The Obesity Epidemic (part 4): Effects of Sugar

In the last three blogs on obesity, I have made the following points:

  1. The entire world is becoming more obese, and the United States is leading the pack.
  2. The most common measure of obesity is “body mass index” (BMI).
  3. There are many negative health outcomes associated with obesity.
  4. Some studies report that obesity can confer health benefits (the “obesity paradox”), but these results don’t hold up when measures of fatness other than BMI are used.
  5. Recent data indicates that about 40% of obese people are actually “fit” and metabolically normal, and do not have increased morbidity, cancer, or cardiovascular disease.
  6. The primary cause of obesity is probably the consumption of excess calories and/or carbohydrates, most likely sugar.
  7. The research concerning the relationship between energy expenditure and obesity is not conclusive—although there is a correlation in the United States between decreased activity and increased obesity, studies indicate that there is no difference in energy expenditure between developed and developing countries.    
  8. The consumption of excess calories during the last three decades of the 20th century can fully account for the obesity epidemic.
  9. Sugar, specifically sugar in soft drinks, is probably the single greatest source of excess calories.

In the last blog we ended with a discussion about fructose (fruit sugar), so let’s pick up there.  

Fructose has been much maligned in recent years, with researchers (and pretty much everybody else) finding correlations between fructose consumption and obesity (!), cardiovascular disease, hypertension, diabetes, and nonalcoholic fatty liver disease.  In 2010, this natural sugar was even hailed by some people as an “environmental toxin.”

One of the topics discussed at the 2012 meeting of American Society for Nutrition was “Fructose, Sucrose, HFCS oh my,” proving that scientists do have a sense of humor.  You can tell from the tongue-in-cheek title of this presentation that the discussion was expected to be controversial and heated.  And by all accounts, it was.  

But unfortunately, after all that intense debate, their conclusions concerning the impact of fructose consumption on health can be summarized in less than ten words:  there is not enough good data to say.  The problem is that fructose studies have many of the same problems as other nutritional studies—in order to clearly see the effects of fructose in the diet, the amount consumed must be unrealistically high, and even then, it is difficult to separate the effects of fructose from the confounding effects of other nutrients such as fats, proteins, complex carbohydrates, and other sugars.  For example, sucrose is about half glucose and half fructose, so if you reduce the amount of sucrose in the diet, you are actually decreasing both glucose and fructose, and thus you can’t blame, or credit, either one of them for any observed effects.  As an added complication, fructose is metabolized in the liver—where half of it is converted into glucose.  So are you seeing the effects of fructose or glucose?  And as I mentioned in the previous blog post (The Obesity Epidemic, part 3), U.S. consumption of fructose has declined in the last decade, but the rate of obesity has not.

Since ingesting sugar increases available energy, it is possible that at least some of the results of the various obesity studies are in fact due to excess energy rather than sugar consumption.  The potential causes and effects are such a tangled mess that the best that science can do in terms of identifying the villain is that it is either the sugar itself or the excess energy that results from the sugar.  And of course, it may be both.

Even so, the culprit identified in the last blog—sugar sweetened beverages (SSBs)— seems to be emerging as a principal bad actor in a number of studies that examined the relationship of SSBs with obesity, diabetes, and cardiovascular disease.  

First, obesity.  The following graph is based an 8-year study involving 50,000 nurses, with the data having been adjusted for age, alcohol intake, physical activity, smoking, postmenopausal hormone use, oral contraceptive use, cereal fiber intake, and total fat intake.

This graph plots the weight of the test subjects at three different times:  in 1991, 1995, and 1999.  At the start of the study in 1991, the nurses were divided into two categories,  either “low” consumers (those who drank less than one SSB a week) or “high” consumers (those who drank more than one SSB a day).  First off, you will notice that there is a glaring anomaly—in 1991 the low consumers weighed more than the high consumers! (Seems to be counter to the point I’m making, doesn’t it?)  Be that as it may, everyone gained weight over the next eight years, and the “high” consumers gained the most.  Even more telling, when some of the “high” consumers switched in mid-stream to became “low” consumers, they lost weight.   And although this particular graph only reflects the correlation between SSB consumption and weight, the researchers also found that higher consumption of SSBs was also correlated with higher rates of diabetes.

A 6-year study of 40,000 black women (1995-2001) had similar results:  the test subjects who drank more SSBs (and fruit juices) also gained more weight and were more prone to diabetes than those who drank fewer sugary beverages.  And just to show how unpredictable these things can be, a similar German study in 2002 found that consumption of SSBs was correlated with weight gain in men, but not in women.   But all in all, the data seems to slant pretty significantly towards a strong correlation between heavy SSB consumption and obesity, as well as diabetes.

Well, so much for the weight-gain literature.  Let’s discuss SSB consumption and diabetes in more detail, which will lead us to cardiovascular disease and the “metabolic syndrome.”

First let’s talk about diabetes.  There are two types—diabetes type 1 (DT1), and diabetes type 2 (DT2).  We are interested in DT2, as it is the one that’s associated with obesity.  DT2 occurs when the body’s cells become “resistant” to insulin or when the body’s insulin production either increases or decreases.

Insulin “instructs” liver cells, for example, to take up glucose and store it as glycogen (a storage form of glucose in animals, as opposed to starch, which is a storage form of glucose in plants).  Alternatively, the liver may take up the glucose and make fatty acids, which then become “fat.”  In this way, insulin regulates blood glucose and is one of the “controllers” of fat accumulation.

So when insulin resistance results in DT2, we get elevated glucose in the blood, as well as elevated insulin—since the insulin is not being taken up by cells.  What seems to be happening is that elevated glucose in the blood over a period of time causes muscle and liver cells to stop taking up glucose, which means there are even higher levels of glucose in the blood.   The body produces more insulin in an effort to bring those levels down, and if muscle and liver cells develop resistance to this increased insulin, uptake is lowered and insulin levels in the blood rise even higher.  Fat cells, particularly those making up the roll around your waist, respond to the elevated insulin by doing what they are supposed to do—store fat.  So you get fatter because there is more insulin circulating around in your body.

Confusing?  Yes it is.  It can even be hard to tell what is a cause and what is an effect.  And that puts us right about where the science is at this point in time.  It does seem to be clear is that obesity and DT2 are correlated, but the exact connection between them has yet to be determined.   Some believe that fructose consumption is the main culprit in the development of insulin resistance, but as I mentioned in part 3 of this series, this view is by no means universal.  
That brings us to a phenomenon known as the “metabolic syndrome” (MSYN).  There are several different definitions of MSYN, but according to the International Diabetes Federation, a diagnosis of MSYN requires obesity along with two of the following: raised triglycerides, reduced HDL cholesterol, raised blood pressure, or DT2.  The World Health Organization’s definition for MSYN requires DT2 and two of the following:  elevated blood pressure, elevated triglycerides and cholesterol, or a BMI (body mass index) greater than 30.  And finally, the US National Cholesterol Education Program requires at least three of the following:  “central obesity” (waist circumference greater than 40 inches in men and greater than 35 inches in women), elevated triglycerides, elevated blood pressure, or elevated glucose (probably DT2).

So MSYN is really a constellation of characteristics that may or may not have an underlying common cause, but certainly together increase the risk of DT2 and cardiovascular disease.  And obesity shows up in all of the definitions of MSYN, though most of them allow for a diagnosis of MSYN in patients who are not obese.

And get this:  44% of the US population over the age of 50 has MSYN.  75% of British patients with DT2 or pre-DT2 have MSYN.   50% of patients with coronary heart disease have MSYN.  


So, given what we have learned so far, it seems reasonable to suppose that if you reduced obesity, you might also reduce DT2 and cardiovascular disease.  Scientists love these kinds of experiments.  You see what’s going on here—if we say that obesity increases the likelihood of DT2 and coronary heart disease, then reducing obesity should have the opposite effect, right?  That is a good check of the theory and a classic “if then” experiment.  

An 11-year project called Look AHEAD (Action for Health in Diabetes) was begun in 2001.  This is/was an “intervention” clinical trial involving approximately 5,000 people aged 45-74 with DT2 and a BMI greater than 25.   Therefore, the test subjects were required to be at least minimally classified as “overweight”; they were also required to be cancer-free and not suffering from  cardiovascular disease.  They were divided into two groups:  one group was placed on an aggressive weight-loss program  involving exercise and a diet restricted to 1,200-1,800 calories per day depending on starting weight, and the other group received normal counseling and treatment for DT2 (called the “diabetes support and intervention” group—or DSE).  The goal for each person in the diet/exercise group was to lose at least 7% of body weight within the first year (e.g., 17.5 pounds for a person weighing 250 pounds).  Those who failed to meet this goal received extra counseling.  

To date, at least two publications have come out of this study.  The first one (2012) looked at the effects of weight loss on DT2, and the other (2013) looked at weight loss and cardiovascular disease.  

The 2012 paper showed that, yup, weight loss was correlated with remission of DT2.  Hooray!  But did the reduction in DT2 result from, say, changes in the diet or changes in the fat cells themselves?  Still unknown.

The 2013 paper is rather more interesting because it showed that weight reduction did NOT result in decreased cardiovascular disease.  There was also no effect on death from any cause.  This disappointing result may be due to the relatively small amount of weight actually lost by the test subjects—only 3.5% after 9 years, down from a high of 8.6% after one year.  Or, it may be because folks in the non-diet/exercise group took statins.  But it is encouraging to note that even though the weight-loss group did not experience lower rates of cardiovascular disease, they experienced a number of other benefits, including reduced urinary incontinence, sleep apnea, and depression, and higher levels of physical fitness.

So there you have it.  Increased sugar consumption in general, and sugar-sweetened sodas in particular, seem to account for the majority of our obesity.  And obesity is correlated with a suite (no pun intended) of conditions known as “metabolic syndrome,” which may, in come cases, be reversible with weight loss.  Unfortunately the same can’t be said for cardiovascular disease.

And of the seemingly endless number of issues involved in a discussion of the obesity epidemic, what is left to cover?  Next up are what I think of as the “minor” explanations for obesity (gut flora and genes, for example), and then finally the dreaded topic I was hoping to avoid—dieting.

Useful References: