, without the moralizing#

The changes that pay off first start with drinks. Swapping a daily soda for water, sparkling water, or unsweetened tea removes the most consistently harmful source of added sugar with essentially no downside. Reach for whole fruit over juice when you can, since the whole version brings fiber and takes longer to eat.

A few more, all optional:

• Read the added-sugars line. US Nutrition Facts labels now break out added sugars separately from total sugars, so you can see what was put in versus what came with the food.
• Change things gradually. Cutting from two sodas a day to one, then to a few a week, tends to stick better than an overnight ban.
• Cook at home when it is realistic. A home-assembled meal usually lands lower on the processing scale without any special ingredients.
• Keep the frame on the overall pattern. No single food is off-limits, and one packaged item does not define a diet.

What the evidence does not say#

A few caveats deserve to be stated rather than buried. Reverse causation is a live possibility in the observational data: early, undiagnosed illness can change what people eat, not only the other way around. Confounding by overall diet and lifestyle is hard to remove entirely. The ultra-processed category is heterogeneous, so lumping fortified whole-grain cereal in with candy can blur real differences. And the number of controlled feeding trials is still small.

One practical question comes up often: are diet drinks a reasonable stand-in? A 2022 systematic review and meta-analysis in *Diabetes Care* (Lee and colleagues) looked specifically at using low- and no-calorie sweetened beverages in place of sugar-sweetened ones and found no signal of cardiometabolic harm, with outcomes broadly comparable to water. Water is still the sensible default, and the evidence here is still developing, but as a substitution to help someone move off regular soda, the data read as reassuring rather than alarming.

Start with what you drink#

Pick the one drink you have most days and change that first. For a lot of people, swapping a regular soda or a sweetened coffee for water or an unsweetened version moves more grams of added sugar than any amount of label-reading elsewhere. From there, hold the ultra-processed question lightly: treat it as a nudge to look at your week as a whole, lean toward foods closer to how they grow, and let the occasional packaged snack be exactly that. The evidence is strongest where the advice is simplest, and that is a reasonable place to put your attention.