Sugar-sweetened drinks and obesity: threat, opportunity, or both?

Following the failure of the ‘eat less fat’ strategy for addressing the obesity epidemic the focus has shifted to sugar-sweetened beverages. Will this approach be more fruitful? Is sugar the problem, or is it liquid calories? Or are we missing something obvious?

Two new studies recently published in the New England Journal of Medicine have shed further light on the issue of soft drinks and weight gain.

Two new soft drink studies

Ebbeling and colleagues studied 224 overweight and obese adolescents who regularly consumed sugar-sweetened soft drinks. Half the subjects participated in a 1-year program designed to decrease consumption of these beverages. Both this intervention group and a control group were followed up for a further year after the intervention was complete. After one year the mean body weight of those in the intervention group was significantly lower (1.9kg) than that of the control group, though significance was lost at two years. So the program worked while underway but its effects did not persist.

The second study by de Ruyter and colleagues involved 641 primarily normal-weight children aged 5-12 years. Over an 18-month period half the children drank a sugar-sweetened soft drink each day; the other half drank an artificially sweetened soft drink. As these were growing children, body weights increased in both groups but those taking the sugar-sweetened soft drink gained a kilogram more than those drinking the artificially sweetened beverage.

Taken together, these studies suggest that sugary soft drinks may indeed increase the likelihood of weight gain. But why is it so?

Image: source

Liquid calories and energy balance

One of the mysteries of modern nutrition is how anyone manages to maintain normal weight in our superabundant food environment. Despite very different diets, food intake varying from day to day and the vast number of calories in and out each year, most people manage to get the balance almost right. Energy intake is fairly tightly controlled by the sensations of hunger, satiation and satiety. When energy intake is too high or too low these sensations adjust to control energy intake. Yet the system of adjustment and compensation seems imperfect when it comes to sugar-sweetened soft drinks.

The main hypothesis explaining this phenomenon is that calories in liquid form do not trigger the same satiety response as calories from solid foods (see Malik 2010), so the compensation for liquid calories is incomplete allowing overconsumption of energy. But is this a reasonable explanation? Aren’t stomach contents liquid? Why would the solid or liquid form of a food outside the body have an effect once the foods are sloshing around in your tummy?

What about milk?

In the best study so far linking individual foods to future weight gain Mozaffarian and colleagues showed that most calorie-containing beverages were associated with future weight gain, including sugary soft drinks, fruit juice and alcoholic beverages. But there was one notable exception – milk. Both whole milk and low fat milk were not associated with weight gain.

Effects on satiety may be the key. A West Australian study published a few years ago looked at how consuming fruit juice and skim milk affected satiety. Even though the volume and calories of the milk and juice were exactly the same, subjects found the milk to be more satiating than the juice. And calories consumed at a subsequent meal were lower after the skim milk.

Image: source

Is it the protein?

Soft drink, fruit juice and skim milk are all liquids and they all contain sugar so what explains the very different effects of these beverages on satiety and weight gain? It is tempting to think that the answer is protein. Some years ago Professor Steve Simpson from the University of Sydney proposed the Protein Leverage Hypothesis which suggested that dietary protein was the key to understanding the obesity epidemic and how it could be addressed. Based on extensive evidence from other species Simpson proposed that human appetite was not driven energy requirements alone: we ate to a protein target.

According to the hypothesis, if the available diet is relatively protein-rich the target for protein will be met well within caloric requirements, satiation will occur and eating will stop. This would explain the apparent benefits of higher protein diets for weight management. On the other hand, if the diet is rich in carbohydrate and fat but low in protein people need to eat a lot of food before meeting the protein target and they will tend to over-consume energy. Seen through this prism the problem with soft drinks is their lack of protein when compared to beverages such as milk.

Liquid meal replacements

Since we are discussing sugar-sweetened beverages and body weight, it is worth remembering that the evidence in favour of liquid meal replacements for the management of body weight is quite strong. A meta-analysis of six studies has shown that weight reduction using meal replacements was more effective than weight reduction using a conventional calorie-reduced diet.

Typically, liquid meal replacements are sugar-sweetened milk drinks. Again, being a caloric liquid or in this case a sugar-sweetened liquid, did not seem to pose a risk for obesity. On the contrary, the milk base of these products appears to offer a benefit.

Presence of harm or absence of goodness?

Nutritionists are conditioned to assess foods and drinks in terms of their sugar content and to pass negative judgement when ‘added sugar’ is present. But how are we to interpret evidence that sugar-containing beverages may be positively or negatively associated with weight gain, or neutral? Frankly, a focus on the sugar content of these beverages explains nothing.

Rather than sugar being an obesogenic agent or liquid calories being a problem, doesn’t the evidence suggest that the key issue is the presence or absence of essential nutrients, especially protein? Wouldn’t this explain why nutrient-poor soft drinks are less satiating and more likely to be associated with weight gain than milk or milk-based meal replacements, both of which are nutrient-rich?


13 thoughts on “Sugar-sweetened drinks and obesity: threat, opportunity, or both?

  1. Interesting discussion as usual.

    Might the next logical experiment (or market entrant) include soft drinks with various types and amounts of protein added in disguise?

    Liquid meal replacements, although they contain some sugars, are consumed in conjunction with a low kilojoule, low carbohydrate diet. This is ketogenic overall. I don’t think that you can fairly suggest that it is the milk protein (in isolation or at all) that is the variable.

    • Hi Trudy
      Meal replacements are sometimes recommended as a weight maintenance measure (one meal replaced) or for weight reduction (2 meals replaced). These approaches are not ketogenic.
      My key point was that the problem with sugary soft drinks is not sugar pe se. Rather it is soft drinks’ lack of essential nutrients. The body’s response to a sugary soft drink seems to be “Keep consuming. I’m not getting what I need”. Protein may or may not be the key. Some people have argued that the calcium in milk is the critical thing.
      I just wish everyone would stop panicking about sugar and think about things more deeply. Regards, Bill

  2. Great reading as usual thanks Bill. I would still like to see less nutrient poor energy dense beverages available for our children to consume. The graduation from the best food for you in the milk bottle as an infant (for those children who were unable to receive breastmilk via the breast) to all manner of forms of liquid offered today in preference to water suggests a dependence on the bottle? Are our children searching for ‘nutrients’ or some psychological comfort? I am sure it is not just thirst!

    • Hi Fiona
      Totally agree with you on the need to see less nutrient poor, energy dense beverages available to children. I just don’t see ‘no added sugar’ and ‘nutrient poor, energy dense’ as meaning the same thing.
      Chocolate milk has the same (high) nutrient density as full cream milk, even though one has added sugar and one doesn’t. Why do dietitians and nutritionists argue against chocolate milk?
      I think we should focus more on nutrient density and less on sugar. Our advice would be more logical and more consistent. Regards, Bill

      • Great discussions. I agree that there should be a focus on what we need as well as what we should have less of, rather than just the latter. In addition, chocolate milk is a great tool for recovery from exercise, but as you stated, it has significantly more sugar with similar amounts of the other nutrients to normal milk. Doesn’t the added sugar and therefore extra energy mean that a chocolate milk has less nutrient density per kJ.
        I think the lowering of added sugar is pertinent as advice and should be in conjunction with appropriate replacements.

        • Hello Cam.
          Compared with full-cream milk, chocolate milk has more sugar but less fat so calories are similar and the nutrient density is almost identical. Yet we tend to judge the beverage with ‘added sugar’ adversely. If chocolate milk is nutrient-rich, who cares if it has some sugar in it?
          Neither the sugar content nor the fat content of the diet affects the risk for conditions of public health significance, but they both have the potential for nutrient dilution. However, whether this potential is realised depends on which sugar-rich or fat-rich foods are included or excluded from diets. Rather than tying ourselves in knots over which sugar-containing foods should be included in healthy diets, why not forget about sugar (and fat) altogether? Just focus on nutrient density. As a general rule, if a food is nutrient-dense, it’s in. If it’s nutrient-poor, it’s out. Nice and simple. Regards, Bill

  3. Hi Bill
    So based on this discussion, (which was really interesting and informative), you would say commercial products such as an up or go, or milo in a low fat milk are not necessarily adverse to an individuals health?


    • Hi Kate
      I’d rather not comment on individual products so let me comment generally on the two concepts you mention.
      The first is a breakfast replacement beverage designed for people pressed for time in the morning. So the likely scenario is no breakfast or a breakfast drink. The breakfast drink has to be better than nothing. It’s certainly nutrient-rich, which is a good thing. And it’s portion-controlled which is also a plus in this day and age. I believe it has added sugar but why should I be concerned about that? How would the sugar content affect any parameter associated with an adverse health outcome? I suspect the glycaemic load of the beverage would actually be less than that of a more traditional breakfast (bread, cereal, juice, etc). And the energy content is controlled.
      Sometimes I think dietitians have an idealised view of the way a breakfast should look – a family sitting around the table eating bread, cereal, juice, etc. But that’s not the modern world with both parents tearing off to work. Whether we like it or not I suspect the breakfast beverage may be where many people are heading. Aesthetically, it’s the pits. But, looked at it scientifically, I think it’s fine.
      In relation to adding sweet chocolate flavouring to milk, I think we should think about the nutritional issues facing children and the beverage alternatives. Some children, especially adolescent girls, have borderline calcium intakes. Ideally they should drink more milk and less soft drink. But let’s face it children like sweet drinks and soft drinks are popular in this age group. If the cost of achieving higher milk intakes is a little added sugar and chocolate flavouring, is that such a bad thing? In this scenario I see sugar as facilitating the intake of important nutrients, a total contrast to the role of sugar in a soft drink.
      That’s why I think a focus on sugar per se is counterproductive. Why not just focus on nutrient density? On this basis I would give both concepts the green light. Regards, Bill

  4. Thank you so much for your input, these have been my thoughts exactly. Plus I feel less guilty about my breakfast. I agree too many dietitians try and impose the ‘idealistic breakfast’, even the next generation (or at least thats what I found amongst the majority of students in my course). I wish I had been reading this blog during my studies, rather than at the end.

  5. Hi Bill,

    Is there evidence demonstrating that a liquid fat beverage would be similar to a sugar-sweetened beverage in its inability to induce satiety? If sugar’s effect here were unique, then we might expect it to have a poor showing against fat-rich liquids with equal protein and calorie content.

    • Hi Will
      From what I have read the satiating potential of fat and carbohydrate are similar and both less than protein. Not sure if this has been tested using beverages. Regards, Bill

  6. Discussions about added sugar need to be in context to the situation. If added sugar in a milk beverage, high in other nutrients, enhances its food of choice, rather than a non nutrient high energy dense product, then it’s a valuable one. Eg, tradesmen who work long hours in physically demanding jobs , often driving between jobs, tend in my experience, to have a liking for chocolate or coffee flavoured milks which provide them with protein, calcium and B vitamins that would be otherwise reduced in their diet if they chose a soft drink of negligible nutritional value except energy. It is only when there is excessive use of any given food that is not expended by physical activity that it becomes problematic as part of the overall diet-style.

    • Hi Liz
      I agree with you – context is everything. That’s why I argue that sugar content is a very poor measure of a food’s nutritional quality. How can anyone argue that sugar in fruit is OK but sugar in chocolate milk is bad? Sugar is sugar. If sugar content is a good measure of nutritional quality we should be able to apply it universally, but we can’t, or we don’t.
      We really need to embrace measures of carbohydrate quality that can be applied universally. For me, glycaemic index is important and can be applied to all carbohydrate-rich foods, sugary or starchy. And nutrient density is important. If any carbohydrate-rich food has a high GI (due to sugar or starch) and it’s not nutrient-dense, there is a good case for restricting it in a healthy diet. Regards, Bill

Leave a Reply

Your email address will not be published. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>