Teen Skepchick Interviews: Diana Williams, Part 2
This post is part of the Teen Skepchick Interviews series, where TS writers talk with amazing women scientists and skeptics about life, the universe, and everything.
Today’s interview is a continuation of last Thursday’s interview with Dr. Diana Williams, who researches neuroendocrine control of food intake, body weight, and metabolism (see part 1 here). Today, we discuss more about dieting, whether it can be harmful and when, as well as the connection between weight and health, whether there really is an obesity epidemic, and what we can do to separate truth from fiction when it comes to claims about weight, dieting, and health.
Does yo-yo dieting cause greater weight gain? Is there really such a thing as a sugar high? Do you need to be concerned about gaining the freshman 15 your first year of college? The answers may surprise you.
If you have questions after reading either part of this interview, Dr. Williams would be more than happy to answer them in a follow-up post. Add them to the comments here or on Facebook, or send a tweet to @teenskepchicks.
Does dieting do any harm?
There’s not much evidence that more reasonable diets do any harm to physical health, and they may be beneficial for people who had less healthy eating habits regardless of whether the diets result in any long-lasting weight change. Because of the lack of long-term success of dieting for weight loss, many people will “weight cycle” over the years. Sometimes also called yo-yo dieting, this is when people lose, gain, lose, gain, etc. You might imagine that this kind of variability in weight over time could be harmful, but the data on this are mixed.
Some studies show no negative impact on things like cardiovascular health or overall mortality (that is, risk of dying from any cause), but others have shown some increased risk with more variable weight. It’s often anecdotally said that yo-yo dieting leads to greater weight gain— that is, dieters end up heavier than they were before they started dieting in the first place. Not all studies have found that to be true, but a recent study by Savage and Birch suggests that we may see individual differences in weight gain after dieting depending on the type of weight-loss strategies that were employed. Women who used unhealthy weight control strategies (such as fasting, skipping meals, use of liquid diets, diet pills, appetite suppressants, laxatives) showed the greatest weight gain over several years compared with women who used only healthy strategies (e.g., reducing caloric intake, eating more vegetables and less meat, increasing exercise, eliminating junk food).
So some particular diets or methods are more harmful than others?
There has been a lot of discussion about the possible cardiovascular risks associated with high protein, low carbohydrate diets, because they tend to encourage rather high fat intake. I’d say the jury is still out on that right now. Some studies show no increase in cardiovascular risk or mortality with those diets, while others have shown increases in markers associated with cardiovascular disease.
Extreme diets that involve fasting—severe caloric or fluid restriction—like some of the recently popular “cleanses” could pose more of an immediate threat. Before starting any diet, but especially one of the more extreme plans, it is a good idea to consult with your doctor to make sure the plan is safe for you.
Is there an obesity epidemic?
There has indeed been an increase in the average weight of Americans and others in Western and Westernized nations over the past several decades. But I don’t like the term “epidemic” in this context because it carries the connotation that obesity is spreading like wildfire and pretty soon most if not all of us will be obese. It’s worth taking a critical look at how we define overweight and obesity, and exactly how weight has changed over time.
“Overweight” and “obese” refer to specific ranges of body mass index (BMI), which is calculated as weight in kilograms divided by height in meters squared: that is, kg / (m × m). (BMI was developed as a demographic tool to describe large populations, and you can read more about the serious flaws in applying BMI at the individual level here: “Top 10 Reasons Why The BMI Is Bogus.”)
The overweight range is BMI of 25-29.9, and the obese range is 30 and above. The World Health Organization further categorizes obesity into 3 classes, with class I at 30-34.9, class II at 35-39.9, and class III at 40 and above. Classes II and III are sometimes called morbid obesity. These definitions are somewhat arbitrary and have changed over time.
I bet that most people have little or no instinctive idea of what a person with a BMI of, say, 31, looks like. The “obesity epidemic” language gives the impression that two-thirds of Americans are super-fat, but that’s far from true. A great way to familiarize yourself with what people in the overweight and obese range actually look like is to check out Kate Harding’s BMI Project, a photo slideshow of volunteers who have given their BMI category.
Whether you count as overweight or obese is a matter of your BMI making it past the threshold for one of those categories. Just a couple of pounds can do that. So it actually wouldn’t take a terribly large average weight gain to move the population statistics on prevalence of overweight and obesity. And that’s exactly what happened over the past several decades. For example, it’s been estimated that in the United States, the average increase in body weight during the 1990s was 7-10 lbs. That was enough to move large numbers of people up into the overweight and obese categories, so we have the one-third overweight and one-third obese prevalence statistics. But when you think about this in terms of average weight change, that doesn’t quite support the “OMG everyone is so obese!” scare language that we so often hear.
Anecdotally, people often claim that they see more very obese people on the street than they used to 10 or 20 years ago, but the fact is that although the prevalence of morbid obesity has increased significantly, that’s still a relatively small percentage of the total population—2.2% in 2001 up to 3% in 2005. You can look at that kind of change in two different ways: (1) the prevalence increased by almost 50%, what a dramatic rise! OR (2) we’re still talking about only 3% of the population here—this is not a super-obese world takeover. That’s not to say that the trend upward is not a concern, but the magnitude of the change is not as large as many people believe.
Is there any connection between weight and health?
Many studies show correlations between BMI and various health problems, including things like cardiovascular disease, type II diabetes, some forms of cancer, and sleep apnea. But it’s not as straightforward as high BMI = unhealthy, or even fat = unhealthy.
A major issue with just about all studies relating BMI to health risks in humans is that they are correlational. They can show that people with higher BMI also tend to have higher health risks, but they cannot show that being fat is the cause of those health problems. It may seem intuitively like that’s the most likely explanation, but it’s just bad science to assume causation from correlational data. In fact, we know from studies on animal subjects that you can cause many of the same health problems by simply giving the animals a high-fat or a high-fat and high-sugar diet while restricting their food intake so that they gain little or no weight. So poor nutrition can be a big factor regardless of fat mass.
Exercise is another factor. Physical activity has many positive effects on health regardless of body weight, and if overweight or obese people are less likely to exercise (as some studies have shown), then that could account for some of the increased health risk. Overweight and obese people face a significant amount of weight bias and stigma, and can receive lower quality medical care as a result. Some avoid seeing doctors altogether. Lack or lower quality of preventive medical treatment likely also contributes to some of the increased risk associated with obesity. Fatness alone may cause problems, but there are many other factors at play here that make it difficult to determine with certainty.
What about the claims about obesity being the second-leading cause of death?
It has been reported in the media that obesity accounts for hundreds of thousands of deaths per year, but that is not a settled fact. Some of the studies that claimed to demonstrate such large effects of obesity on mortality actually had fatal flaws in sampling and statistical analyses that could have caused a large amount of overestimation.
For example, some of those studies removed many of the deaths of thin people from the analysis because those individuals had suffered from some identifiable illness, but included all the deaths in the overweight and obese groups. It’s pretty obvious how that could lead to a biased estimate. In 2005, one of the best studies on this topic was published by Flegal and colleagues, in which they used a more representative sample and properly controlled for other potential causes of death. They observed that there was actually no increase in mortality in overweight compared with “normal” weight populations, and that the increase in mortality with obesity was lower than had been previously reported. This remains controversial, but I haven’t seen any more convincing analysis to the contrary.
It is worth noting the strong confirmation bias that exists in this area. Confirmation bias is the tendency to find support for pre-existing beliefs. If we already believe that obesity causes poor health outcomes and death, then we are more likely to find support for that hypothesis. This effect is especially relevant when the scientific and statistical analysis is complex and difficult for non-specialists to understand.
A great example of this phenomenon happened recently after there was a great deal of media attention given to the notion that obesity is “contagious” from friend to friend, based on a social network analysis by Christakis & Fowler published in 2007. In fact, their statistical methodology was seriously flawed and thoroughly debunked by the statistician Russell Lyons, first on his website and ultimately published in the journal Statistics, Politics, and Policy this year (others have published criticism of Christakis & Fowler’s methods as well). Whereas the notion that obesity is a socially contagious disease was very well received, the debunking got relatively little attention and took years to publish, finally landing in a less prestigious journal than the initial paper claiming an effect.
There is no doubt that overweight and obese people can be in good health. Family history (genetic predisposition) still accounts for most of the risk for cardiovascular disease and type II diabetes, and it is by no means the case that all fat people will wind up with one of these disorders. True, plenty of overweight and obese people are not in great shape, eat a lot of junk food, and are sedentary. But the same is true of many thin people. On an individual level, using BMI or body weight as a proxy for health is simply inaccurate.
What are the most interesting recent studies related to diet and health?
I’m a behavioral neuroscientist, so the research that I find most exciting is about how our brains mediate behavior. Over the past decade or so, there’s been an increasing amount of research on the rewarding value of food. Tasty foods, often high in sugar and fat, are particularly rewarding, and we know that it’s far easier to eat large amounts of delicious food than it would be to eat something relatively flavorless, even if both foods provide essential nutrients. Recent evidence suggests that hormones that regulate food intake, like leptin, do so in part by acting in brain areas that control reward processing. These are the same brain regions and neurotransmitters that are responsible for drug addiction. In fact, drugs of abuse are tapping into the brain’s system for mediating natural rewards like food and sex. If we didn’t have strong motivation for those things, the species wouldn’t survive.
We know quite a lot about drug addiction and how the brain responds to drugs of abuse, and the application of some of that knowledge to food reward has been very useful. Previously, the focus had been more on mechanisms by which the brain detects stomach and intestinal fullness and fat mass. Without a doubt, those mechanisms are important, but they interact with these food reward processes. You may be quite full after a meal but still choose to eat a delicious-looking brownie. These kinds of studies will help get us closer to understanding the neural circuits and chemicals that control our conscious decisions about whether to eat and what to eat.
What popular myths about food and our bodies are not rooted in evidence or even run counter to it?
The biggest one, which we’ve been discussing all along here, is the myth that we have full control over our body size. I’ve said plenty about that, so here are a few more.
There is no such thing as a “sugar high.” Eating candy will not give you energy. People often report that they feel energized by sugar, but this is just another example of confirmation bias. In controlled experiments, there is absolutely no effect. If anything, there is some evidence for the “sugar crash,” where you are more tired several hours after eating a lot of sugar. The popular energy drinks are really all sugar and caffeine. The sugar works against “energy” if anything, and the caffeine doesn’t work for everyone, much less if you are a regular consumer of caffeine.
“Cleanses” and ideas about eliminating toxins always seem to be popular. There is absolutely no scientific basis for the health claims of these types of programs. Any effect on weight is simply the result of the fasting that is done during the cleanse. And the idea that you need to do something special to “clean out” your intestines is just plain silly.
There are hundreds of weight loss scams like human chorionic gonadotropin (hCG), various other herbal and homeopathic remedies, and gadgets that claim to “zap” away fat. They are all total crap. Too good to be true usually is. Don’t trust late-night TV advertisements. One of my favorites is called Leptoprin, a pill that contains nothing that might actually cause weight loss. Their whole gimmick is that the name of the product sounds like the real hormone leptin, so it will seem more science-based. In fact, these products come with instructions that you follow a very low-calorie diet, so any weight loss that occurs while following their plans is strictly due to the food restriction. If the pharmaceutical industry ever develops a drug treatment that really produces significant weight loss safely, it’ll be an FDA-approved prescription drug, not sold via an 800 number or spam e-mails.
Finally, here’s one on a more positive note. Everyone has probably heard of the “freshman 15” phenomenon, where incoming college students supposedly gain 15 or more pounds during that first year. It turns out that this does not actually happen. A recent study by Zagorsky and Smith found that the average weight gain during the first year of college is 2.5 to 3.5 lbs, and that young people of the same age who are not going to college gain about a half pound less during the same year. This small weight gain is probably more related to people at this age being on their own for the first time, and heavy alcohol drinking was the major predictor of weight gain. Those who don’t binge drink gain very little. So, worries about large weight gain when you go to college are probably not necessary.
Do you have any advice for separating the hype from the reality when reading news stories about diet and health?
This is unfortunately very tough. The media coverage on these topics falls somewhere in between slightly misleading and outright false. If you can find the original study that the report is about, that’s a start. You might not have free access to the journal where it was published, but you can actually e-mail the corresponding author and request a copy. Many journal articles will be difficult if not impossible to fully evaluate if you are not a trained scientist in the field, but you can often spot obvious omissions or false spin in the news story immediately when you compare the scientists’ own summary of the work with the news report.
If you have a good doctor or nutritionist, talk to them about this stuff. Not every doctor and nutritionist is up to date on all the science, but they can usually at least help you figure out whether the new information is credible.
What is the best evidence-based advice you can give teens about body weight, diet, and health?
Don’t smoke cigarettes. If you don’t smoke now, don’t start. If you do smoke now, quit. Many people say that they started smoking as teens in part because they believed that it would help them control their body weight. And the weight gain that often occurs when people quit is a major reason people cite for not wanting to quit. It is true that nicotine has a small but real suppressive effect on body weight, and it’s true that people who quit smoking often experience some weight gain. However, the negative effects of smoking outweigh the possible negatives of being 10 lbs heavier.
Avoid fast food and high-sugar beverages. Fast food is like a perfect storm of bad nutrition. It’s not going to kill you to eat it once in a while, but frequent consumption is clearly unhealthy. For example, one study from 2004 found that people who eat fast food meals twice per week were far more likely to develop insulin resistance (a pre-diabetic condition) than those who ate fast food less than once per week. Frequent consumption of high sugar beverages (soft drinks or juice) is associated with increased risk for cardiovascular disease and diabetes. Again, these are mostly correlational data in humans, but experiments in animals strongly support the idea that these foods and drinks are detrimental to health.
Learn to cook. I mean really cook, with fresh ingredients, not just throwing some frozen stuff into the microwave. If you can cook for yourself, then you will be able to rely far less on restaurants and pre-packaged foods, which tend to be high in fat, sugar, and sodium.
Incorporate exercise and physical activity into your daily life. Exercise protects against cardiovascular disease, stroke, hypertension, type II diabetes, and other problems. It’s also a mood lifter and may protect against depression and stress. You get these benefits whether you’re fat or thin, so it’s good for everyone.
See Dr. Williams’s response to questions here.
Disclaimer: No statements made in this interview should be taken as medical advice.
Dr. Diana Williams received her PhD from the University of Pennsylvania Department of Psychology in 2003, and trained as a postdoctoral fellow at the University of Washington School of Medicine, Department of Medicine, Division of Metabolism, Endocrinology and Nutrition, from 2003-2008. In 2008, she joined the faculty of Florida State University as an Assistant Professor in the Department of Psychology and Program in Neuroscience. Her laboratory focuses on the neural and endocrine control of food intake and body weight.
See also the references and recommended books and websites at the end of part 1.
Brownell KD, Rodin J. Medical, metabolic, and psychological effects of weight cycling. Arch Intern Med. 1994 Jun 27;154(12):1325-30. Review.
Savage JS, Birch LL. Patterns of weight control strategies predict differences in women’s 4-year weight gain. Obesity (Silver Spring). 2010 Mar;18(3):513-20.
Obesity epidemic, BMI trends, relationship between BMI and health:
Christakis NA, Fowler JH. The spread of obesity in a large social network over 32 years. N Engl J Med. 2007 Jul 26;357(4):370-9.
Flegal KM, Carroll MD, Ogden CL, Curtin LR. Prevalence and trends in obesity among US adults, 1999-2008. JAMA. 2010 Jan 20;303(3):235-41.
Flegal KM, Graubard BI, Williamson DF, Cooper RS. Reverse causation and illness-related weight loss in observational studies of body weight and mortality. Am J Epidemiol. 2011 Jan 1;173(1):1-9. Review.
Flegal KM, Graubard BI, Williamson DF, Gail MH. Excess deaths associated with underweight, overweight, and obesity. JAMA. 2005 Apr 20;293(15):1861-7.
Friedman JM. Obesity: Causes and control of excess body fat. Nature. 2009 May 21;459(7245):340-2.
Lyons R. The Spread of Evidence-Poor Medicine via Flawed Social-Network Analysis. Statistics, Politics, and Policy. 2011 2(1) DOI: 10.2202/2151-7509.1024
Puhl RM, Heuer CA. The stigma of obesity: a review and update. Obesity (Silver Spring). 2009 May;17(5):941-64.
Shiraev T, Chen H, Morris MJ. Differential effects of restricted versus unlimited high-fat feeding in rats on fat mass, plasma hormones and brain appetite regulators. J Neuroendocrinol. 2009 Jul;21(7):602-9.
Sturm R. Increases in morbid obesity in the USA: 2000-2005. Public Health. 2007 Jul;121(7):492-6.
Duffey KJ, Gordon-Larsen P, Steffen LM, Jacobs DR Jr, Popkin BM. Drinking caloric beverages increases the risk of adverse cardiometabolic outcomes in the Coronary Artery Risk Development in Young Adults (CARDIA) Study. Am J Clin Nutr. 2010 Oct;92(4):954-9.
Malik VS, Popkin BM, Bray GA, Després JP, Willett WC, Hu FB. Sugar-sweetened beverages and risk of metabolic syndrome and type 2 diabetes: a meta-analysis. Diabetes Care. 2010 Nov;33(11):2477-83. Review.
Morrow ML, Heesch KC, Dinger MK, Hull HR, Kneehans AW, Fields DA. Freshman 15: fact or fiction? Obesity (Silver Spring). 2006 Aug;14(8):1438-43.
Pereira MA, Kartashov AI, Ebbeling CB, Van Horn L, Slattery ML, Jacobs DR Jr, Ludwig DS. Fast-food habits, weight gain, and insulin resistance (the CARDIA study): 15-year prospective analysis. Lancet. 2005 Jan 1-365(9453):36-42. Erratum in: Lancet. 2005 Mar 16;365(9464):1030.
Zagorsky JL, Smith PK. The Freshman 15: A Critical Time for Obesity Intervention or Media Myth? Social Science Quarterly 2011 92(5):1389-1407.