Replacing sugary drinks can make a small difference to weight. Compare sweeteners with sugar and water, and separate trial results from long-term associations.
The drinks shelf offers sweetness with sugar and sweetness without it. You reach for the lower-calorie bottle, then remember a headline saying artificial sweeteners make people gain weight. The bottles are easy to compare. Sorting out the claims about them takes much longer, especially when a small change to an everyday habit has become a debate about which ingredient deserves the blame.
Trials that reduce sugar-sweetened drinks have found small average reductions in weight.[2] Observational studies following people who drink low-calorie sweetened beverages have also reported associations with weight gain.[15] Both findings belong in the discussion. Understanding the apparent disagreement starts with a practical question: what did the drink replace? A sweetened alternative replacing a sugary drink is a different choice from adding it to a routine that already includes water.
This feature examines changes in drinks and their relationship with weight. For the separate question of drinking water before meals, see 500 ml of water before meals and weight loss; for alcohol, see Alcohol and weight management. Here, sugar-sweetened drinks, low- or no-calorie sweetened alternatives, and water share the same shelf. The research becomes more useful when we look beyond their labels to the people, comparisons and periods actually studied.
Cutting sugary drinks produces a small average weight difference
Nguyen and colleagues systematically selected research on sugar-sweetened drinks in children and adults and pooled results in a meta-analysis, keeping observational studies and randomized trials separate.[2] The adult evidence included 21 observational studies involving 448,661 people and 16 randomized controlled trials involving 1,343 people.[2] For the sugary-drink trials, random assignment determined which participants changed their intake and which served as controls.[2] The comparison therefore asks what happens after an assigned change, rather than simply identifying differences between people who already make different choices.
In the adult trials, adding sugar-sweetened drinks to the diet produced an average weight gain of 0.83 kg relative to the control groups. The 95% confidence interval was 0.47 to 1.19 kg. Trials that reduced those drinks found an average difference of −0.49 kg relative to controls, with a 95% confidence interval of −0.66 to −0.32 kg.[2] For sugary-drink reduction, that interval shows the uncertainty around the pooled weight difference of −0.49 kg.[2] The estimate and its interval belong together: the first tells us the size of the difference, while the second shows how precisely it was estimated.
These are average differences between groups across trials. They do not predict how much a particular person will lose after changing a particular bottle. Also, the size of the gain in the addition trials was different from the size of the reduction in the subtraction trials.[2] They were separate sets of comparisons with their own control conditions. It would be misleading to treat them as opposite sides of an identical experiment, or to expect removing a drink to reverse an exactly equivalent gain.
In the adult observational studies, each additional daily serving of sugar-sweetened drinks was associated with 0.42 kg higher body weight. The 95% confidence interval was 0.26 to 0.58 kg.[2] A serving is the intake unit used in the research. This association describes differences linked to consumption; it is not a guaranteed weight change caused by adding or removing a serving. Its numerical resemblance to the trial estimate does not make the two measures interchangeable.
The trials examine differences following intervention. Observational studies examine drinking patterns within people’s existing lives, where food choices and other circumstances can also matter. Observation alone cannot fully separate a drink from those surrounding conditions. In this review, the trial findings pointed in the same overall direction as the observational evidence, making reduction of sugary drinks a reasonable candidate for weight management. The average effect was still small.[2] That scale of change suits a repeatable daily choice better than a demand for a dramatic result on tomorrow morning’s scale.
初日の汗は、予約した人だけが持ち帰れる。
Look at the whole diet after removing sugar
Research on sugary drinks is easily turned into a claim that sugar has a special weight-gain effect regardless of the rest of the diet. Te Morenga and colleagues examined dietary sugars and body weight using trials and observational studies. In adult trials where food intake was not strictly controlled, reducing sugars was associated with an average weight reduction of 0.80 kg. Increasing sugars was associated with an average gain of 0.75 kg.[3] These results concern changes made within freely chosen diets, rather than an experiment holding every aspect of intake constant.
For the reduction, the 95% confidence interval was 0.39 to 1.21 kg; for the gain, it was 0.30 to 1.19 kg. The picture changed when sugars were exchanged for other carbohydrates while energy intake was kept the same. That comparison found no change in weight: the mean difference was 0.04 kg, with a 95% confidence interval of −0.04 to 0.13 kg.[3] The authors interpreted weight changes after changing sugars as appearing to work through changes in energy intake.[3]
Removing sugar from a drink and reducing intake across the whole day are therefore separate things to check. Suppose you replace the sweet drink bought on the way home with water, then use that decision as a reason to add dessert after dinner. The drink has changed, but so has another part of the diet. The equal-energy comparison is a reminder to consider drinks and food together.[3] It does not require judging every meal; it asks whether the intended replacement has actually remained a replacement.
Labels can help with that task. They describe the product you are holding, which is useful when deciding between drinks. But wording about the absence of sugar cannot establish the energy content of your whole diet or determine where your weight will go.[3] A bottle’s label does not grade the rest of dinner. For everyday decisions, the relevant question is what changes across the eating pattern after that bottle enters it.
This distinction also explains why the research is more specific than a general instruction to choose something labeled sugar-free. The review separated reducing sugars under ordinary eating conditions from exchanging carbohydrates at equal energy.[3] The difference between those designs is central to the conclusion. When you apply the findings, keep the surrounding food choices in view rather than attributing the entire outcome to the name of the sweetening ingredient.
Evaluate sweeteners as replacements for sugar
The everyday term artificial sweetener groups together substances that provide sweetness with little or no energy.[20] Research uses related terms such as low-calorie, no-calorie, non-sugar and nonnutritive sweeteners. Some studies compare beverages, while others replace sugars within the diet. Those designs should remain distinct. A result from replacing daily dietary sugars does not automatically answer a question about adding a particular drink, and a beverage comparison does not represent every way a sweetener might be consumed.
McGlynn and colleagues combined 17 randomized controlled trials involving 1,733 adults with overweight or obesity who were at risk for, or had, diabetes. Their average age was 33.1 years, and 77.4% were women.[12] The comparisons included 12 trials with 601 participants replacing sugary drinks with low- or no-calorie sweetened drinks; 3 trials with 429 participants replacing sugary drinks with water; and 9 trials with 974 participants comparing low- or no-calorie sweetened drinks with water.[12]
Replacing sugary drinks with low- or no-calorie sweetened drinks was associated with an average body-weight difference of −1.06 kg. Its 95% confidence interval was −1.71 to −0.41 kg.[12] Body-fat percentage was also lower by an average of 0.60 percentage points, with a 95% confidence interval of −1.03 to −0.18 percentage points.[12] Weight and body-fat percentage measure different things. The weight difference cannot simply be relabeled as an amount of fat lost. For body weight in this replacement comparison, the certainty of the evidence was rated moderate.[12]
The other comparisons were less decisive. Replacing water with low- or no-calorie sweetened drinks showed no clear association with body weight. Replacing sugary drinks with water also showed no clear difference in this analysis. Certainty for body weight in both of those comparisons was low.[12] This does not justify a broad verdict that water is ineffective. It means the review’s comparisons had different amounts of supporting research and different levels of certainty. An absence of a clear difference in a limited comparison must stay attached to that comparison.
Results beyond weight also differed. When low- or no-calorie sweetened drinks replaced water, glycated hemoglobin A1c, a blood-glucose control marker, was higher by an average of 0.21 percentage points. The 95% confidence interval was 0.02 to 0.40 percentage points. Systolic blood pressure was lower by an average of 2.63 mmHg, with a 95% confidence interval of −4.71 to −0.55 mmHg.[12] These outcomes moved in different directions even though the weight comparison showed no clear difference. A weight finding cannot establish that blood glucose and blood pressure will also improve. If you are receiving treatment, discuss drink choices with your doctor in the context of those other measures.
A newer meta-analysis by Li and colleagues combined 19 trials. Participants replacing daily caloric sugars with nonnutritive sweeteners had an average weight difference of −0.79 kg, with a 95% confidence interval of −1.55 to −0.04 kg.[1] Results varied substantially between studies. Differences were significant in studies lasting less than 18 weeks, while effects in longer studies were small and not statistically significant.[1] A pooled estimate therefore needs its time frame beside it. A small overall result does not establish an equally reliable benefit over longer periods.
In the subgroup with obesity, the average difference was −0.646 kg, with a 95% confidence interval of −1.839 to 0.546 kg.[1] That interval included the possibility of no difference. Alongside the small improvement found in McGlynn’s analysis of adults with overweight or obesity, this shows why the participants, intervention, duration and comparator all matter.[1][12] Different reviews need not have identical answers when they assemble different comparisons. For someone routinely drinking sugary beverages, a sweetened alternative can be considered a replacement tool. For someone already drinking water, these findings do not establish a reason to add sweetened drinks for weight loss.[1][12]
It also matters whether a review reports body weight or body mass index, usually shortened to BMI. BMI is calculated by dividing weight in kilograms by height in meters squared. In Toews and colleagues’ review, a small favorable result for adult BMI came from 2 studies involving 174 participants. The mean difference was −0.6, with a 95% confidence interval of −1.19 to −0.01. Evidence was rated low or very low certainty.[19] A large review can contain a particular outcome supported by a small number of participants. The size of the whole review does not make every estimate inside it equally well supported.
The same review found no evidence of an effect from non-sugar sweeteners in adults or children with overweight or obesity who were actively trying to lose weight.[19] That conclusion must sit alongside, rather than disappear behind, the favorable replacement findings.[12] Comparing higher use with lower use or no use is a different question from specifically replacing sugar-sweetened drinks. Removing the comparator from a headline makes those questions sound identical. At the shelf, the useful starting point is your existing drink: look for evidence about replacing that drink, then check how confidently and for how long the result was established.
Numbers: sugary drinks and replacement
Reducing sugary drinks in adult trials: −0.49 kg versus controls; 95% confidence interval −0.66 to −0.32 kg.[2]
Replacing sugary drinks with low- or no-calorie sweetened drinks: −1.06 kg; 95% confidence interval −1.71 to −0.41 kg.[12]
Exchanging sugars for other carbohydrates at equal energy: 0.04 kg; 95% confidence interval −0.04 to 0.13 kg.[3]
At 12 weeks in an adult weight-management program: −5.6 kg with water and −5.8 kg with nonnutritive sweetened drinks.[17]
At 52 weeks in the same trial of 493 adults: −6.1 kg with water and −7.5 kg with nonnutritive sweetened drinks.[6]
初日の汗は、予約した人だけが持ち帰れる。
1
2