Home Carbohydrates Study: Chocolate cake with breakfast leads to better weight loss than low...

Study: Chocolate cake with breakfast leads to better weight loss than low carb diet over the long term


This study was in the news today, I just had to post it – because it is very interesting. Food for thought. Cake at breakfast with protein led to greater weight loss, significantly more over the long-term for clinically obese people, in a 32 week study. The difference was not in the short-term, but in the second half of the study; 16 – 32 weeks.

Is it the extra carbs at breakfast? (with the protein.)

Is it a bigger meal to start the day?

Is it being able to have food that is usually denied when on a ‘diet’?

Having a treat first thing in the morning leads to less desire for treat food later on?

Having a treat / or is it more carbs with one meal each day leads to greater fat loss in the long run?

Having carbohydrates with breakfast suppressed grehlin far more, is this a key to appetite control and eating less for the rest of the day?

Is this more evidence that having some carbs is better than very low carb?

Top Off Breakfast with — Chocolate Cake?
Tuesday, February 7, 2012

A full breakfast that includes a sweet dessert contributes to weight loss success, say TAU researchers

When it comes to diets, cookies and cake are off the menu. Now, in a surprising discovery, researchers from Tel Aviv University have found that dessert, as part of a balanced 600-calorie breakfast that also includes proteins and carbohydrates, can help dieters to lose more weight — and keep it off in the long run.

They key is to indulge in the morning, when the body’s metabolism is at its most active and we are better able to work off the extra calories throughout the day, say Prof. Daniela Jakubowicz, Dr. Julio Wainstein and Dr. Mona Boaz of Tel Aviv University‘s Sackler Faculty of Medicine and the Diabetes Unit at Wolfson Medical Center, and Prof. Oren Froy of Hebrew University Jerusalem.

Attempting to avoid sweets entirely can create a psychological addiction to these same foods in the long-term, explains Prof. Jakubowicz. Adding dessert items to breakfast can control cravings throughout the rest of the day. Over the course of a 32 week-long study, detailed in the journal Steroids, participants who added dessert to their breakfast — cookies, cake, or chocolate — lost an average of 40 lbs. more than a group that avoided such foods. What’s more, they kept off the pounds longer.

The scale tells the tale

A meal in the morning provides energy for the day’s tasks, aids in brain functioning, and kick-starts the body’s metabolism, making it crucial for weight loss and maintenance. And breakfast is the meal that most successfully regulates ghrelin, the hormone that increases hunger, explains Prof. Jakubowicz. While the level of ghrelin rises before every meal, it is suppressed most effectively at breakfast time.

Basing their study on this fact, the researchers hoped to determine whether meal time and composition impacted weight loss in the short and long term, says Prof. Jakubowicz, or if it was a simple matter of calorie count.

One hundred and ninety three clinically obese, non-diabetic adults were randomly assigned to one of two diet groups with identical caloric intake — the men consumed 1600 calories per day and the women 1400. However, the first group was given a low carbohydrate diet including a small 300 calorie breakfast, and the second was given a 600 calorie breakfast high in protein and carbohydrates, always including a dessert item (i.e. chocolate).

Halfway through the study, participants in both groups had lost an average of 33 lbs. per person. But in the second half of the study, results differed drastically. The participants in the low-carbohydrate group regained an average of 22 lbs. per person, but participants in the group with a larger breakfast lost another 15 lbs. each. At the end of the 32 weeks, those who had consumed a 600 calorie breakfast had lost an average of 40 lbs. more per person than their peers.

Fig. 2. Body weight by Diet Intervention Group. The p-value is for general linear model repeated measures comparisons. HCPb = energy-, carbohydrate- and protein-enriched breakfast diet group, white squares: □ LCb = low carbohydrate breakfast diet group, black squares: ■.
Fig. 3. Ghrelin suppression after breakfast meal challenge at baseline, Week 16 and Week 32 by diet intervention group. The p-values are for GLM repeated measures comparison by group. HCPb = energy-, carbohydrate- and protein-enriched breakfast diet group, white squares: □ LCb = low carbohydrate breakfast diet group, black squares: ■.

Realistic in the long run

One of the biggest challenges that people face is keeping weight off in the long-term, says Prof. Jakubowicz. Ingesting a higher proportion of our daily calories at breakfast makes sense. It’s not only good for body function, but it also alleviates cravings. Highly restrictive diets that forbid desserts and carbohydrates are initially effective, but often cause dieters to stray from their food plans as a result of withdrawal-like symptoms. They wind up regaining much of the weight they lost during the diet proper.

Though they consumed the same daily amount of calories, “the participants in the low carbohydrate diet group had less satisfaction, and felt that they were not full,” she says, noting that their cravings for sugars and carbohydrates were more intense and eventually caused them to cheat on the diet plan. “But the group that consumed a bigger breakfast, including dessert, experienced few if any cravings for these foods later in the day.”

Ultimately, this shows that a diet must be realistic to be adopted as part of a new lifestyle. Curbing cravings is better than deprivation for weight loss success, Prof. Jakubowicz concludes.

Journal Reference:

  1. Daniela Jakubowicz, Oren Froy, Julio Wainstein, Mona Boaz. Meal timing and composition influence ghrelin levels, appetite scores and weight loss maintenance in overweight and obese adults. Steroids, 2011; DOI: 10.1016/j.steroids.2011.12.006



Although dietary restriction often results in initial weight loss, the majority of obese dieters fail to maintain their reduced weight. Diet-induced weight loss results in compensatory increase of hunger, craving and decreased ghrelin suppression that encourage weight regain. A high protein and carbohydrate breakfast may overcome these compensatory changes and prevent obesity relapse.


In this study 193 obese (BMI 32.2 ± 1.0 kg/m2), sedentary non diabetic adult men and women (47 ± 7 years) were randomized to a low carbohydrate breakfast (LCb) or an isocaloric diet with high carbohydrate and protein breakfast (HCPb). Anthropometric measures were assessed every 4 weeks. Fasting glucose, insulin, ghrelin, lipids, craving scores and breakfast meal challenge assessing hunger, satiety, insulin and ghrelin responses, were performed at baseline, after a Diet Intervention Period (Week 16) and after a Follow-up Period (Week 32).


At Week 16, groups exhibited similar weight loss: 15.1 ± 1.9 kg in LCb group vs. 13.5 ± 2.3 kg in HCPb group, p = 0.11. From Week 16 to Week 32, LCb group regained 11.6 ± 2.6 kg, while the HCPb group lost additional 6.9 ± 1.7 kg. Ghrelin levels were reduced after breakfast by 45.2% and 29.5% following the HCPb and LCb, respectively. Satiety was significantly improved and hunger and craving scores significantly reduced in the HCPb group vs. the LCb group.


A high carbohydrate and protein breakfast may prevent weight regain by reducing diet-induced compensatory changes in hunger, cravings and ghrelin suppression. To achieve long-term weight loss, meal timing and macronutrient composition must counteract these compensatory mechanisms which encourage weight regain after weight loss.


► Diet-induced weight loss results in compensatory changes that encourage weight regain. ► Breakfast composition may overcome the obesity-related defect in ghrelin suppression. ► Diet induced increase of ghrelin, appetite and cravings were prevented by enriched breakfast. ► Enriched breakfast may be a strategy to maintain weight loss and prevent weight regain over time.

For more diet and nutrition health news from Tel Aviv University, click here.

Keep up with the latest AFTAU news on Twitter: http://www.twitter.com/AFTAUnews

Great post here go over and read it: ‘The winding path of paleo’
An accurate and humourous look at paleo dilemmas


  1. Well, this is really neat. And should get people in the Paleo community talking, or riled up, depending who you poll! I wonder what the overall carb intake was after breakfast? If you have a large (>50g) carbohydrate shot first thing, but after consuming some protein, and then don’t eat much carbohydrate for the rest of the day you would still be coming in relatively low but not very low for the day.

    • I noticed years ago that whenever I went very low carb, and tried to stick to it, I inevitably ended up with cravings for sweet things. Or after a meal – I keep looking for something sweet. A bit of sweet food or starch sometimes, not necessarily every meal made all the difference at keeping my appetite better regulated.

  2. “However, the first group was given a low carbohydrate diet including a small 300 calorie breakfast, and the second was given a 600 calorie breakfast high in protein and carbohydrates, always including a dessert item (i.e. chocolate).”

    I wonder why they do one group with 300 and the other with 600 calories for breakfast. Also makes me wonder what happened for the rest of the day. Seems a bit odd.

    I agree with you Julianne about low carbing and cravings. I had the same experience. I don’t low carb any longer and don’t get cravings.

  3. This is one of those topics where everyone seems to have a view: you shall have a carb-free-breakfast (paleo), you should have no breakfast (IF/LG), you shall have a balanced breakfast (zone), you shall have dessert for breakfast (this study).
    Would be great if people could finally acknowledge that there is not one right answer to this qn, otherwise one would have found it already – it all depends on the context.
    Having said this, if eating sugar/fructose, then the first meal (whether in the morning or at lunchtime if IF’ing) might be a good place to do so as there is space in the liver. Incidentally wrote about that yday http://thorfalk.wordpress.com/2012/02/09/fructose-is-always-bad-or-not/

    • And it seems to vary between individuals.
      With regards to paleo being carb free, while I acknowledge many people think that, paleo is actually carb agnostic. Personally – I rarely have a no carb meal at breakfast. This is usually when I have fruit with protein – which as you point out is probably the best time to have fruit if one eats it. Added to that I go to the gym most days and feel better for a banana before hand.

  4. Interesting article and very different to Poloquin’s take on a meat and nut breakfast. I have tried low carb Paleo for the past 12 months, but particularly with struggling with stress and burnout, have felt sugar and refined carb cravings more intensely. Now going to experiment with Paul Jaminet’s Perfect Health Diet to try and reduce the carb cravings and keep to more healthy carbs. I think I would struggle with the sugar that is suggested in the paper above though!

  5. Hi, here is what the study says:

    Subjects were assigned to one of two isocaloric weight loss diets which differed primarily in the composition of the breakfast meal:

    a) Low carbohydrate diet (LCb): a low carbohydrate diet with a low calorie, and low carbohydrate breakfast; and

    b)High carbohydrate- and protein-enriched breakfast diet (HCPb) with similar composition at lunch and at dinner to the low carbohydrate diet, but with a calorie-carbohydrate-and protein-enriched breakfast. In this group, the breakfast also included a “dessert” on a daily basis. The “dessert” was a sweet food selected from the following list: chocolate, cookies, cake, ice cream, chocolate mousse or donuts.

    Men were instructed to consume 1600 kcal while women were instructed to consume 1400 kcal daily. Composition of the diet interventions is presented in Table 1. In order to maintain daily energy intake constant, the dinner in the HCPb was reduced from 600 to 300 kcals for women and from 700 to 400 kcals for men (Table 1). All subjects were counseled by a registered dietitian who instructed subjects how to keep daily diet intake checklists for all foods consumed. The subjects’ body weights and dietary intake checklists were monitored every 4 weeks, and dietary adjustments were made as necessary.

      • I forgot to mention that in my opinion, a third group should have been tested: In order to reduce any confounding variable, a third group eating the same breakfast at dinner (as in the high carb group). This would have helped determining if it was the inclusion of sweet in the breakfast per se or the inclusion of sweets in a large meal on the dietary basis. This could have made possible a direct comparisson between the low carb and high carb groups.

  6. @Chris – it says high carb, high protein vs low carb, and it isocaloric to the other one, so it should be (comparatively) low fat

    @Julianne- if ever you get the real papers: the graphs seem just to nice for some nutritional study, normally they are a bit more “oscillating” – anything fishy going on?

  7. This statement also jumped out at me…
    “…the first group was given a low carbohydrate diet including a small 300 calorie breakfast, and the second was given a 600 calorie breakfast high in protein and carbohydrates, always including a dessert item…”

    Why is the fact that cake was eaten at breakfast seen as more significant than the fact that twice as many calories were eaten at breakfast?

    You can’t change multiple variables and then pick on one as the cause for the observed outcomes. It might make for a splashy headline but it is scientifically bogus.

    As a person with Type 2 Diabetes: I see it often discussed on the Diabetes Forums that our tolerance to Carbohydrates changes through the day, with breakfast-time being when most seem to be the least tolerant… something like an apple might send the Blood Glucose very high early in the day but might have less BG impact in the evening, for example. I see no reason why this same daily variability would not apply to everyone and I offer this as caution in assuming that adding refined carbs (like chocolate cake) to breakfast might be a good long-term strategy.

    If you need a chocolate fix early in the day (and I often indulge myself) you might try some 85% cocoa chocolate 😉

    • See outline from Lucas Tafur.
      Essentially – they were eating the same calories, both on low carb diets, except one group shifted 300 calories in the form of a dessert to breakfast.
      As you say two variables changed – the number of calories at breakfast and the addition of carbohydrates, and possibly fat.

      I don’t that means it is scientifically bogus, but it does lead to a number of un-answered questions.

      The study authors have drawn some conclusions form these results, but they might be the wrong conclusions. I think another study changing one variable at a time would be useful.

  8. I had seen that the total daily calories was the same but does that override the variable of having more of those calories at breakfast rather than later in the day?

    Perhaps bogus was too strong a word… it was aimed at the headline rather than their conclusion.. which as you rightly say, is only one of several possible interpretations including:
    * eating more of your daily calories at breakfast (regardless of macronutrient makeup) can lead to greater weight loss in the longer term
    * eating more of your daily calories at breakfast with a specific macronutrient makeup, can lead to greater weight loss in the longer term
    * eating a dessert at breakfast (regardless of overall calories) can lead to greater weight loss in the longer term
    * something else specific about the macronutrient makeup when eaten at breakfast (regardless of calories) can lead to greater weight loss in the longer term

    and so on… changing more than one variable leads to too many possibilities to safely draw any single conclusion

    My own intuitive assumption (that would need to be tested) is that: eating more of your total daily calories at breakfast leaves you more satisfied and less hungry for the remainder of the day. But based on personal experiences I am not suggesting that it is as simple as just calories from any source… in other words: the macronutrient makeup is also important. For example I find bacon, eggs and avocado at breakfast more satisfying for a longer part of the day than the same number of calories from breakfast cereal, skimmed milk and orange juice. For me it is the difference between going ’till lunch or even into the evening before needing to eat again, versus looking for a snack mid-morning.

    I can also relate to enjoying something sweet to conclude a meal (I wonder how much of that is learned from my upbringing?) but the idea that having dessert at breakfast leads to better long term weight loss needs to be tested in a more methodical way than it was in this study 😉

    Perhaps this is all discussed in the full text but I am too frequently frustrated by the apparent lack of researchers in exploring alternate interpretations for their observations. I wonder how often they go in with preconceptions and see what they expected to see and nothing else? Then the media come along and make it worse by inventing a catchy headline, which may not be the direct fault of the researchers but surely they should have considered this eventuality and tried to head it off? Instead we end up with misleading “science”… surely we deserve better?

  9. @FrankG – great analysis; the study seems rather badly designed indeed.

    There is also something really fishy going on in the data, or maybe we just dont have all the info: do they stop dieting after 16 weeks? if yes, why do the breakfast cakers continue to lose weight (fig 2)? and why do the other guys gain weight almost as fast as they have lost it? and in any case how is weightloss and gain so nicely linear over 3 and even 6 months? there should be some tapering off after a while when weightloss becomes more difficult.
    Finally an average weightloss of 20kg over 32 weeks is pretty dramatic on a 1600cal/1400cal diet, especially given that the average starting weight is “only” 90kg – granted this is high, but it is not outlandishly high, and at the end they should come at least into the lower teens bodyfat%

Leave a Reply