The truth about fructose and metabolic syndrome: Part 2 – What Dr Lustig got wrong about sugar consumption
After the previous introduction of the topic Fructose and metabolic syndrome, I would like to proceed to the actual content of Dr Lustig’s videos: Sugar: the bitter truth (2009) to which I referred initially and which made me to choose this topic for my dissertation. Since then I have watched another video of his, named Sugar Pandemics Policy vs. Politics vs. Law (2013). In both of these lectures Dr Lustig repeatedly presented some data and diagrams, to which I would like to write my comments. I am going to use the relevant diagrams from his slides here and contradict his claims with my findings, based on the previous research of the topic. After that you can make your own opinion about the reliability of Dr Lustig claims about the trends of sugar consumption by the Americans and what can be behind the obesity and the metabolic syndrome. I have cropped the diagrams from the printscreen captures, together with the timer on the bottom, so that you can easily find the particular video sections (2013) if you have wished to check some details on your own. Most of the presented materials in these two videos can also be found in Dr Lustig proposal for the U.S. Government intervention, dated back to 2011, see the reference list.
The trends of the sugar consumption in the U.S. – What Dr Lustig showed us
Dr Lustig repeatedly said:
“It is not just we are eating more. We are eating more sugar.”
This was in the relation to the obesity trends. I agree with both of these sentences to some extent. Let’s focus on the first part: We are eating more. My comment on this is that it is a very simplistic statement. The problem is not about eating more or less. It is about the positive energy balance which has been gradually created since early 20th century by not only eating more but also by moving less. Hence, the modern Americans consume more than they manage to burn off and the surplus energy is stored as fat.
However, it is not true that the modern Americans eat more than the Americans ate over 100 years ago. According to the Dietary Studies In The New York City in 1894 and 1896 publication of Atwater and Woods (1898), the average New York worker consumed nearly 4000 kcal and remained lean! In contrast, the average American person energy intake today is around 2600 kcal. The truth is that our ancestors had significantly higher energy output because their main transportation was by foot; they did not have kitchen appliances or take-aways, so the women were sweating when whisking the eggs or working the dough and people in the past worked more manually instead of having office jobs or spending their free time in front of the TV or computers while munching on caloric snacks. The dangers of sedentary lifestyle were recognized back n 1950 in the U.S. when the physical activity was promoted by the Council of the Fitness and Health (Caballero, 2007). Similarly for the UK, David Matthews (2010) wrote about the rise in obesity among the suburban middle-class society in the UK in 1930s or 1950s, due to increasingly modern and comfortable lifestyle, which included more food and reduced physical activity due to the car ownership and other reasons. It was the overall modern lifestyle, not only eating more, that led to the increased body weight.
We are eating more sugar – is the second part of the first statement. Here I am placing a diagram which Dr. Lustig used in his 2013 presentation.
When talking about it, he had put an emphasis on the whole curve and how much the sugar consumption had increased since 1820 until recently. Certainly, the world eats more sugar than a century or two ago.
Back in 1930s the American life insurance companies already determined the association between the body weight and the mortality, so they used the BMI to apply the insurance premiums on the clients (Callabero, 2007). This happened in the decade when the sugar consumption reached a peak since 1820, according to the diagram on your right. However, the obesity was not that frequent around 1950s as it is today, despite the consumption of sugar was already remarkably higher than in 1820. Please look at the curve spot above the 1960-70s and remember that. I will come to that later. There was a relative plateau of sugar consumption between 1920s and 1970s, with a short drop during the WWII. Centers for Diseases Control and Prevention (CDC, 2014) stated that “There was a dramatic increase in obesity in the United States from 1990 through 2010” while other sources pushed the increasing trend back to 1980s as you will see later. Nonetheless, the rise of obesity correlates with the rise in sugar consumption since 1980s. However, does this correlation prove the causation?
We are eating more sugar, I repeat Dr Lustig’s statement. When briefly looking at the diagram it looks that they (the Americans) are. However, this diagram reaches only up to year 2005 or so. When you look above the year 2000 you can see that since then the trend in sugar consumption was decreasing. Indeed, Dr. Lustig admitted at 10:55 minute that there was a decrease in sugar consumption in the last 5-6 years but he also said that it was nothing when compared to the rest of the period (since 1820). That is true. If only we had an overview of the trends in consumption of other macronutrients, so that we could get a whole picture, especially during the last few decades when the obesity and other metabolic diseases, such as type 2 diabetes, have become a major threat to the sustainability of the healthcare system. Is it really the problem of only the sugar or fructose? There are other macromolecules that provide us with energy: proteins and fats. What did Dr Lustig say about fat in 2009?
“The fat is going down, sugar is going up and we are all getting sick.“
In his 2013 lecture, Dr. Lustig repeated the same thing: the Americans were told by the government to go low-fat in order to reduce the risk of cardiovascular diseases while the carbohydrate consumption had increased. What carbohydrate, he asked? SUGAR, he answered his rhetoric question. By this statement Dr Lustig gave the audience an impression that the sugar was the main contributor to the increased consumption of carbohydrates and that the other carbohydrates other than sugar were negligible. Virtually all his media activity, and less known attempts to influence the government policy, has been about how sugar had made the American nation sick. He had supported his answer with nice and clean diagrams, of which I have placed two here. Let’s have a look.
Dr Lustig said that the consumption of fat remained virtually the same while the intake of carbohydrate had increased. He demonstrated it with this diagram borrowed from the original article of Chanmugam et al. (2003). The bars represent the changes in consumption of food items or food groups in grams between two nationally representative dietary surveys conducted in 1989/91 and 1995/96. This diagram represents the consumption by adult Americans older 18 years and over. Similarly, two additional diagrams were presented in that lecture, and these were about the children and adolescents. These additional two diagrams also indicated a small increase of fat consumption (by 5 grams) and 35% increase in consumption of fruit drinks and 41% of soft drinks during the same period of 1989-1996. Was it a lot? I cannot say, because in one slide the change in fat was presented in grams and in other slide the sugar consumption via the soft-drinks was presented in percentage. From this practice it is difficult to get an idea about the extent of the sugar problem in comparison to fat.
Back to the diagram above, the soft drinks and reduced fat milks had the highest increase in consumption among American adults, whereas the full-fat milk consumption showed the biggest decrease (on top of the chart). These changes were within 100 grams of the total amount on average. Those 100 grams represent the product, not the sugar or fat within. If milk contains 3% fat, then of the 100g of milk the fat represents only 3 grams, which is not too much. The third highest increased consumption was of beer, then fruit drinks and tea. Grain mixtures were the last ones that were circled in red as carbohydrates. It looks good so far, except that this was far before 2013 when Dr Lustig delivered this lecture to the audience and this was within a short period of time. He presented 20 years old data as the current trends.
The second diagram compared the trends in the fat consumption and the obesity prevalence. By looking at the diagram while listening to Dr Lustig, an untrained observer would easily be tricked. Why? Because what is showed here is the 10 years old data presented as the current trend again. Or even 20 years, I cannot figure out which point on the x-axis belong to the particular decade threshold and a proper reference to the original material has not been provided. Never mind. My question is: did Dr Lustig have not the access to more recent information about the U.S. dietary patterns and the metabolic diseases in 2013? I do. And I will show you later how manipulative and misleading are these slides and the whole hunt for fructose as well.
There is another problem with this diagram and it has already been picked by others. It is the questionable practice in presenting the rise in the obesity trends along the PERCENTAGE of the fat in the diet. I guess that the author had decided for this practice because both variables were presented in percentages. If someone wants to compare two or more sets of the data, they must have something in common in order to be comparable. However, for this form of presentation to be justifiable the trend in the percentage and the actual AMOUNT of the particular nutrient consumption should match. Had this condition been met, at least within the presented time period? According to the diagram, the drop in percentage of fat calories look impressive, suggesting some 10% decrease. Had the same happened with the actual amount of fat consumption? We already know that the intake of energy had been increasing over the past few decades. Those who understand percentages know that when the total amount changes, so does the amount represented by the percentage even when the percentage figure has not changed. Therefore the percentage says nothing about the actual amount consumed when we do not have these additional figures. I do have these figures and the reality is not what this diagram suggests.
What the data actually says about SUGAR
The following two diagrams will show you how we have been manipulated to see a picture which the presenter wanted you to see. I have made similar diagrams for my dissertation, based on the widely available raw data of the United States Department of Agriculture (then USDA 2013, now 2014 after the recent update). Because some of the figures have changed in their data sheets, I have decided to use the most recent data so that anyone can check them on their own. The data represents an estimation of the consumption of individual commodities/nutrients per capita, adjusted for loss. In other words, what you see in the diagram below is what the USDA expects that the average American person had consumed in the particular years and the organisation did everything possible to adjust for potential losses on the way from the producer to the stomach of the consumer. This adjustment is still not perfect, but it is the best we have and what also Dr Lustig and others have been using as well. The other most comprehensive source of the consumption data are the nationally representative dietary surveys (NHANES) and their data will also be presented here. The ‘dw’ on the vertical axis in the diagram below means the dry weight.
This first diagram illustrates the overall trends of the total and individual caloric sweeteners available per person per year in the U.S. between the years 1966 up to 2012. You can compare what has been presented to you by Dr Lustig in 2009 and 2013: up to year 2005 the latest, but most of his data focused on the the comparison of sugar and fat consumption based on dietary surveys data from 1989-1996 only. Why so narrow time span in the latter case when there was more recent data available? I don’t know. This last period was presented separately for adults (discussed earlier) and also adolescents (described but not shown here). Dr. Lusting constantly kept showing 10-20 years old data. Now, you can see in the diagram that the total caloric sweeteners (not artificial sweeteners) availability has been increasing up to 1999 and then it decreased down in 2012 to the level only little higher than in 1966. In contrast, Dr Lusting showed you only the increasing trend saying that we ARE EATING more sugar and less fat in 2009 and also 2013.
So, looking at the figures, how much sugar do Americans consume today in comparison to 1966 when the obesity on such a large scale was non-existent in the U.S.? According to the USDA (2014) data, in 1966 the total amount of added sugar consumption was 128 pounds per capita per year, whereas in 2012 the amount was 138.9 pounds, increase by mere 11 pounds. In comparison, in 1999 the estimated amount was 169.9 pounds, the highest in the history. Hence, in 1999 the average American consumed 41.9 pounds of sugar more than in 1966 and 31 pounds more than in 2012. Does this look like “the sugar is going up” to you? To me it looks like a pretty large decrease for the last 13 years.
Because I aim to be unbiased and provide as accurate data as I possibly can, I should mention this as well: this data does not include some other caloric sweeteners on the U.S. market, such as fruit juice concentrates (Popkin et al., 2012), sugar alcohols (sorbitol, maltitol), which are also converted to fructose in the liver (Oliveira and Burini, 2012), or sugar contained in imported blends and mixtures, according to the footnote of the USDA 2014 Sugar and sweeteners (added) spreadsheet. All these details point towards under-estimating of the sugar/fructose availability for the metabolism of the Americans when looking only at the consumption presented by the USDA. Nonetheless, the USDA data is still perceived as overestimating due to inaccurate loss adjustment and despite all these additional details it is unlikely that they would have the power to significantly change the curve of the total added sugars consumption or the availability of sugars for the human metabolism.
While some may object that the USDA data may be inaccurate, I have also looked at two analyses of the nationally representative dietary surveys, similarly as Dr Lustig presented for years 1989 – 1996. However, my two additional sources cover a longer period of time and also the more recent years.
Chun et al. (2010) analyzed and compared the consumption of total and added sugars among the American population, based on the NHANES I (1971-1975) and NHANES III (1988-1994). This is a significantly longer time span than what Dr Lusting showed us (1989-1996) and presented as a long-time trend. Chun et al. found out that the total energy intake increased by 144 kcal per day and total carbohydrate by 40g per day (+18%) on average. However, the consumption of added sugar (the one we are interested in here) increased by only 9g/d (+12%), which are only two teaspoons of refined sugar, to give you an idea. Moreover, Chun et al. also stated that “the contribution of added sugar to the total carbohydrate intake decreased“. Therefore, when Dr Lustig asked a question that what carbohydrate we eat more of and he answered SUGAR, he was not saying the truth. Based on Chun et al,:
- it was the non-sugar carbohydrate that increased more in absolute amounts and
- the proportion of added sugar to the total carbohydrates actually decreased.
Therefore, while it is true, that the American population had been consuming more sugar in grams for that particular period (those two teaspoons), when taking this from the perspective of total carbohydrates over a longer period of time, Dr Lustig’s statement was false.
These figures were for the whole American population aged 1 year up to the elderly. When looking individually at the adolescents up to year 18 and adults aged 19 years and over, there were differences. While the energy consumption among the youngsters decreased by 3%, it increased by 11% in adults. Splitting also the sugar consumption into the age groups, the added sugar consumption increased by 5% in children and by 18% in adults (giving together the average 12% mentioned in the previous paragraph). How does this compare to the 35% and 41% increase in the consumption of fruit juices and soft drinks among young Americans between 1989 and 1995 highlighted by Dr Lustig? Quite a difference, is it not? These percentages presented by Chun et al. are not of the total energy consumption, these are only the comparative changes in consumed amounts within over 20 years up to the same point in time (mid 1990s). In this case nobody can argue that these time periods covered by Dr. Lustig (1989-1996) and Chun et al. (1971-1994) do not match at all. I consider looking at a longer period of time as more appropriate, because the dietary surveys have been conducted every couple of years and the consumption of individual nutrients can fluctuate from one dietary survey to another, similarly to the curves based on the USDA data. Sometimes there is increase, the other time there is decrease, and the magnitudes of these trends can also vary quite a lot. This was probably the main problem of the short period of time selectively presented by Dr Lustig as a long term – it was simply misleading, outdated and out of context.
Chun et al. (2010) further reported that the increase of ENERGY from added sugars was very minor: only 1%. This was pronounced as too little for being responsible for the increase in the obesity prevalence during those 20 years. This is very important to understand, because knowing the changes in consumption in grams is good, but again, if we have no idea about the total energy intake and how much have added sugars contributed to this energy along other nutrients, we may end up with a wrong picture overall. In comparison, the increase of energy intake from the total carbohydrates was 4% in adolescents and 5% in adults. So again, another evidence proving Dr Lustig wrong when suggesting that it was primarily sugar the consumption of which was behind the enormous increase of the metabolic diseases in the past decades.
Nonetheless, we are still in mid 1990s, when the sugar consumption was on the rise. How about the later years? Look at what Welsh et al. (2011) found out: between NHANES 1999-2000 and 2007-2008, the total consumption of added sugars decreased by mean 23.4 grams per day, representing a decrease 3.5% by which added sugars contributed to the total energy intake. Please bear in mind that the total energy intake was still increasing. Moreover, the soda consumption, what Dr Lustig constantly presented as the main evil in the current American diet, decreased also. In fact, two thirds of the decreased added sugar consumption was due to the decrease in soda intake, corresponding to net mean reduction by 14.6 g/d. These figures refer to the actual sugar content within the soft-drinks, not the volume or weight of the drinks, which includes water. The only increased source of added sugar was from the energy drinks, representing tiny 0.15 g/d, which is virtually nothing. Now I am asking again: does this look to you that the consumption of sugars in the U.S. is going up? I understand that in 2009 Dr Lustig could not have seen this particular study because it was published in 2011, but how about his lecture in 2013? And how about other data indicating decreased sugar consumption well before 2009?
Regardless of the trends discussed above, the consumption of added sugars in the U.S. is still higher than was estimated to be safe and the soft drinks still contribute the most to the added sugar intake. However, it would be totally incorrect repeating that the sugar consumption is going up when it does not, and to link the ever increasing obesity and other metabolic diseases to sugar, which was actually going down, not up, including the added fructose.
So, if the sugar consumption has been decreasing, while the total energy was increasing (including the non-sugar carbohydrates), what else could contribute to the present ill health of the U.S.?
What the data actually says about FAT
The second diagram I promised to show here is related to Dr Lustig’s claims that the Americans eat more sugar and less fat, also presented earlier by one of the slides, where he compared the reduced percentage of fat while the obesity kept rising. The data for my diagram here was obtained from the same source (USDA, 2014) but from a different spreadsheet named Calories. This diagram only shows the trends up to year 2010, because newer data for some of the calories sources were missing. Nonetheless, you can clearly see that the availability of energy has increased since 1970s (blue curve). You may recognize a little flattened curve of added sugars (red). Although it looks less dramatic, the trend is there: just before 2000 there was a peak and then a steady decline. The purple curve represents the grains, providing mainly glucose to the body and representing most of the non-sugar carbohydrates. The remaining carbohydrates are in fruits and vegetables, including potatoes, of which the consumption remains to be minor in comparison to the grains and the consumption of which has not dramatically changed over the period of 40 years (I have checked it myself in the raw data).
What can be actually seen here is the agreement of the USDA data about the grains availability/consumption with the data of Chun et al. (2010) presented earlier. The diagram shows that the grains (carbohydrates other than added sugars) contributed to the increased energy intake the most for the whole 40 years period. Look how all of these curves started virtually at the same point in 1970 and how they followed their own path for the next 40 years. So, the statement of Dr Lustig that it is the sugar carbohydrates we eat so much of has been disproved by both: the USDA corporate data and also the nationally representative dietary surveys, especially when the added sugars were decreasing whereas the grains remained almost stable since the point above year 2000.
What about FAT? Look carefully at the green curve: until 1999 it almost shared the trend with added sugars but from that point, the sugar consumption went down whereas the fat consumption shot up and kept the trend, contributing the most to the increased total energy intake. In fact, even the grains consumption was on the mild decline whereas the fat was the single nutrient responsible for the increased energy intake among the U.S. population since 2000. Can you see how these two curves: the Total kcal and the Added fats, oils and dairy fats are similar towards the right margin? In comparison, the consumption of meat, eggs and nuts as one category and dairy as another category remained almost stable, the consumption of vegetables has decreased whereas the fruits intake has slightly increased according to USDA (not shown in the diagram). Even if someone was tempted to argue, that added fats are not exactly the same as total fats, because the whole milk consumption went down and so did the fat in meat, it is unlikely that these would significantly change the scenario illustrated in the diagram.
There is another funny point in these trends. Do you remember the diagram with the 10% decrease in fat intake while the obesity has been going up? Then compare the curve of added fats and oils to the curve of the grains up to year 2000: do you see the increasing gap between these two? While the consumption of calories has been steadily increasing, so was the consumption of grains, but the fat and also the sugar did rise to much lower extent. Could we say in this case that the relative intake of grains has increased and the intake of fat has decreased as the percentage of total energy? Sure, it is exactly what had happened. And exactly the same applies for sugars/sweeteners. So, if I have used Dr Lustig’s method, I could as well claim that the sugar consumption had decreased by 10%, because the sugar and fat shared the trends quite closely, did they not? Have a look once more time. How bizarre!
Again, if someone had doubts about the reliability of the USDA data and what I have presented above, let’s have a look at what the dietary surveys said. The analysis of Austin et al. (2011) confirmed that the percentage of fat calories among the adult Americans had decreased between 1971 and 2006 by 2.9%, but there was hardly any change in the actual amount of fat intake in grams. The NHANES surveys indicated that while in men the absolute daily intake of fat had decreased by mean 20 kcal, among women it actually increased by 27 kcal per day. In addition, Chun et al. (2010) had found zero mean change in fat consumption in grams between 1971 and 1994. Would you consider this as a significant decrease in fat consumption so that you could say that the obesity had been rising while fat consumption had fallen down as much as 10%, as it was presented in one of the diagrams discussed earlier? I really do not understand why Dr Lustig sticks to that old and misleading diagram and presents it over and over again in his lectures. To make it even worse, Welsh et al. (2011) further confirmed, that between NHANES 1999/2000 and NHANES 2007/2008 also the percentage of fat calories of the total energy intake has slightly increased, exactly by 0.8% for the American population aged 2 years and older. Hence, there is nothing like the fat is going down. Fat has been going up while the sugar clearly went down.
What does fructose have to do with this? As you already know, fructose is present in the sugar, HFCS and the honey and various syrups, but its largest source in the modern Americans diet is the sugar and HFCS. While it is true that the America consumes more sugar than a century or two ago, its consumption was on the decline together with added caloric sweeteners since 1999. While I agree that the large amounts of ingested fructose produces adverse metabolic responses in animals and humans, I also think that blaming only fructose for the ever increasing metabolic issues, while its consumption was on a decline during the past 13 years (up to 2012 supported by data), is incorrect. And that is the reason why I have decided to write this article: to oppose Dr Lustig’s claims by showing the real figures and suggest that there is another nutrient than sugar or fructose that might have been driving the metabolic diseases in the past decades.
From the last diagram it is apparent that the sugar consumption had been relatively stable in comparison to a substantial increase of energy obtained from grains and there was a remarkable increase in the availability of fat calories since 1999, when the sweeteners availability started to decrease. This fat calories rise is logical when you realize how widespread have the fast-food and fries shops become in the U.S during the past two decades. In addition, the diet sodas have become very popular, even among morbidly obese fast-food addicts, who can give up the extra load of sugar but not the taste of sweeteness in addition to fat and salt. This taste combination had been reported to trigger the response in our brain which is similar to a dose of cocaine back in 2009 or even earlier. I remember reading about it in some papers that time.
I would not have realized the flaws in these presentations of Dr Lustig without being an insider and by doing my own research. I was originally as amazed as anyone of the general public who firstly watched his famous video of the ‘bitter truth about sugar’. Now you have learned the bitter truth about the bitter truth about the sugar. And now I know that I must check the facts before I will believe any person presenting such impressive data in the media, regardless of them being members of the general public or highly educated and skilled professionals.
In the next chapter (Part 3) I will look into the isotope tracer studies of fructose metabolism and what evidence we have to say that fructose is such a villain in metabolic health of people.
In a nutshell, I have proved that Dr Lustig has repeatedly been misleading the audience about the current and recent dietary trends in the United States:
- It is not true that the consumption of sugar in the U.S. is going up, because it has been going down for over a decade and this was confirmed by the data of USDA (2014) and a couple of studies based on the nationally representative dietary surveys data.
- It is not true that the obesity went up while the fat consumption went down, because the diagram, which Dr Lustig used to support his claims, was outdated and misleading. The fat consumption went up during the past few decades and this was again confirmed by both data sources as above.
- It is true though, that Americans still consume too much of added sugars, but they consume a lot of processed grains and other products, also containing added fats, which are contributing to the current prevalence of obesity and metabolic diseases to a greater extent than the sugars/fructose alone.
Atwater, W.O., Woods, C.D., (1898) Dietary Studies in New York City in 1895 and 1896. USA, Washington: Government Printing Office 1898
Austin, G.L., Ogden, L.G., Hill, J.O. (2011) Trends in carbohydrate, fat, and protein intakes and association with energy intake in normal-weight, overweight and obese individuals: 1971-2006. American Journal of Clinical Nutrition 93, pp: 836-843
Caballero, B. (2007) The Global Epidemic of Obesity: An Overview. Epidemiologic reviews 29 (1), pp: 1-5
CDC (2014) Adult obesity facts [Online] Available at: http://www.cdc.gov/obesity/data/adult.html (Accessed 15.4.2014)
Chanmugam, P., Guthrie, J.F., Cecilio, S., Morton, J.F., Basiotis, P.P., Anand, R. (2003) Did fat intake in the United States really decline between 1989-1991 and 1994-1996? Journal fo the American Dietetic Association, 103 (7), pp: 867–872
Chun, O.K., Chung, C.E., Wang, y., Padgitt, A., Song, W.O. (2010) Changes in intakes of total and added sugar and their contribution to energy intake in the U.S. Nutrients 2, pp: 834-854
David Matthews (2010) Obesity and the modern lifestyle: What can we learn from history? [Online]. Available at: http://www.historyextra.com/feature/obesity-and-modern-lifestyle-what-can-we-learn-history (Accessed 16.4.2014)
Lustig, R. (2011) Rationale for Policy Intervention [Online] Available at: http://efnep.ucanr.edu/files/134615.pdf (Accessed 17.4.2014)
Oliveira, E.P., Burini, R.C. (2012) High plasma uric acid concentration: causes and consequences. Diabetology and Metabolic Syndrome, pp: 4-12
Panchal, S.K., Brown, L. (2011) Rodent Models for Metabolic Syndrome. Journal of Biomedicine and Biotechnology. Volume 2011, Article ID 351982, 14 pages
Popkin, B.M., Adair, L.S., Shu, W.N. (2012) NOW AND THEN: The Global Nutrition Transition: The Pandemic of Obesity in Developing Countries. Nutrition Reviews 70 (1) pp: 3-21
USDA (2014) Food Availability (Per Capita) Data System – Overview [online]. Available at: http://www.ers.usda.gov/data-products/food-availability-(per-capita)-data-system/.aspx#.U05XAvldWSo (Accessed: 16.4.2014)
Welsh, J.A., Sharma, A.J., Grellinger, L., Vos, M.B. (2011) Consumption of added sugars is decreasing. American Journal of Clinical Nutrition 94, pp: 726-734