If you like to stay abreast of the latest in nutrition research, you’ve probably heard about studies showing that dietary saturated fat intake doesn’t increase your risk of heart disease. You’ve probably also heard the phrase “take everything with a grain of salt”, meaning you shouldn’t take everything you hear at face value.
However, when it comes to your health, being a skeptic is simply not enough. Learning to think critically and understand nutrition studies will empower you to know when you’re being misled and when you should take findings seriously.
The purpose of this article is to deeply evaluate one of the biggest and most cited studies on saturated fat and heart disease. Giving yourself the ability to “cut through the fog” of nutrition studies will help you to make the best dietary choices for you and your family.
The Research in Question
Several studies concerning saturated fat/cholesterol intake and coronary heart disease (CHD) have come to the forefront of many blogs and news outlets over the past few years. The most cited ones are meta-analyses, which means researchers analyze a group of similar studies altogether to develop a singular conclusion.
Coronary heart disease (CHD) is the buildup of plaque in the arteries and is type of cardiovascular disease (CVD). Cardiovascular disease leads to heart attacks, strokes, angina, and is the number one cause of death worldwide. (source) With that in mind, information concerning this disease can literally result in life or death.
One of the most cited studies about dietary fat and cholesterol is this large meta-analysis published in the American Journal of Clinical Nutrition in 2010. A quick look at the results section of the study reveals the following:
During 5–23 y of follow-up of 347,747 subjects, 11,006 developed CHD or stroke. Intake of saturated fat was not associated with an increased risk of CHD, stroke, or CVD. The pooled relative risk estimates that compared extreme quantiles of saturated fat intake were 1.07 (95% CI: 0.96, 1.19; P = 0.22) for CHD, 0.81 (95% CI: 0.62, 1.05; P = 0.11) for stroke, and 1.00 (95% CI: 0.89, 1.11; P = 0.95) for CVD. Consideration of age, sex, and study quality did not change the results.
Upon first glance, it appears like these results are valid and reasonable. After all, it spans over a multitude of studies and evaluates lots of participants.
However, there are several reasons why these results cannot be taken at face value. Many major authorities in the field – specifically heart health experts – have expressed concerns about the study design, and therefore the results of this meta-analysis. Taking a closer look, it is easy to understand why.
Who’s Questioning These Results and Why?
In several letters to the editor of The American Journal of Clinical Nutrition (cited below), many well-respected nutrition experts expressed their concerns. Peter Scarborough and Mike Rayner from the British Heart Foundation are included in this list. The concerns they express illuminate the flaws of the research in question and are as follows:
If all this is confusing to you – don’t worry – there’s a video that below that will help to explain.
1. Serum Cholesterol Concentration
Researchers adjusted for serum cholesterol concentration in many of the studies included in the meta-analysis.
By “adjusting for serum cholesterol”, researchers basically rule out the impact that participant’s cholesterol levels have on their CVD risk. This is a problem because your it has long been established that cholesterol scores positively impact your risk for CVD. (source)
Because everyone has different starting cholesterol levels based on their genetics, even people consuming the same amount of saturated fat will end up with different cholesterol scores because they started in a different place. However, if all these individuals have rising cholesterol levels, they are increasing their risk regardless of their baseline scores. (source)
The meta-analysis in question isn’t going to find an increased risk by design because it is not considering people’s baseline cholesterol scores and observing the changes when additional saturated fat is consumed.
It is also well established that replacing saturated fats with other fats and reducing dietary cholesterol will improve blood cholesterol levels. (source) The study cited is a meta-analysis of metabolic ward experiments. A metabolic ward experiment is when participants are essentially held in a room and food intake is strictly administered over the course of the study.Compare this to an epidemiological meta-analysis (like the one in question) where researchers aren’t controlling people’s food behavior, but rather trying to observe associations based on methods such as FFQ (food frequency questionnaires) or recall surveys. This basically involves asking participants what they ate the day before. Without a doubt, the metabolic ward experiments are a much more powerful way to observe the impact of certain foods.
2. Flawed Methods
Many of the studies in the meta-analysis had extremely weak dietary assessment methods such as 24-hour dietary recall. Such methods have been proven to be unreliable and produce bad data as they don’t reflect long term dietary habits. (source)
3. Conflict of Interest
As the final piece of the letters to the editors point out, there is a major conflict of interest as the senior author has been involved in advisory activities for the dairy industry and obviously all dairy products contain saturated fat.
If you want to dive deeper into these letters to the editor they can be viewed direct from the AJCN’s website. (source)
Where Did the Recommendations to Lower Saturated Fat Intake Come From?
This video by Dr. Michael Greger of NutritionFacts.org does a great job of explaining all this in simplest terms possible. He references this exact study and breaks down in detail why it was specifically set up to fail. If you were confused up to this point, this video is likely to clear up a lot of things.
Dr. Greger compares these studies similar to the tobacco industry studies of the old days, creating confusion among consumers so that they can justify continuing to consume their products – in this case – food products that contain saturated fat.
Another Example of a Poorly Designed Study
Poor study designs aren’t only limited to epidemiological studies. Studies such as the one we discuss below on eggs and serum lipid profiles (cholesterol scores) are also extremely flawed. Again, it is important to examine cited studies closely and not take the touted results at face value.
In this study, 29 participants who had diabetes were split into two groups to test a 1 egg breakfast versus a group eating 40g of oatmeal. The study showed that there was no significant difference in the serum lipid profile of the participants and the egg breakfast also reduced certain markers of inflammation.
One might look at this and say that this means that eating eggs doesn’t have any impact on serum cholesterol and improves inflammation. However, consider this before coming to that conclusion:
One egg is only about 75 calories (source) which equates to 3-4% of a 2000 calorie diet. This study doesn’t address what else the participants ate during the trial period for the other 96% of their diet, massively diluting the entire intervention and rendering the entire thing pretty much meaningless.
In addition, this study design used 24-hour diet recalls to assess the rest of their dietary habits, which we discussed earlier.
As the proverbial icing on the cake, the oatmeal subjects were given 472 mL of milk which nearly double in calories the amount of oatmeal! This study is more about testing eggs vs. milk then it is about testing eggs vs. oatmeal.
Unsurprisingly, this study was funded by the Egg Nutrition Center.
What Does a Meaningful and Impactful Study on Saturated Fat and CHD Look Like?
Learning to think about the design flaws in a study isn’t enough – it is just as important to learn to identify a study that is meaningful and well-designed. For the purpose of this article, we’re going to take “A Way to Reverse CAD” by Caldwell B. Esselstyn Jr. This study has many features that speak to the applicability of the findings to reality.
To summarize the study: 198 patients diagnosed with CVD were given nutrition counseling. This counseling instructed them to adopt the following lifestyle:
Eliminate: Oils, fish, meat, fowl, dairy products, avocado, nuts, and excess salt, sugary drinks, excess caffeine, and fructose.
Adopt a Core Diet of: Whole grains, legumes, lentils, and whole vegetables/fruits.
Because participants were given detailed levels of counseling and tracked for adherence to that counseling it allows for a much clearer picture of results than analyzing people filling out questionnaires about food recall and food frequency. This makes the results clear and easy to assess.
So, what happened? 177 people or 89% of the patients who were adherent to their dietary counseling. Out of these 177 people only 1 or 0.6% of them experienced another cardiac event. Compare this to the people that weren’t adherent to the plan. In this group, 13 out of 21 or 62% of those patients experienced cardiac events.
Studies like this, where participants make major changes to dietary habits and researchers closely monitor the impact of those changes, are the most accurate way to assess the impact of diet on a person’s health.
The fact that 62% of the patience that didn’t adhere to the whole food plant-based diet experienced another cardiac event compared to the 0.6% of patients that did is staggering. It serves as yet another proof-point that too much saturated fat/cholesterol in the diet, especially for someone who already has heart problems, can literally be life or death.
For the newcomer, understanding nutrition research can seem overwhelming and many opt to give up and become apathetic. However, the real-world dietary interventions and metabolic ward experiments that we’ve talked about are a great starting point. Unsurprisingly, as shown by the simplicity of Dr. Esselstyn’s study, the most trustworthy study designs are the ones that ring true to your common sense and logic.
Finally, it is important to note that the biggest body concerned with heart health is unanimous on this issue. The American Heart Association (source) recommends limiting saturated fats (as long as you’re not replacing them with refined carbohydrates such as sugary cakes and cookies) and criticizes any publications saying otherwise. Given all the material that we’ve just covered, it’s easy to understand why. Be sure to check the cited source for their full statement.