Saturated Fat Back in the Gutter; Failing to see the bigger picture

Saturated Fat Back in the Gutter; Failing to See the Bigger PictureSaturated fat is the bad boy once again according to a paper recently published online in the Journal of the American College of Cardiology (1).

The authors of the paper, led by Li Yanping from the Department of Nutrition at the Harvard School of Public Health, conclude that unsaturated fatty acids, especially polyunsaturated fatty acids (PUFAs) and/or high-quality carbohydrates can be used to replace saturated fatty acids (SFAs) to reduce the risk of coronary heart disease (CHD).

Although in line with current dietary recommendations, these results contradict a systematic review and meta-analysis published last year by Chowdhury and coworkers which found no significant association between the intake of SFAs and the risk of coronary heart disease (2).

Furthermore, last February, The Scientific Report of the 2015 Dietary Guidelines Advisory Committee (DGAC) in the USA concluded that  “while the body of research linking saturated fat intake to the modulation of LDL and other circulating lipoprotein concentrations is significant, this evidence is essentially irrelevant to the question of the relationship between diet and risk for cardiovascular disease (3).”

However, things tend to change rapidly (except the guidelines) in the world of nutrition science. Now, after what appeared to be a successful rehabilitation, saturated fat is back in the gutter.

But a closer look at the Yanping paper raises questions about some more obvious conclusions not addressed by the authors.


The Yanping Study

Yanping and coworkers studied combined data from the large, observational Nurses’ Health (NHS) and Health Professionals Follow-up Studies (HPFS). Approximately 127 thousand individuals free of diabetes, cardiovascular disease, and cancer were followed for 30 years. Diet was assessed by a semiquantitative food frequency questionnaire every four years.

At baseline, those who consumed the highest amount of SFAs were slightly younger, had a higher body mass index, a lower prevalence of physical activity, were more likely to be smokers, and consumed more cholesterol. They also tended to have a higher energy intake from monounsaturated fatty acids (MUFAs) and trans fats and lower energy intake from carbohydrates.

A multivariable adjusted model was used to adjust for known risk factors for coronary heart disease, including body mass index, family history of diabetes and myocardial infarction, menopausal status and hormone therapy, regular use of aspirin, smoking status, physical activity, presence of hypertension and hypercholesterolemia, and percentage of energy from protein and cholesterol.

10283570_mA total of 7,667 cases of CAD were documented during the follow-up period.

Higher intakes of PUFAs and carbohydrates from whole grains were significantly associated with a lower risk of CHD. In contrast, carbohydrates from refined starches and added sugars were associated with higher risk of CHD. Intake of SFAs and MUFAs was not significantly associated with the risk of CHD in the multivariate model.

Replacing 5% of energy intake from SFAs with equivalent energy intake from PUFAs, MUFAs or carbohydrates from whole grains was associated with a 25%, 15% and 9% lower risk of CHD respectively. Replacing SFAs with trans fat or carbohydrates from refined starches and added sugars did not significantly affect the risk of CHD.

Interestingly, participants generally replaced calories from SFAs with calories from low-quality carbohydrates rather than calories from PUFAs or high-quality carbohydrates.

The authors conclude that their findings “provide epidemiological evidence of the current dietary guidelines, which recommend both replacing saturated fatty acids with monounsaturated and polyunsaturated fatty acids and replacing refined grains with whole grains. 

They claim that consuming less SFAs will lower the risk of coronary heart disease as long as they are replaced with PUFAs or high-quality carbohydrates such as whole grains but not if they are replaced with trans fats or low-quality carbohydrates such as white bread, white rice, or potatoes.

A Closer Look at the Data

The study participants were divided into quintile categories (five groups of equal size), based on their nutritional intakes. So, for each macronutrient, there were five groups. If we take SFAs for example, the lowest quartile is the group (20% of the study population) with the lowest intake of SFAs, and the highest quartile is the group (20% of the study population) with the highest intake of SFAs.

Macronutr

There were 16% more cases of CHD in the group with the lowest fat intake (ca. 26% of calories) compared with the group with the highest intake (ca. 41% of calories).

Similarly, there were 12% more cases of CHD in the group with the lowest intake (ca. 9% of calories) of SFAs compared to the group with the highest intake (ca. 15% of calories).

Remember that, although slightly younger, the group who consumed most SFAs had a higher body mass index, a lower prevalence of physical activity, were more likely to be smokers, and consumed more cholesterol.

The group with the highest (ca. 56% of calories) carbohydrate consumption had 25% more cases of CHD than the group with the lowest (ca 37% of calories) carbohydrate consumption.

These numbers are not dealt with by the authors of the paper.

Saturated Fat – The Bigger Picture

Surely the study by Yanping and coworkers will encourage authors of dietary guidelines to stick with their previous recommendations highlighting restriction of SFAs and that they should be replaced with PUFAs and high-quality carbohydrates.

However, it is important to understand that, due to the study’s observational nature, it cannot prove causality. Despite the use of complex statistical methods, it is difficult to rule out confounding factors that may affect the results.

Furthermore, there are many types of SFAs with different metabolic effects and their association with CHD varies. For example, while palmitic acid appears to be associated with adverse metabolic effects, other SFAs may be associated with metabolic benefits (4).

Dietary SFAs are derived from many different foods, and these have many other ingredients and characteristics that modify their health effects. Therefore judging the health effects of foods entirely from the amount of SFAs may be dreadfully misleading.

When looking more closely at the data it appears that individuals with the highest intake of dietary fats have a lower risk of CHD than those with the lowest fat consumption.

Furthermore, those with the highest carbohydrate intake appear to have a higher risk of CHD than those with the lowest carbohydrate consumption.

Despite my deep respect for the use of statistical methods to adjust for confounding factors, I find it hard to see how statistical acrobatics might reverse this picture.

So, the numbers appear to suggest that; less fats-more heart disease, more fats-less heart disease and more carbs-more heart disease, less carbs-less heart disease. Yes, I know; correlation does not prove causation but…

I suspect the Yanping study may convince many of us that choosing PUFAs rather than SFAs and choosing whole grains rather than starches and added sugar is preferable. But what’s most important about this paper are the questions raised but regrettably left unanswered by the authors.

The lights are on, but nobody’s home.



36
Leave a Reply

avatar
14 Comment threads
22 Thread replies
0 Followers
 
Most reacted comment
Hottest comment thread
16 Comment authors
Carroll H Hoagland IIICarolMieFromPAAnn Childers Recent comment authors

This site uses Akismet to reduce spam. Learn how your comment data is processed.

  Subscribe  
newest oldest most voted
Notify of
Will
Guest
Will

Damned if ya do, damned if ya don’t!

David_Brown
Guest
David_Brown

You’re more likely damned if you do. Do what? Follow the advice of the Dietary Guidelines Advisory Committee and Harvard epidemiologists to swap solvent extracted seed oils for saturated fats. (1) The problem is, there are deleterious effects associated with high PUFA intake that have nothing to do with heart attack risk. For example: “Over recent decades, more and more of the oil in the typical American diet has come first from corn, and then from soybeans. In both cases, that oil is an unusually concentrated source of omega-6, linoleic acid. The direct metabolic implication of this is immediately clear:… Read more »

Leah
Guest
Leah

Great article! I am a Nutrition Diagnostics Master’s student at Cox College, and this is the information that my cohort and I are just beginning to learn. It has taken my mind by storm. The “saturated fats are bad, whole grain carbohydrates are good” mantra is one that is still being taught in almost every undergraduate nutrition program in the country. It is going to be a long road ahead to try to forge these new ideas, especially with researchers suffering from growing pains. Thanks for the comprehensive report!

Axel F Sigurdsson
Guest
Axel F Sigurdsson

Thanks Leah.
I think you’re right. There’s a long road ahead.

rubagreta
Guest
rubagreta

Go for it, Leah!

Ashkar Lessirey
Guest
Ashkar Lessirey

Nice catch 🙂

Axel F Sigurdsson
Guest
Axel F Sigurdsson

Thanks Ashkar.

George
Guest
George

Thanks for extracting that data for us. There’s one thing I didn’t understand. The HRs for SFA and MUFA across the quintiles were almost exactly the same, with slight advantage to SFA, yet they predict benefit from replacing SFA with MUFA.
How is this possible?

Axel F Sigurdsson
Guest
Axel F Sigurdsson

George.
They perform some type of substitution analysis. The statistical approach is pretty difficult to understand; “When estimating the effect of substituting 1 type of fat or 1 source of carbohydrate for another, we included energy contributions from the 2 nutrient types as continuous variables in the same multivariable adjusted model. The differences in their coefficients and covariance were used to estimate the HR and 95% CI of the substitution.”
Still trying to figure it out 🙂

George
Guest
George

Even a stage magician needs to show there’s nothing up his sleeve. We’re not being shown that here, it has the appearance of a “proprietary formula”. Another thing that bothers me – among many – there simply isn’t a 5%E difference between upper and lower quintile in most cases. And even if there were, a projection of benefit from the 5%E increase based on the real data could only apply to the lowest quintile. But they’re essentially telling people in every quintile of PUFA intake to increase PUFA by 5%E. Another issue – the exclusion criteria for plausibility of diet… Read more »

bob_johnston
Guest
bob_johnston

The real problem to me is that science today in some disciplines is so poor that a person has no idea what to believe, even if one does dive into the study an tear it apart. Peer review is no panacea to fixing the system either. We know that a good portion of studies today are fraudulent or biased and as long as there’s no effort made by the journals to fix this situation I think it wise to believe that everything being published is suspect. Because my distrust has grown so strong in the system I think my rule… Read more »

Peter T Capozzoli
Guest
Peter T Capozzoli

Wait a minute, There were 127,000 people in the study, followed over 30 years and there were only 7,667 that had CAD on follow-up? … that’s only 6% with CAD. Is there any real ABSOLUTE differences in the CAD risk based on diet? 25% or 9% of 6% does not move ABSOLUTE risk that much. Oh and, if the saturated fat group was younger, doesn’t that skew results (bad habits or not)? How young were these people? It looks like they are saying the optimum fat intake in % of diet is So could someone break down the correct %’s… Read more »

rubagreta
Guest
rubagreta

Let’s see. I went on a higher-fat (with lots of eggs), low-junk carb (grains), low-refined sugar, no-fruit juice regimen (it’s not a diet because I eat when and what I want), and dropped 20 pounds with no effort. And so have thousands of others.

And as an added benefit, I know longer take a statin.

I call Bullshit.

Will
Guest
Will

You may be confusing weight loss with health outcome.

rubagreta
Guest
rubagreta

Don’t they go hand-in-hand?

George
Guest
George

Not needing a statin indicates that health markers have improved, either that or that intelligence has increased.

Will
Guest
Will

Firstly, congrats on being off statins, that’s a great achievement.

Re my comment, simply meant that we can lose weight a number of ways…doesn’t mean it’s always healthy. You can google fellow who lost weight by just eating twinkies but prob not the best way to do it. If the weight loss you’ve experienced also works with your lipid profile then you’re on the right track, just my two cents.

Best of luck

Ann Childers
Guest
Ann Childers

Agree about lipid profile. Triglyceride to HDL ratio is telling and may reflect propensity for large particle LDL (a good thing). 2:1 or less is great. Dr. Haub of KSU is the twinkie guy. He also exercised like crazy and drank Muscle Milk, facts he underemphasizes. So we don’t really know what Twinkies and other junk foods contributed in his case; for his health’s sake I’m glad he didn’t go full twinkie jacket. Apart from jounalistic sensationalism there isn’t much of a case report there. We know high carb, low calorie diets can result in weight loss. Any teenager who’s… Read more »

Christopher Palmer
Guest
Christopher Palmer

The basic problem with the Yanping analysis and the many others of the ilk is that epidemiological dietary studies are fraught with prospects of confounding factors. Indeed poor health outcomes and CAD risk may well be influenced by the composition of the diet. Thereore it is not wrong to study diet. But factors other than diet may also have influence. If the diet is studied to the exclusion of concern for other non-dietary influential factors then analyses and conclusions could well be misleading. There also exists a great weakness in implicating cause by association, where any causal pathway may be… Read more »

George
Guest
George

Interesting thing about the cholesterol-fed rabbit model of CAD is that MUFA often comes off worse than SFA, and butter high in CLA is protective, as are choline and chondroitin.

Mie
Guest
Mie

What’s truly interesting is people who talk about animal feeding studies when we have ample evidence from human studies (metabolic ward) of the effects of fatty acids.

Less SAFA & sugar/junk carbs –> better health. Simplified, yes, but still valid.

FromPA
Guest
FromPA

Could you cite a study where they reduced saturated fatty acids and that resulted in lower heart disease (or “better health” in any way)? In the meantime, here’s my study: The women’s health initiative (diet arm), where they took 59,000 women and randomized them into two groups. One group was actively counseled to eat less overall fat, less saturated fat, more fruits and vegetables, etc., and they did (and reduced that “evil” red meat too, by a statistically significant amount). After 8 years and half a billion dollars, there was no statistically significant difference between the two groups in heart… Read more »

Mie
Guest
Mie

“Could you cite a study where they reduced saturated fatty acids and that resulted in lower heart disease (or “better health” in any way)?” If by “reduced” you mean “exchanged safa for pufa”, then e.g. LA Veterans. Mere reduction means nothing because the calories still come from somewhere – and that “somewhere” is the key here. Safa –> Western carbs = not a good idea. About WHI: the study goes to show that people cannot make the required changes and/or that the changes they make aren’t what they should be (e.g. reducing total fat isn’t the way to go). What… Read more »

Mie
Guest
Mie

Axel, you stated:

“I suspect the Yanping study may convince many of us that choosing PUFAs rather than SFAs and choosing whole grains rather than starches and added sugar is preferable.”

The wording is a bit funny/awkward. Did you mean to imply that you have a problem with this recommendation?

BTW, I find it … somewhat disingenuous that you continue referring to Chowdbury et al without acknowledging the problems inherent in it. Why do you allow comments here if you choose to ignore/not to reply the points made by the commenters?

Axel F Sigurdsson
Guest
Axel F Sigurdsson

Thanks for the comment MIe.
No, I don’t have a problem with the “choosing PUFAs rather than SFAs” recommendation. I think it’s valid for individuals with atherosclerotic heart disease and as a general recommendation. However, I don’t think it’s of key importance, to some extent misleading, and that we should focus on foods rather than macronutrients.
Many (most) systematic reviews have inherent problems and I don’t doubt that Chowdhury’s one of them. However, they can still be cited or referred to.
Finally, I never ignore comments, but due to lack of time I’m not able to respond to all of them.

Mie
Guest
Mie

“However, I don’t think it’s of key importance, to some extent misleading, and that we should focus on foods rather than macronutrients.” Could you clarify the first two a bit, please? I agree with on the third one. (… which is, BTW, precisely the direction where the current recommendations – at least Nordic/Finnish guidelines – have been heading for quite some time. However, if you focus on entire foods – or better yet, DIETS – you more or less end up in a situation where both SAFA and sugar intake get minimized.) “Many (most) systematic reviews have inherent problems and… Read more »

Axel F Sigurdsson
Guest
Axel F Sigurdsson

Mie. I believe the evidence suggesting replacing SFAs with PUFAs is weak which is one of the reason I don’t consider this recommendation to be of key importance. Furthermore, PUFAs are heterogenous. Some, like EPA and DHA may have more benefits than other. SFA’s are also heterogenous, some appear to have positive beneficial effects while others don’t. This is one of the reasons I use the word misleading. The evidence that SFAs lower triglyceride-rich lipoproteins and lipoprotein(a), and the lack of evidence for a relationship between SFA intake and coronary heart disease, and no proven harm for most major food… Read more »

Mie
Guest
Mie

“I believe the evidence suggesting replacing SFAs with PUFAs is weak ” And yet you continue to utter statements which are much less supported by evidence than the abovementioned (e.g. in this very message of yours). Selective, at best. “Furthermore, PUFAs are heterogenous. Some, like EPA and DHA may have more benefits than others.” … which (heterogenuity) is a moot point as there’s evidence of both n-3 and n-6 fatty acids being beneficial when consumed within intakes recommended in (virtually all) the guidelines. “SFAs are also heterogenous, some appear to have beneficial metabolic effects while others don’t.” You’re referring to… Read more »

Axel F Sigurdsson
Guest
Axel F Sigurdsson

Thanks Mie.
Don’t know what I’d do without you 🙂
But, I never said that we should increase SFA intake.
And, I always try to choose fish, nuts, vegetables and olive oil rather than processed meat and junk carbs, I think everybody should. But, I don’t think it has anything to do with the amount of saturated fats. There are other measures to assess whether a food is healthy or not. And I love some butter and cream in between.
By the way. Do you believe that all studies that don’t support your opinion are flawed?
And I think his name is Chowdhury, not Chowdbury.

Mie
Guest
Mie

“Thanks Mie. Don’t know what I’d do without you :)” Ignore the key points in someone else’s posts? How about you try a little something called “argumentation”? 🙂 “But, I never said that we should increase SFA intake.” If SAFA is harmless – as you claim – why shouldn’t we? If there’s next to no evidence for the benefits of fat exchange – as you claim – why shouldn’t we? Or how about you define “key importance”: does fat exchange (SAFA for PUFA) matter or not? “And, I always try to choose fish, nuts, vegetables and olive oil rather than… Read more »

Peter T Capozzoli
Guest
Peter T Capozzoli

https://money.cnn.com/2015/10/26/news/red-meat-processed-cancer-world-health-organization/
Meat bad, Vegetable good. Gorg not eat dinosaur any more… eat green stuff.

Auggiedoggy
Guest
Auggiedoggy

From years of trial and error I have determined that I do far better on a low saturated fat diet. This article won’t change my mind one iota.

Auggiedoggy
Guest
Axel F Sigurdsson
Guest
Axel F Sigurdsson
Carol
Guest
Carol

Is cheese a good source of saturated fats?

Carroll H Hoagland III
Guest
Carroll H Hoagland III

Correct, I have watched Harvard Medical go down the tube … with many doctors being paid by BigPharma and BigSugar, just really hard to get info, but some is coming out. … and this is no diff, as I do not believe much of what comes out of Harvard anymore.

There is much more positive evidence coming out of LCHF dieting, since “There is NO known Carbohydrate Deficiency Disease”, and no proof that we need any carbs.

70 Going On 100

Tweet
Pin
Share274
Share12