Archive for the ‘Dietary Research’ Category

Triglycerides report from AHA – History Does Not Agree

April 26th, 2011

The American Heart Association (AHA) recently published a report entitled Triglycerides and Cardiovascular Disease, chronicling the rising rates of serum triglyceride levels and its role in cardiovascular disease in order to “update clinicians on the increasingly crucial role of triglycerides in the evaluation and management of CVD risk and highlight approaches aimed at minimizing the adverse public health–related consequences associated with hypertriglyceridemic states.”

This report is interesting and important for physicians to be aware of, but the major concepts are absolutely predictable with a basic understanding of serum cholesterol responses to carbohydrates in the diet. Simply stated, when people eat carbohydrates their HDL (good cholesterol) goes down and their triglycerides go up. This is uncontroversial, and so consistent that researchers use triglycerides and HDL as objective measures of carbohydrate consumption. Dr. Frank Sacks of Harvard Medical School explains in a recent paper on low carbohydrate diets that “HDL is a biomarker for dietary carbohydrate.” High triglycerides and low HDL means the subjects are eating lots of carbs. The AHA’s report confirms this as well, explaining that “very high intakes of carbohydrate (>60% of calories) is accompanied by a reduction in HDL cholesterol and a rise in triglyceride.” Perhaps the most interesting quote in the report comes in the introduction:

It is especially disconcerting that in the United States, mean triglyceride levels have risen since 1976, in concert with the growing epidemic of obesity, insulin resistance, and type 2 diabetes mellitus.

It is quite disconcerting, but it is EXACTLY what should be expected. If it is true that triglycerides increase in response to carbohydrates, then at some point around 1976, there should have been an increase in carbohydrate consumption. And there was. It was in response to the first ever Dietary Goals for the United States, issued in 1977 by the U.S. Senate Select Committee on Nutrition and Human Needs. Here are the first few recommendations:

  • Increase carbohydrate consumption to account for 55 to 60 percent of the energy (caloric) intake.
  • Reduce overall fat consumption from approximately 40 to 30 percent energy intake
  • Reduce saturated fat consumption to account for about 10 percent of total energy intake

So the recommendation to eat more carbohydrates happened almost precisely at the same time that triglyceride levels began to increase to “disconcerting” levels. The next question is if people actually followed this advice and ate more carbohydrates. And they did:

As you can see, since about 1976 carbohydrate intake increased and dietary fat intake decreased. Here is another graph of carbohydrate intake over the past 30 years (from Gross et al 2004)

As explained in the introduction this report will be of value to the Adult Treatment Panel IV (ATP IV) of the National Cholesterol Education Program (NCEP), from which evidence-based guidelines will ensue.

So what does the NCEP recommend in order to lower our triglycerides? Why nothing more than the exact recommendation we received in 1977:

Very high intakes of carbohydrate (>60% of total calories) are accompanied by a reduction in HDL cholesterol and a rise in triglyceride …. These latter responses are sometimes reduced when carbohydrate is consumed with viscous fiber …; however, it has not been demonstrated convincingly that viscous fiber can fully negate the triglyceride-raising or HDL-lowering actions of very high intakes of carbohydrates…Carbohydrate intake should be limited to 60 percent of total calories.

You can read more about triglycerides, HDL and carbohydrates here.

Try PhotoCalorie for free: Sign up for free at, or download the free iPhone app

Follow us on twitter: @PhotoCalorie, or on facebook:

Dietary Research

Sugar: the new Fat

February 14th, 2011

We’ve been hearing the same thing for the past 30 years: Fat is bad. It has more than twice as many calories as protein or carbs AND it shares the same name with the very characteristic so many of us have acquired during this obesity epidemic.

It has been the general consensus because it makes sense. If people are gaining weight by eating too many calories, then eliminating fat (the most potent calorie-contributor) from the diet should ameliorate the problem. The recommendations soon followed: “Choose lean meats; Use low fat salad dressing; Eat fat free potato chips.” After 30 years of trying, Americans – as well as the rest of the world – have not been very successful. 75% of Americans are projected to be overweight or obese by 2020.

Inadequate advice or poor compliance are the two obvious explanations for this failure. The latter has been incriminated thus far: the majority of Americans just aren’t listening. Yet much evidence suggests they may have been listening quite well.

Clinical Trial Evidence

The appearance and sudden popularity of the Atkins diet in the 1990s had dieters running to the meat department, leaving carbs in the dust. The apparent success of this diet, mostly ascertained from anecdotal evidence, had the overweight population excited and health experts worried. A diet characterized by high amounts of meat and fat was deemed impossible to be effective and a serious health risk.

At the time, few clinical trials had been done analyzing the efficacy and safety of such a diet, which understandably led to extreme skepticism among dietitians and doctors. Recent years have seen numerous studies comparing a calorie unlimited, low carbohydrate diet to various other low-fat, low calorie diets.In other words, a battle between two notorious opponents: Eat until you are full and limit carbs Vs. Eat until you reach a calorie limit and restrict fat. Since fat has 9 calories per gram and protein or carbs have 4 calories per gram, a high fat diets seem destined to fail.

Yet to the surprise of many, when compared to other diets, the calorie unrestricted, lowest carbohydrate diet group generally — but not always — loses more weight. With few exceptions, their HDL increases and their blood triglyceride levels decrease without having any significant effect on LDL (bad cholesterol). When subjects keep their carbohydrate intake lower than 50-75 grams per day, they seem to be most successful.

Often times the various groups fare the same, both losing approximately the same amount of weight. But NEVER, in dietary clinical trial history, has the low-fat, low-calorie diet group lost more weight (more).

The High-Fat Paradox

The very idea that a diet characterized by high-fat foods and unlimited calories can do as well, or better, than a low-fat, calorie-restricted diet poses a challenge to the current weight-loss recommendations. However, conceding to this evidence, and altering the recommendations would mean the advice from the last 30 years may have been premature.

Yet it seems that slowly, the anti-carb message is seeping in. Here is a recent public service announcement from the New York City Health Department:

Researchers at Harvard seem to agree.

Fat is not the problem. If Americans could eliminate sugary beverages, potatoes, white bread, pasta, white rice and sugary snacks, we would wipe out almost all the problems we have with weight and diabetes and other metabolic diseases.

-Dr. Walter Willett, chairman of the department of nutrition at the Harvard School of Public Health

The country’s big low-fat message backfired. The overemphasis on reducing fat caused the consumption of carbohydrates and sugar in our diets to soar. That shift may be linked to the biggest health problems in America today.

-Dr. Frank Hu, professor of nutrition and epidemiology at the Harvard School of Public Health.

Compliance to the high-carbohydrate recommendations

According to the center for disease control, since 1975 Americans have eaten less fat and more carbohydrates:

Perhaps it is possible that the blame does not lie entirely on the individual, due to the fact that this change in eating behavior is EXACTLY what we were asked to do:

Dietary Research, Nutrition and Weight Loss, Policy

Do Clinical Trials Support The 2010 Dietary Guidelines’ Bad Fats Recommendation?

January 17th, 2011

MyPyramid: Steps to a...more confused you?

The 2010 Dietary Guidelines suggest lowering our saturated fat intake even more. For the past 15 years, Americans have done a tremendous job of nearly achieving the current goal of no more than 10% of their calories coming from saturated fats. However, due to the less than satisfactory reduction in heart disease rates, the new guidelines suggest we should eat even less:

given that in the US population 11-12 percent of energy from SFA [saturated fatty acids] intake has remained unchanged for over 15 years, a reduction of this amount resulting in the goal of less than 7 percent energy from SFA should, if attained, have a significant public health impact.

For a 2,000 calorie diet, this means eating about 15 grams of saturated fat a day, a value that seems unattainable for the omnivore. The equivalent of a glass of milk and two 6 ounce pieces of chicken breast; or one 9 ounce piece of steak:

9 ounce ribeye

Given these rather strict limitations on saturated fat, it seems logical to assume that the clinical trials supporting this relationship are clear-cut and abundant. However, this is not the case.

Clinical Trials

Since the 1950s, there have been a relatively small number of large, long-term clinical trials examining the potential benefits of decreased saturated fats in the diet as a primary focus. All major trials (along with a study description and link to the original article) since 1966 are listed here. Some took place in mental institutions, some were not randomized, and some also involved major cocontaminant interventions such as weight loss, exercise, or increased fruit and vegetable consumption. Many show benefits to replacing saturated fats with polyunsaturated fats, while others do not.

It is therefore very difficult to come up with a straight answer, and very simple to cherry-pick research that best fits your hypothesis. If there is a benefit to decreasing saturated fat intake, increasing polyunsaturated fats, or both, the evidence is mixed at best. The larger trials seem to suggest no major effect, while others do show significant benefits.

If we were to focus on the largest (i.e. > 100 subjects), randomized, most famous trials ever done lasting longer than 1 year, we are left with very few to assess that meet the following 2 criteria:

1) The only significant intervention involved a reduction in fat and saturated fat and an increase in polyunsaturated fats
2) They ask the question: does this diet reduce heart disease? (defined as heart attacks or death from heart disease)

Listed in reverse chronological order:

Women’s Health initiative (2006) – 48,835 women, 8 years, no significant difference between intervention and control.

Diet and Reinfarction trial (1989) – 2,033 men, 2 years, no significant difference between the groups given and not given fat and fiber advice. No significant differences in ischaemic heart disease between intervention and control (intervention was only advice in this trial)

Minnesota Coronary Survey* (1989) – 4,393 men and 4,664 women, double-blind, mean 1 year, no significant reduction in cardiovascular events or total deaths from the treatment diet

Finnish Mental Hospital (1972) – 12 years, physicians not blinded, significant decrease in coronary heart disease (CHD)death in men ( 5.7 deaths /1000 person-years vs 13 deaths /1000 person-years in the control. Non-significant decrease in CHD in women. (Not randomized, although included here because this is main experiment cited in support of diet-heart hypothesis)

Los Angeles Veteran’s Trial* (1969) –  846 subjects, up to 8 years, non significant difference in primary endpoints –  sudden cardiac death or myocardial infarction. When cerebral infarcts were added, it reached significance. More non-cardiac deaths in experimental group

Oslo Heart Study (1968) – 412 men, 5 year, significant decrease in CHD with intervention. When stratified by age, the results were significant only in subjects younger than 60.

Double blind

A full list of all the trials done supporting and refuting the saturated fat-heart-disease relationship, and a more in depth description of each can be found here . There are many others that did not meet the criteria I defined above.

Extra weight loss in high saturated fat groups adds complexity

To further complicate things, the diets that are typically characterized by high amounts of saturated fats seem to result in the most weight loss. When researchers compare a calorie unlimited, low-carb, high saturated fat diet to a traditional low calorie, low-fat diet, the low carb group generally — but not always — loses more weight. With few exceptions, their good cholesterol levels go up and their triglycerides go down. Despite having an unlimited calorie budget and often consuming 3x the amount recommended saturated fats, the subjects tend to lose more weight and rarely increase their bad cholesterol levels. ( For more on this and a list of all major clinical trials, see Low Carb Diets.)

Recommending such low levels of saturated fat, primarily found in meats, may have indirect consequences. Since saturated fats are mainly found in protein-dense animal products, decreasing saturated fat intake to very low levels  by definition encourages low-protein diets, which seem to be less effective for weight loss and satiety (feeling full). Given our seemingly unyielding obesity epidemic, this may not be the best approach. Such a drastic decrease in one nutrient of our diets can lead to a large increase in another. This is exactly what has happened the past 30 years with carbohydrates. Especially refined ones:

Gross et al. 2004. American Journal of Clinical Nutrition

For a more in depth analysis of this relationship, see our page on saturated fats and heart disease in our new research library, Data Driven Dining.

Try PhotoCalorie for free: Sign up for free at, or download the free iPhone app

Follow us on twitter: @PhotoCalorie, or on facebook:

Dietary Research, Nutrition and Weight Loss, Policy

Do More Expensive Wines Taste Better? Evidence from Blind Tastings

December 20th, 2010

No! In fact it seems that the average wine drinker prefers expensive wines slightly less, than the cheap stuff. These are the results of  a study published by economists of the American Association for Wine Economists, entitled Do more expensive wines taste better? Evidence from a large sample of blind tastings.

Could this actually be possible? If people don’t see the price of the wine, they actually enjoy the taste of the cheaper wine? This is the conclusion from the more than 6,000 US blind tastings compiled by food and wine critic, Robin Goldstein, who concluded that “on average, individuals who are unaware of the price do not derive more enjoyment from more expensive wine. In fact, they enjoy more expensive wines slightly less.”

To come to this astounding conclusion, researchers used wine ranging in price from $1.65 to $150. 506 participants tasted wine flights made up of 523 different wines, presented in a double-blind manner. In other words, neither the subjects or the people serving the wine knew which wine was which. After tasting the various wines, the subjects were asked, “Overall, how do you find the wine?” The subjects could respond with: “Bad”, “Okay”, “Good”, and “Great,” to which the researchers converted to a 1-4 scale.

The results for wine connoisseurs were different. Those categorized as “experts” were able to distinguish between cheap and expensive wines, although this result was only slightly significant.

This study was recently profiled on the Freakonomics podcast, and less recently in the New York Times Freakonomics Blog. Steven Levitt did a similar experiment years ago, during his time at Harvard:

On Tuesday afternoons we had wine tastings. I asked if I could be allowed the opportunity to conduct one of these wine tastings “blind” to see what we could learn from sampling wines without first knowing what we were drinking. Everyone thought this was a great idea. So with the help of the wine steward I selected two expensive bottles from the wine cellar and then I went down the street to the liquor store and bought the cheapest bottle of wine they had made from the same type of grape.

I thus had two different expensive wines and one cheap one. I tried to make things more interesting by splitting one of the expensive bottles into two different decanters. Thus, in total the wine tasters had four wines to taste, although in reality there were only three different wines, with one sampled twice by each taster. I gave them a rating sheet and each person rated each of the four wines.

The results could not have been better for me. There was no significant difference in the rating across the four wines; the cheap wine did just as well as the expensive ones. Even more remarkable, for a given drinker, there was more variation in the rankings they gave to the two samples drawn from the same bottle than there was between any other two samples. Not only did they like the cheap wine as much as the expensive one, they were not even internally consistent in their assessments.

There was a lot of anger when I revealed the results, especially the fact that I had included the same wine twice. One eminent scholar stormed out of the room stating that he had a cold — otherwise he would have detected my sleight of hand with certainty.”

Although much less scientific than the previous experiment, Levitt’s scheme came to similar results: Most people can’t tell expensive wine from cheap, which brings a smile to college students’ faces nation-wide.

Andre´ and 3-buck-chucks really are the highest quality beverages in the world!!

Dietary Research

75% of us will be overweight by 2020 says new report

October 30th, 2010

A new report released by the Organization for Economic Cooperation and Development has shed more light on the terrifying reality of our seemingly unyielding obesity epidemic. Entitled “Obesity and the Economics of Prevention: Fit Not Fat,” a team of health economists summarized research on topics like obesity trends, intervention effectiveness, and the economics behind them.

The graph below are the current levels of obesity and predicted future figures if we continue at our current pace:

75% of Americans are supposed to be overweight by 2020. These numbers pose a terrifying threat to the world’s health and economy. Currently in America, the direct cost of medical costs of obesity are about $93 billion dollars (read more), not to mention the indirect costs of the many associated diseases.

What is going on here? Despite the billions of dollars invested in weight loss research and interventions, we can’t seem to slow this trend.

What happened in the late 70s that caused this sudden spike in obesity?  The problem with going back in time and finding an event which coincided with this trend is that you will likely find a correlation, not a cause.  Never the less, some researchers have provided some insight.

In a study on refined carbohydrate consumption in the US, researchers found a dramatic rise in carbohydrate intake in the late 1970s, and correlated it it with obesity rates:

The Center for Disease Control’s National Health and Nutrition Examination Surveys (NHANES) found a similar trend. From about 1976 to 2000, the percent of calories from carbohydrates steadily increased, while the percent from fat and protein slightly decreased.

These trends have caused many to believe carbohydrates are the cause of the obesity epidemic.

But to assume causation from a graph like the one above, would be to make a fundamental mistake in the world of statistics. It would be like a college student theorizing that sleeping with your shoes on causes headaches the following morning, when actually it was those five appletini’s the night before that did it.

We are on pace to overwhelm our healthcare system and decrease the lifespan of our entire population. Something needs to change.

Dietary Research, Nutrition and Weight Loss, Trends

Connected Health Conference 2010

October 22nd, 2010

I had the pleasure to attend the Connected Health Conference this past friday.  The title of the conference was The Way Forward: Reform’s New Focus on Health and Wellness, Independent Aging, Chronic Condition Self-Care and the Tools That Support Them and it was one of the best conferences I’ve attended in a while.  I came away with a number of new ideas to promote behavior change for PhotoCalorie – some of which we’ll roll out shortly, so stay tuned!

Here are some highlights from my notes:

BJ Fogg:

  • Words to live by “put hot triggers in front of motivated people.”  Hot triggers are anything that allows a user to take action – “click now!” vs. billboard that says “drink milk” that you see while you are driving.
  • Start small and then build off your success.  You learn how things work.
  • Formula for behavior change: Behavior = motivation * ability * trigger.  Start with triggers, then ability, and finally motivation.  Motivation is the hardest to address.

Sheena Iyengar and her book The Art of Choosing.

  • People can’t decide when they are presented with more than 7 choices (+/- 2).  They end up doing nothing.
  • Solution is to decrease the number of choices and make the choices clear and meaningful.  A successful choice results in higher user satisfaction and trust.

Kevin Volpp at U. Penn.

  • Economic incentives can be effective to induce behavioral changes but you need to be careful.  In weight loss studies, when the incentive was discontinued people gained the weight back.
  • Lots of discussion about carrot vs. stick approach.  The “stick” approach has not been studied as much mainly because it is riskier for employers since the employees are likely to complain (read: lawsuits).
  • Anticipated regret is another type of motivator.  For example, “You didn’t reach your weight loss goal for this week but if you had you would have won XXX prize.”  This creates a sense of loss and people are more likely to participate next time.

David Rose from Vitality created Glow Caps to improve medication adherence – it’s pretty clever.  Here are his slides.

ANT+ appears to be really nice protocol to connect to all kinds of wireless sensors.  Imagine having your phone communicate with your scale or a sensors on your bike communicate with your computer that track calories burned, power, and distance traveled shown on a map.  ANT’s parent company is Garmin.

Random Facts

  • 30% of people have apps but only 23% use them.
  • 1 out 10 adults didn’t know if they had an app.
  • 1 out 10 users have some type of health app (broadly defined).
  • 50+ age group is the fastest growing social network.  Apps need to be designed for this age group.
  • Obesity costs $73B in lost productivity .
  • Japan has penalties for poor health such as large waist sizes.

Behavior Change, Conference, Dietary Research, Ideas

How much beef fat did you remove before eating in High School?

October 17th, 2010

It seems like every day, a new study emerges finding a link between diet and disease:

Heavy Drinkers Outlive Non-drinkers

Low-carb diets may cause cancer?

These studies found an association between a part of the diet, and a particular disease. In other words, the first one says people who drink heavily may live longer than non drinkers, and the latter claims people who eat a low-carb diet are more likely to get cancer.

Below is a screen shot of an actual food frequency questionnaire created at Harvard University to determine high school eating behavior:

These questions were given to Nurses, asking them to recall what they ate 40 years previous. Here are some others:

  • Between the ages of 13-18, how much beef fat did you remove before eating?
  • How often, on average, during high school did you eat dark meat fish?
  • How many Yams did you eat, on average, per month?

How could someone possibly remember what they ate that long ago. And if they do remember, are we to believe these subjects are being perfectly honest, and not reporting more of the healthy foods and less of the bad ones? Fortunately, this question has been studied, and that it precisely what happens. But to be fair, most such studies don’t ask about your eating habits 40 years ago. Maybe just in the past few years.

Graph Taken from Chris Masterjohn's blog of the daily lipid.

The same person who created these questionnaires, Dr. Walter Willet, also looked at their accuracy with 173 nurses in 1980(read abstract of study here). These nurses were part of the famous Nurse’s Health Study, filling out food frequency questionnaires about their diet. Then, for 4 separate weeks out of the year, the women were given scales and asked to measure out all the foods they ate during that week, and record it in a food journal.

In the bar graph, (taken from Chris Masterjohn’s excellent blog), each food has two bars. The left most represents the FFQ filled out at the start of the year, and the right most bar is the FFQ they did at the end of the year.

It turns out that when they compared the FFQ’s report of the foods eaten to the weighed food records, the foods most accurately reported were tea and beer, while things like meat, fish, bacon, and hamburgers were at best 25% accurate. The researchers also found that the foods deemed unhealthy, like butter, whole milk or processed meats were under reported by 10-30%, while some fruits and veggies were over reported by up to 50%!

This is the level of accuracy we have in our long term studies.

These studies, sometimes involving over 100,000 people are very difficult to perform and extremely expensive. Yet despite all this money and effort, it may be difficult or impossible to confidently say the results represent actual eating behavior, and not just bias.

So take each of these ‘potential diet-disease’ relationship articles with a grain of salt, even if it is by Harvard researchers, reported by the New York Times.

Dietary Research, Nutrition and Weight Loss