For more than five decades we’ve been brainwashed to believe that saturated fat causes heart disease. It’s such a deeply ingrained belief that few people even question it. It’s just part of our culture now.
Almost every day I read or hear about someone proudly that they have a “healthy” diet because they don’t eat butter, cheese or red meat or any other foods high in saturated fat (nevermind that red meat isn’t particularly high in saturated fat, but that’s a subject for another post). Or I might overhear someone at the grocery store saying how much they prefer whole fat yogurt to the low-fat version, but they eat the low-fat stuff anyways because they want to make the “healthy” choice.
What most people don’t realize is that it took many years to convince people that eating traditional, animal fats like butter and cheese is bad for you, while eating highly-processed, industrial vegetable oils like corn and soybean oil is good for you. This simply defied common sense for most people. But the relentless, widespread campaign to discredit saturated fat and promote industrial oils was eventually successful.
Well, that’s what I’m telling you. We’ve been duped. Lied to. And we’ve suffered greatly as a result. Not only have we suffered from being encouraged to eat packaged and processed foods made with cheap, tasteless vegetable oils and refined carbohydrates (low-fat cuisine), but these very foods we were told would protect us from heart disease actually promote it!
The recent review I’m talking about is a meta-analysis published this week in the American Journal of Clinical Nutrition. It pooled together data from 21 unique studies that included almost 350,000 people, about 11,000 of whom developed cardiovascular disease (CVD), tracked for an average of 14 years, and concluded that there is no relationship between the intake of saturated fat and the incidence of heart disease or stroke.
Let me put that in layperson’s terms for you:
Eating saturated fat doesn’t cause heart disease.
Like what you’re reading? Get my free newsletter, recipes, eBooks, product recommendations, and more!
There. That’s it. That’s really all you need to know. But if you’d like to read more about it, John Briffa and Chris Masterjohn have written articles about it here and here.
I wonder how long it will take for this information to trickle down into the mainstream culture? Unfortunately it’s not going to happen overnight. Paradigm shifts don’t work that way. But I’ve seen some positive signs, and I do believe the tide is turning. Let’s hope it doesn’t take another 50 years.
To read more about heart disease and cholesterol, check out the special report page.
Sign up for a free e-mail series
Debunking 5 common (but dangerous) myths about cholesterol that could be putting you at risk.
Better supplementation. Fewer supplements.
Close the nutrient gap to feel and perform your best.
A daily stack of supplements designed to meet your most critical needs.
There are so many comments addressing the Okinawan diet without stating the date line. Like most of the world, diet is changing toward the Western Diet.
The diet that has been studied, and the one I am commenting on, is what brought so many people today to over 100 years old.
The Centurions, primarily female, did not eat much pork or any other meat. It was only consumed on holidays or special occasions. Sweet potatoes are primarily what they survived on till some years after WWII. As the economy improved they had more rice, tofu, natto and as always had the vegetables from their home gardens. Over 80% of their daily calories were from whole plant-based foods and the number of calories were about 10% less than the Japanese (Okinawa became part of Japan in the early 70s) which is less than the Western diet. So this group of Okinawans were somewhere around 1800 daily calories.
The only reason to discuss the diet that the young Okinawans are now consuming is to also mention that the level of disease has also increased in the young group.
Chris. This is somewhat of a weird article. You seem to be denying that saturated fats contribute to heart disease. Whilst at the same time you are trying to educate people about the importance of lowering their LDL-P to lower their risk of heart disease risk.
From where I am standing, I see your advice, and it doesn’t help many people on the internet who are trying to find solutions to their long term health problems. It’s like you’re intentionally confusing people about the issue of heart disease.
I’m assuming that you know about all the studies that show that LDL-p and APo-B are raised by the increased saturated fat that many paleo opportunists such as yourself recommend?
LDL is NOT the cause of cardiovascular disease.
ONLY Oxidized LDL is a problem. And LDL_P are LDL particles
To prevent oxidizing LDL particles –Eliminate SUGAR.
SUGAR is the main culprit for lots of diseases lie cardiovascular and diabetes. Just stop eating sugar–as much as you can.
Every time you eat sugar you are oxidizing your LDL-P particles and making it dangerous.
LDL-P is not dangerous in itself –as it carries cholesterol which the body and the brain desperately needs. You do NOT want to get rid of LDL-P. You just want to stop oxidizing it. So stop pigging out on sugar!!!
Blaming cholesterol and LDL for Cardiovascular disease is like blaming food for people getting sick by eating spoiled food.
It appears to me that the Okinawan diet for many changed around 1960 and so what is being suggested as a diet to follow ranges from the 1600s to 1960.
Chris, do you have any thoughts on the criticisms of this study by representatives from the Harvard School of Public Health that it is misleading a number reasons: http://www.hsph.harvard.edu/nutritionsource/2014/03/19/dietary-fat-and-heart-disease-study-is-seriously-misleading/
Apparently, even of of the study’s authors thinks that the results recommend fish and vegetable oils over saturated fats. http://news.sciencemag.org/health/2014/03/scientists-fix-errors-controversial-paper-about-saturated-fats
Thoughts?
Interesting discussion on saturated fat. Hi Jake, what is your thoughts about information from this book?
“The Big Fat Surprise. Why Butter, Meat and Cheese Belong in a Healthy Diet – Nina Teicholz”
Regarding Harvard and other universities—I have been disappointed in them in that they have NOT been embracing the not-so-recent debugging of various “theories’ such as the dietary cholesterol theory–which says eating cholesterol raises blood cholesterol.
But I KNEW what their plan was. They would keep supporting the false theories–but then little-by-little they would start dropping them–so that no one would be able to say that they are promoting false theories.
Surely enough–we are NOW reading that leading cardiology researchers are saying stuff like “serum cholesterol is not as much causal in cardiovascular diseases as previously thought.” 🙂 WE have known for along time that inflammation is primarily causal. So little by little they will drop the BS theories without the public realizing that they had been misinforming them for decades.
Let’s wait for the next back-tracking statement from the “experts” regarding saturated fats. 🙁
I would like to know who paid for this “research”.
The reason heart disease is hardly found in certain places in the world like Okinawa is because they eat very little animal flesh and products. There are plenty of studies that show that eating animals increases your risk of heart disease and many other diseases. To think otherwise is insanity. Watch videos by Greger, M.D., or read the work or Ornish and McDougal.
Incidentally, only one diet has reversed heart disease. Just one, and it certainly didn’t involve animal products. See “Prevent and Reverse Heart Disease” by Caldwell Esselstyn, M.D.
Stating that Okinawan people eat little animal products is simply untrue!
The diet of the people of Okinawa consists mainly of pork(literally every part of the animal), not so much fish as you would expect for an island nation, but always combined with fresh vegetables such as bittermelon Personally, I think that the latter is preventing them from getting cardiovascular diseases-or at least it is a factor that definitely helps..
You are sadly mis-informed about the Okinawan diet that produced such a high percentage of people (Primarily women) to live over 100 years with clean arteries.
The diet that is important existed from the 1600s to around 1960. It is true that many of the younger people do not follow the old ways but it is also showing up in increased levels of disease.
Pork was consumed primarily on holidays in the original diet. Sweet potatoes made up a very large percent of daily calories which by the way were fewer than in Japan. The old Okinawans averaged a couple of hundred fewer calories on a daily basis than in Japan.
The dairy industry funded the study…
Of course they did. I feel like with all the info on healthy plant based diets, those powerful industries are going to be fighting back hard to get people on their side again. Too much money to be lost.
I am not nearly as interested in reading about what does not cause heart disease as I am to know what does cause it…
Looking at the cultures that do have low rates of heart disease I find that they eat less meat, little if any dairy, not many eggs but heavy usually on sweet potatoes or whole grains. Needless to say they have almost no processed foods…
Societies that have low heart disease eat lots of veggies—which are anti-oxidants–especially veggies like kale.
Use of spices like Turmeric, basil, oregano are also anti-oxidants.
Anti-oxidants prevent LDL from becoming oxidized and it is OXIDIZED LDL that is a problem—NOT ordinary LDL.
Why don’t the medical meat-heads, our doctors, tell us that??
Because then they would not be able to write prescriptions for Statins and they would not make as much money. 🙁
As it is often said: “Follow the money!”
I understand you people like to think in terms of stats and all…. but then that is science.
What I do not read here is the main problem with fats….
Canola oil seems to be the rage for being less saturated fat but we know the proportion of omega fats is damaging.
We know that commercial fats are damaging due to tooooo much omega 6 overall.
We also know that olive oils are mostly fake from Italy so cannot be trusted.
I buy one brand of coconut oil and a different brand of olive oil, both known to be totally organic.
Other than that, nuts, avocados, chia seeds, are basically unprocessed sources of fats with the added benefits of protein.
We need to eat fat in our diets…. for nutrients, taste and I believe it is essential for the cells. It is really a mess out there but it can be sorted by the lay person.
I hate the processed world of fats as this is the main issue.
Dianne, while I believe there is likely some truth to your emotional, rather than scientific, belief in organic and such there is no evidence to support what you say. It is based on what you believe not what you know. Again, I would like to believe and even use my purchasing power to get the purest least processed oils and such that I can there is still no evidence to support the arguments that it makes any difference.
Everything Diane said does line up with the science so i don’t get the criticism. We do need fats, this is proven and they are best from natural, least processed, sources.
There is no question but that at least half of the EVO is fake…just google fake olive oil!!!
The guy who runs the Plant Positive website did a pretty interesting breakdown of this meta-analysis
http://plantpositive.com/blog/2012/3/26/tpns-52-54-the-best-low-carb-research-money-can-buy.html (starts at 6:42 in the first video and continues into the next few videos further down the page)
Apparently when you actually read the studies that make up the meta-analysis, they aren’t very favorable to saturated fat
PP Link: None of the studies have been very positive when it comes to saturated fats. The problem is when the “meta analysis” is performed and it does NOT show “statistical significance”. I am not certain what your link argues or demonstrates but this study and most studies either show a negative impact on high levels of intake on saturated fats or a “trend” towards bad outcomes (this study included). Even though it doesn’t reach ‘statistical significance’ if you were a betting person you would not bet on saturated fats.
Here is some other food for thought. The high fat diets that help people stay in ketosis to avoid seizures tend to have negative effects on those persons, which is why they are not a “goto” treatment but rather a last option. High fat diets lead to a HUGE increase in kidney stones, slower growth in children, higher rates of bone fracture among others. The higher rates of bone fractures is now understood to be a negative effect of an increase in saturated fats in the bone. Same is true for obese diabetics who have higher bone densities but an increased risk for fractures.
I’d suggest to stay away from this “plant positive” guy, because he does not know what he is talking about and has a serious double standard.
From the link he cites Stamler’s critique. Most of his criticisms were irrelevant to the actual meta-analysis, and those that were relevant (dietary methods, over-adjustment, endpoint) were already addressed by Siri-Tarino:
http://ajcn.nutrition.org/content/92/2/460.full
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2904039/
http://www.ncbi.nlm.nih.gov/pubmed/20711693
Next, PP claims that “One concern was that the authors did not distinguish between carbohydrates of different quality”. However, the authors clearly stated that there was insufficient statistical power to assess this question.
There was a study that tried to answer this question which found that carbohydrates with high-GI in place of saturated fat was associated with a statistically significant increased risk of myocardial infarction, but no significant associations for medium-GI or low-GI carbohydrates in place of saturated fat – http://www.ncbi.nlm.nih.gov/pubmed/20375186
In other words not only is there no independent association between safa and CVD, there is also no credible evidence that replacing safa with anything else is beneficial. Furthermore, the replacement argument is irrelevant since most people tend to eat a mix of fats or eat saturated fat WITH quality carbohydrates. It’s not an either-or situation.
Z.M.: First it is important not to cherry pick the information presented. The meta analysis on saturated fats that we are all speaking of definitely showed a signal (as just about every trial regarding saturated fats has shown) for increased risk with diets high in saturated fatty acids. However, it did not reach statistical significance.
Regarding the cherry picking. You have selected an article presented in the american journal of clinical nutrition, which is notorious for presenting information strongly supported by industry. In fact all three of the articles you presented are by the same author. Here are Krauss (one of the main authors) disclosures “RMK receives research support from the National Dairy Council, National Cattleman’s Beef Association” are you surprised that they try to put a spin on SAFA are good? Let us assume that we accept the articles and commentaries that YOU have presented. Here is what they conclude “Epidemiologic studies and randomized clinical trials have provided consistent evidence that replacing saturated fat with polyunsaturated fat, but not carbohydrates, is beneficial for coronary heart disease. Therefore, dietary recommendations should emphasize substitution of polyunsaturated fat and minimally processed grains for saturated fat.” However, they also point out that substituting carbs with low glycemic index in place of saturated fatty acids is also likely good. This coming from persons with clear incentive to say SAFA’s are good.
Let me state that the evidence is very clear that replacing saturated fatty acids with PUFA’s has proven time and time and time again to be beneficial.
Below are is a link with a snippet from it below the link.
http://www.hsph.harvard.edu/news/press-releases/saturated-fat-polyunsaturated-fat-cut-heart-disease-risk/
This link finds “A new study by researchers at Harvard School of Public Health (HSPH) provides the first conclusive evidence from randomized clinical trials that people who replace saturated fat in their diet with polyunsaturated fat reduce their risk of coronary heart disease by 19%, compared with control groups of people who do not.”
Take a look at current populations who have longevity. They are NOT populations with high saturated fat diets. The diets of the past are fun for conjecture, but there is much debate regarding what they at and in what proportions. We know that ancient man had heart disease, this is not new (look at the HORUS study if you not clear on this point). It seems that atherosclerosis is part of aging. Diet certainly matters, but man currently lives longer than at any time in history, but when you look at populations who live longest they tend to eat little meat, lots of veges and are much more active.
For more info take a look at the seventh day adventist study with regards to diet. Veges/fish win out every time. This study was important because of the reliability of the diets they followed, unlike most diets which rely on questionable forms.
Casey: ” However, they also point out that substituting carbs with low glycemic index in place of saturated fatty acids is also likely good. ”
Casey, the study they cited in support of this found insufficient evidence for it. I repeat, the study found that carbohydrates with high-GI in place of saturated fat was associated with a statistically significant increased risk of myocardial infarction, but NO significant associations for medium-GI or low-GI carbohydrates in place of saturated fat – http://www.ncbi.nlm.nih.gov/pubmed/20375186
Casey: “Let me state that the evidence is very clear that replacing saturated fatty acids with PUFA’s has proven time and time and time again to be beneficial.”
It is not clear at all. Check out my review on the topic to save me from a lengthy post – http://diettrialclaims.blogspot.com/2014/07/the-evidence-on-saturated-fat.html
Z.M. I am well aware that replacing saturated fats with polyunsaturated fats is “unclear” to you. Your blog is poorly analyzed and omits too many thing or glosses over evidence. You even mention that there were findings that replacing SFA with POLYs might be good but you try to dismiss them. None of your posts answers any of my statements above.
You cite O’Sullivan but you don’t seem to mention that he found a SIGNIFICANT increase in the risk of mortality with increased meat consumption. What about the post commentary by Skeaff and miller.
“to examine the ef- fect on CHD death and CHD events of replacing SFA with MUFA, PUFA or carbohydrate. The main finding was a significantly decreased risk of CHD death and CHD events when PUFA replaces SFA. The multivariate-adjust- ed hazard ratio for CHD death per 5% TE incremental substitution of PUFA for SFA was 0.87 (95% CI 0.77–0.97); for CHD events, the hazard ratio for the same fat substitu- tion was 0.74 (95% CI 0.61–0.89). This result from the pooling of observational studies, along with supportive evidence from clinical trials of lower CHD risk in high P/S diets, and the effects of PUFA to lower LDL choles- terol and the total:high-density lipoprotein ratio, led the Consultation to conclude there was convincing evidence of lower CHD risk when PUFA replaces SFA”
The bottom line is
1) it is very clear that replacing SFA with PUFA’s is beneficial (or if you don’t like such a firm statement…….the evidence STRONGLY suggest replacing SFA with PUFAs is beneficial.
2) even if you don’t believe that SFA are that harmful, there is no evidence to suggest they are beneficial. Saying they aren’t horrible is NOT the same thing as saying they are good for you.
3) epidemiological evidence demonstrates clearly that regions with extreme longevity tend to have a primarily plant based diet with occasional meat, and lots of exercise. Time to provide any evidence of longevity in populations with high saturated fat diets.
4) take a look at the 7th day adventist study. This is one of the more reliable studies regarding diet because it is not a simple questionnaire taken on one day. The reliability and validity of this study is more powerful than many. They found that 1)not smoking 2) eating a plant based diet 3)eating nuts several times per week 4) regular exercise 5) maintaining normal body weight gave them an average life span of 7 to 10 years beyond the average. Even more if you consider quality of life years.
The evidence is clear. Smoke, mirrors and misunderstanding or misinterpretation of data can’t change the facts.
Casey: “You even mention that there were findings that replacing SFA with POLYs might be good but you try to dismiss them.”
Not exactly, it is more specific than that. The best evidence for PUFA is specifically fish/marine fatty acids. Strong conclusions are not possible even for this for reasons stated on my blog.
Casey: “You cite O’Sullivan but you don’t seem to mention that he found a SIGNIFICANT increase in the risk of mortality with increased meat consumption”
Because I’m evaluating saturated fat, not meat consumption, which is another topic.
Casey: “The main finding was a significantly decreased risk of CHD death and CHD events when PUFA replaces SFA.”
Look at the actual evidence in their paper:
1) Intake of PUFA and linoleic acid was associated with increased risks of CHD mortality, but this depended on the type of analysis and the risks were small.
2) No associations were found for ALA, but long chain n-3 PUFA intake or fish consumption was associated with a lower risk of CHD death.
3) For trials, no significant effects on CHD death and total mortality were found with either lower-fat diets or high P/S diets. This was despite the fact that the analysis included trials that employed multiple interventions disadvantaging the saturated fat groups, and included a trial with serious methodological problems, which should have been excluded.
4) Long chain n-3 PUFA intake or fish consumption was associated was a decreased risk of CHD events, CHD death, and all-cause mortality, but only after exclusion of DART 2, a trial with serious methodological problems.
As I concluded, saturated fat, ALA, LA, and total PUFA showed no convincing relations with risk. Long chain n-3 PUFA intake or fish consumption may reduce risk. In other words, the strongest evidence from their paper is for marine fatty acids specifically.
Skeaff and Miller is only one review, and you have to put their findings in context with everything else. Taking everything into consideration, it is not possible to come to strong conclusions on fatty acids in general.
“Not exactly, it is more specific than that. The best evidence for PUFA is specifically fish/marine fatty acids. Strong conclusions are not possible even for this for reasons stated on my blog.”
Yup exactly, your blog is a poor reference. I should remind you that just because YOU “ZM” seem to have interpreted data in a very skewed manner and not very objectively hardly is a point for reference. Your first post in this discussion provided 3 references (were they your best?) that as I mentioned refute what you say (please refer back). Now, regarding POLYs and Omega 3’s. 2 points.
1st: lets assume that your belief is true (it is not) and we aren’t certain whether it is the non omega-3 PUFA or the total consumption of POLYs. Lets say the data isn’t clear which (but it is clear) then your argument should be. “while it isn’t clear that PUFA’s in general without excluding OMEGA-3’s are healthier when substituted for SAFA’s the only evidence we have at this time seems to suggest that substituting POLY’s for SAFA’s seems to be good. However, I personally have a belief that it is the benefit of omega-3 that provides that benefit.” However, this is not your argument, but it is what you are saying.
2nd: Lets look at just the study of more than 30,000 persons when they compared the CIRCULATING levels of fatty acids in the blood. They compared the top 1/3 of the selected fat versus the bottom 1/3.
here they are in relative risk.
saturated fatty acids 1.06
this means that those who consumed a large amount of saturated fatty acids versus low had about a 6% increase in the risk of a cardiovascular event. (nope, it did not meet statistical significance, but it is consistent with previous studies and is a clear “signal”)
Omega-3 Highest 1/3 RR of 0.84
this means there was about a 16% reduction in cardiovascular events in the higher consumers. No surprise there
Omega-6 highest 1/3 RR 0.94
this means there was about a 6% reduction in cardiovascular events with the highest consumption of NON omega -3 PUFA’s.
Here was a meta analysis done in 2010 that made a point of including OMEGA-6 and not omega-3 in their analysis to help distinguish between the two (take a look at it, it has plenty of links to other research that demonstrates the same)
If that isn’t good enough take a look at the 7th day adventist study
or perhaps the recent mediterranean diet study
or any epidemiological data.
The evidence is clear that replacing SAFAs with POLYs is a good idea if you want to reduce risk of vascular disease.
“Because I’m evaluating saturated fat, not meat consumption, which is another topic.”
Not in your blog you weren’t you were promoting the benefits of food from that study that didn’t show a relationship to heart disease. The point of commenting on why you seem to have omitted an important finding is that you have a clear biases and aren’t looking for the truth otherwise you would have commented on that point. I suspect it is because meats tend to be where we get much of the saturated fats in our diet and when people think about saturated fats they think of meats. So lets reiterate. The study you seem to think is helpful to your argument states there was a clear harm from increased consumption of meat.
Taken as a whole it is very clear that.
1) replacing polyunsaturated fats for saturated fats can provide risk reduction in heart disease, whether those polyunsaturated fats are omega-3 or n-6’s.
2) calling a study bad because they disprove what you would like to be true doesn’t help anything.
3)In almost all of these studies there is a clear “trend” or signal that increasing saturated fatty acids might be harmful. (for those who don’t know when they perform these meta analysis they really are fraught with errors for many reasons, but in general they have demonstrated that there is no clear increase in risk from saturated fats. However, there is certainly a “trend” or “signal” that there is harm. It is sort of like reading about the odds of getting heads when flipping a certain coin. Lets say every study they do with this “coin” does not show a statistical difference between the number of heads versus the number of tails. However, if you look at the trials themselves you see that in almost every one of the studies heads appears more often than tails (lets say 52% of the time on one study and 51% on another and 54% on another and then 57% on another and 52% of the time on another) even though the research doesn’t demonstrate a statistical significance between head or tails you would be a fool to bet on tails. There is probably something wrong with that coin that trends towards heads.
Again, the best evidence we have is looking at communities or people who seem to live much longer than others. The basics of their lifestyle are….
1) don’t smoke
2) exercise
3) eat primarily plant based diets
4) +/- drink alcohol in moderation
5) don’t eat much meat (see it is relevant, hard to get a lot of saturated fat in your diet if you don’t eat meat)
Casey, it is tedious replying to your lengthy posts, especially when you are citing the same studies I have already commented on in my blog in more detail. Furthermore, I think the reader would see quite clearly who is really interpreting the data in a “skewed manner” after reading your posts and my blog post on the subject.
I initially responded to a specific claim which was misleading, and did not intend to get into an argument with someone who has already made up their mind, and in my view is thoroughly misrepresenting and misinterpreting the evidence.
Given this, I’m quite content to let it rest here and let the undecided readers make up their minds.
I have to call you out on (I am talking to km) “Casey: “You cite O’Sullivan but you don’t seem to mention that he found a SIGNIFICANT increase in the risk of mortality with increased meat consumption”
Because I’m evaluating saturated fat, not meat consumption, which is another topic.
In your blog you write. “O’ Sullivan conducted a cohort analysis on specific food sources of saturated fat, but found no association between butter, milk, cheese, dairy and all-cause/CVD mortality.”
Comments: Saturated fat showed no convincing relations with risk.”
You did intentionally omit the evidence that in O’Sullivans study he found that meat was statistically associated with increased risk of mortality.
On your blog when you use some of O’Sullivans finding to suggest foods high in saturated fats are not bad even though O’Sullivan actually found that increased meat consumption is associated with increased mortality.
You clearly have an agenda.
B.Dobson: “You did intentionally omit the evidence that in O’Sullivans study he found that meat was statistically associated with increased risk of mortality.”
For meat intake, the authors rightly did not attribute the small risk increase to saturated fat. So it in fact further supports my case, and as I said before is a different subject.
I had to navigate to get to the article you were referencing casey. You may want to post the actual link to the paper itself unless you wanted the article you actually linked to to be read. Very informative. 19% reduction in cardiac events with increasing polyunsaturated fats in place of saturated. Cardiac deaths were decreased by 20%. ZM does have a post about this article but there is nothing beneficial or new to be gained by reading his comments. Everything he says and more is addressed in the commentary. Great study! ZMs links were good too but they did seem to support your (casey) argument.
B.Dobson: “19% reduction in cardiac events with increasing polyunsaturated fats in place of saturated. Cardiac deaths were decreased by 20%”
This is Mozaffarian’s analysis. To understand these reviews you have to understand the studies included in it:
Mozaffarian’s review on trials claimed that PUFA in place of SFA reduces CHD events, but once again it included trials with multiple interventions and a trial with serious methodological problems (24, 37, 38, 50), already recognized by others (29, 32). It also left out other trials without justification (49) and did not have info at the time for cause specific endpoints from one other trial (33). Despite all this, there was no reduction in all-cause mortality.
Have you ever read the studies included in Mozaffarian’s review? Furthermore, why cherry-pick one review and ignore all other reviews with a fairer selection of trials?
Thanks for your comment, because it indicates that I should do a review of the individual trials first, then analyze the meta-analyses so the reader would better understand them.
I do not have an agenda, and have never promoted any particular diet.
ZM: Smoke and mirrors. This is a desperate attempt to get people to read your blog. There are methodological flaws in all trials, bar none. You seem to discount the flaws in trial when they approve what you believe, but you embrace the flaws when they are what you do not want to believe.
When Burt recognized that what I said about you omitting meat from O’Sullivan’s research you said it wasn’t relevant to saturated fats even though on your blog you only included the research because it was ONLY relevant to saturated fats. Now you argue that the “author” discounts the effect of meat, guess what? He does the exact same things for the foods you included in your comments. So why would you include foods from the same trial that did not show an increase in mortality from a trial and say the data and research is good but then omit data from the same trial and say it was bad? Yes the author does recognize the flaws in the research (but you didn’t) so you shouldn’t have included any of it if you didn’t want demonstrate bias. You can’t have it both ways. The bottom line is, if you are going to use O’Sullivan’s research you have to take the good with the bad, but don’t take what you want and then dismiss the rest.
Regarding “Mozaffarian’s review” You added nothing to the conversation. Nothing. You have little blurb about something, but they discuss all the flaws paper and explain everything as many authors do. The results stand. It is very clear that replacing saturated fats in a persons diet with polyunsaturated fats regardless of whether they are n-6’s or n-3’s (n-3’s are better) lowers both cardiovascular mortality and morbidity. In your blog you provide a link to a paper from 1965 that has no relevance to this research. Do you even read the things you reference? You argue that they didn’t include reasons for inclusion, but you are wrong yet again. Did you even read the paper? They (just like most) included criterion for inclusion.
Here is the actual link to the research:
http://www.plosmedicine.org/article/info%3Adoi%2F10.1371%2Fjournal.pmed.1000252
If someone wants to know about the research and the limitations go read for yourself. Never rely on a person telling you what the research says. Avoid at all costs a “blog” summarizing and then giving you their own interpretation of the data or you will end up with important data being omitted, unimportant data being highlighted, research being discounted and conclusions derived by personal opinion. The author’s of most research will tell you the flaws and what you can and can’t take from the research. You can also read critical reviews of research online, but you should only do that from a well established source.
As it stands:
Polyunsaturated fats in place of saturated fats will reduce cardiovascular disease.
Low glycemic index (this is according to YOU) in place of saturated fats will likely lower risk.
Most civilizations with longevity tend to have extremely low saturated diets.
Nowhere do you address the fact that persons and cultures that live longest don’t consume much saturated fats.
Casey: “Now you argue that the “author” discounts the effect of meat, guess what?”
Meat intake was associated with a small increased risk, and the authors mentioned various possibilities, but did not actually attribute the increased risk to the saturated fat content. Get it? It’s saturated fat we are talking about here.
Casey: “You argue that they didn’t include reasons for inclusion, but you are wrong yet again. Did you even read the paper? They (just like most) included criterion for inclusion.”
The exclusion of that 1965 trial was unjustified. This was also noticed by Ramsden et al – http://www.ncbi.nlm.nih.gov/pubmed/21118617
Casey: “Low glycemic index (this is according to YOU) in place of saturated fats will likely lower risk.”
Never claimed such a thing.
Casey: “Nowhere do you address the fact that persons and cultures that live longest don’t consume much saturated fats.”
The point of the blog post was to evaluate prospective cohort studies, randomized trials and their reviews or meta-analyses, all at the top of the hierarchy of evidence.
KM: I am not certain you understand the research you are presenting. You provided this information
“There was a study that tried to answer this question which found that carbohydrates with high-GI in place of saturated fat was associated with a statistically significant increased risk of myocardial infarction, but no significant associations for medium-GI or low-GI carbohydrates in place of saturated fat – http://www.ncbi.nlm.nih.gov/pubmed/20375186”
In this study that you provided there was (not statistically significant) decrease in MI when you replace saturated fats with low glycemic foods. Perhaps you don’t understand how to interpret the research but here is the conclusion of the author that you recommend we read.
“This study suggests that replacing SFAs with carbohydrates with low-GI values is associated with a lower risk of MI, whereas replacing SFAs with carbohydrates with high-GI values is associated with a higher risk of MI.”
So I will state again.
“Low glycemic index (this is according to YOU) in place of saturated fats will likely lower risk.”
I am glad that you concede that you have no evidence that cultures and people who seem to live longest lived primarily on plant based diets.
I do find it interesting that you agree that the mediterranean diet (One that is high in PUFA’s and low in SAFA) provides health benefits, but for some reason you don’t believe this has anything to do with the lower SAFA levels in the diet. (The mediterranean diet is essentially replacing both carbohydrates and SAFA with POLY’s and MONO’s, and has clear improvement in cardiovascular outcomes).
In reference to your 1965 study. Now I see what you are doing. You are trying to read what other people are writing and copying them. I suspect you haven’t read much of the research. Take a look at Mozaffarian’s paper. It includes the 1965 in a chart and they give reason for excluding it along with 50 + other papers. Before putting faith in what others write you should understand the information yourself.
Regarding your understanding of the “meat” controversy. I have no problem with saying that there might have been other confounding causes outside of saturated fatty acids that caused the increased risk of mortality (even though this is consistent with other studies) HOWEVER, this is also what the author says about the dairy not just the meat. So if you want to discount the clear HARM this study found with meat you can’t at the same time praise the lack of harm (not benefit) that came with increased dairy, and that is what you did. You wrote and acted as if this study demonstrated that saturated fatty acids weren’t harmful based on this study while ignoring the harm from meat. If you discount the meat in this study you have to discount the dairy.
To summarize:
evidence demonstrates that replacing SAFA’s with PUFA’s will likely be beneficial to your cardiovascular system.
Replacing SAFA with low glycemic index carbs MIGHT be beneficial.
Clear evidence from existing populations with longevity demonstrate high fruit/vege and low meat diets are healthiest.
There is no reply button to your latest post so I’m replying here.
Casey: ” This study suggests that replacing SFAs with carbohydrates with low-GI values is associated with a lower risk of MI, whereas replacing SFAs with carbohydrates with high-GI values is associated with a higher risk of MI.”
I interpreted it correctly because the result for low-GI was not statistically significant. You will notice I said “but no significant associations”. I don’t know if you have read other papers, but other researchers also interpret it the same way I do, because it is not correct to cite non-significant findings to support your case. A consistent standard is imperative to avoid bias, and I cannot come to strong conclusions based on a non-significant result whether it supports my beliefs or not.
Casey: “I am glad that you concede that you have no evidence that cultures and people who seem to live longest lived primarily on plant based diets.”
Hmm…why are you always putting words in my mouth?
Casey: “but for some reason you don’t believe this has anything to do with the lower SAFA levels in the diet.”
because there is no evidence that the benefits are due to safa restriction.
Casey: ” Take a look at Mozaffarian’s paper. It includes the 1965 in a chart and they give reason for excluding it along with 50 + other papers. Before putting faith in what others write you should understand the information yourself.”
It seems that you do not understand Ramsden’s point. By their reason for exclusion (multiple interventions) the STARS study should have also been excluded also for having “multiple interventions”, which can only mean in this context “multiple arms”.
Casey: ” So if you want to discount the clear HARM this study found with meat you can’t at the same time praise the lack of harm (not benefit) that came with increased dairy, and that is what you did.”
It’s not about discounting anything, but rather finding evidence for safa’s so called harm. In this field it is very difficult to discount anything because you can always come up with an ad hoc explanation for why no evidence was found, which is why the burden is on those who want to demonize a specific nutrient. So no evidence was found for for the harmful effects of safa whether it was meat or dairy. Since this is the case, non-causation has to be assumed until proven otherwise, or in statistical terms, no evidence against the null.
“This study suggests that replacing SFAs with carbohydrates with low-GI values is associated with a lower risk of MI, whereas replacing SFAs with carbohydrates with high-GI values is associated with a higher risk of MI.”
KM: I see your limitation in understanding the research. You “cherry pick” as you like to say data. The statement I made stands. This study clearly “suggests” that replacing saturated fats with low GI value carbs is beneficial (and this is consistent with most trials). Nope it did not meet statistical significance (0.05 P) but there was a clear trend (please read my posts). If you walked away from that trial believing only that high GI foods in place of saturated fats is bad then you really need to start over. Again, the author rightfully argues and even states that this study suggests low GI foods in place of saturated fats was beneficial.
I am glad that you concede that you have no evidence that cultures and people who seem to live longest lived primarily on plant based diets
You have no response or even any evidence to support the contrary so your omission in a concession.
Casey: “but for some reason you don’t believe this has anything to do with the lower SAFA levels in the diet.”
because there is no evidence that the benefits are due to safa restriction
Interesting way of looking at a diet that makes a point of avoiding many foods high in saturated fats. Allow me to rephrase. The Mediterranean diet, which is one of the only diets demonstrated to provide primary and secondary cardiovascular benefits which also happens to be a diet high in Polyunsaturated fatty acids and lower in saturated fatty acids and is currently a clear favorite for reducing heart disease and also happens to meet the low glycemic index on carbs but it has nothing to do with saturated fatty acids. I might also point out that this points to the fact that almost any culture we look at that has longevity eats this style of diet. How’s that?
Maybe this. Even though KM hasn’t seen the P value on them yet every diet that seems to decrease cancer and decreased the risk of cardiovascular disease has lower saturated fat and higher polyunsaturated fat as a staple.
with regard to o’sullivan she found “High intakes of meat and processed meat were significantly associated with an increased risk of mortality ” just because there are ways to dance around the fact and say the data could be confounded or whatever you want to say. The fact remains that meat met statistical significance for harm. You even try to state that you can find fault with any study (you can) but you don’t get to praise ones that say what you want and then discount ones that say what you don’t want. I see that you changed your blog on this issue, which was good but your bias is clear. You should have written that she found a statistically significant increase with increased meat consumption but none with dairy.
With regard to
It seems that you do not understand Ramsden’s point. By their reason for exclusion (multiple interventions) the STARS study should have also been excluded also for having “multiple interventions”, which can only mean in this context “multiple arms.
at least now you recognize that they did in fact give reason for exclusion. However, you then try to infer what they meant by multiple interventions. You seem to think it has to do with an “arm” of the trial. Nope. Let me help.
In that trial (which you certainly have to know was not a very good trial. poor compliance. They used a P value less than .1, great googly moogly) when they said multiple interventions they meant that in the placebo arm they essentially did nothing. In the trial arms that both tried to enforce a diet and tried to enforce consumption of different oils (as you know the oil compliance was poor at absolute best and the diet part was not likely very good) so each arm had multiple interventions making it difficult to interpret and apply meaningfully.
Casey: “Nope it did not meet statistical significance (0.05 P) but there was a clear trend ”
If all you have is a “trend” or a “suggestion” then your evidence is extremely weak and insufficient. Citing non-significant results for your case is simply desperate.
Casey: “I am glad that you concede that you have no evidence that cultures and people who seem to live longest lived primarily on plant based diets”
Stop trying to change the subject, and put word in my mouth. I never conceded anything.
Casey: “You should have written that she found a statistically significant increase with increased meat consumption but none with dairy.”
Again, the issue here is saturated fat, not meat per se.
Casey: “at least now you recognize that they did in fact give reason for exclusion.”
I always knew their reason for exclusion. Again, multiple interventions in this context can only mean multiple arms. STARS was a true “multiple intervention” trial in the normal sense since it involved other changes besides fat intake, confounding the comparison. Whereas the Rose trial was much better controlled since the changes were largely confined to dietary fat. Therefore, they could not have excluded Rose because of “multiple interventions” in the usual sense of the word (unless there are crazy), since STARS is the true multiple interventional trial here, which they included.
WOW, you must be new to research, or you are blinded by a love of SAFA. Lets approach 2 of the topics that seem to confuse you.
It is funny that you attempt to suggest that meat and all the other foods included in O’Sullivan’s research is not related to saturated fat. Here is the title of the research.
“Food sources of saturated fat and the association with mortality: a meta-analysis”
She was attempting to study foods high in saturated fats and determine their association with mortality. You seem to want to report the part you like and gloss over the part you don’t.
“Saturated fat showed no convincing relations with risk. Dairy and specific dairy sources are generally not associated with risk, whether high-fat or not (59). O’Sullivan’s analysis is one of the few cohort reviews that reported on total mortality. It is consistent with studies that reported on cardiovascular outcomes as well as total mortality, finding virtually no evidence of mortality increases associated with saturated fat intake”
This is from your blog. You say “saturated fat” in the beginning when referring to the dairy part but deny saturated fats have anything to do with meat. Interesting that YOU want do that when the author did not. I would agree that there are issues outside of saturated fat that could be the culprit, but that would apply to YOUR dairy also. You can’t say her research applies to dairy but not meat. You either take the good and the bad or you don’t take either. You cherry pick. Instead of telling the truth you use the word “virtually”. You seem compelled to point out the evidence that when dairy foods are used as a surrogate for saturated fat it did not demonstrate an increase risk of cardiovascular problems. However, from the same author from the same study you discount the absolute statistical significant finding that meat (when used as a surrogate for saturated fat)
Here is what O’Sullivan says
“High intakes of meat and processed meat were significantly associated with an increased risk of mortality”
The second point I want to comment on is your misunderstanding of research. Multiple interventions does NOT mean multiple arms or even multiple risk. The reason rose’s work (besides the obvious for anyone who has read that research) was not included and the STAR was is because of the multiple interventions. Apparently this needs some explaining.
In Rose’s work there was a placebo arm (simple enough) the other two arms had multiple interventions BOTH DIET AND OIL intake. There was NO single intervention. This trial would have been included if they had included a 3rd arm that only had diet included. (see, one intervention) Rose had more than 1 intervention in each arm. This has nothing to do with the number of arms. In the STAR trial there were more than one are BUT THEY HAD ONE WITH A SINGLE DIETARY INTERVENTION Hopefully now you go back and understand the research a little better.
Casey: “In Rose’s work there was a placebo arm (simple enough) the other two arms had multiple interventions BOTH DIET AND OIL intake. There was NO single intervention.”
First of all, to correct your last post, they used an alpha level of 0.05, which is standard. The increased risk approached the significance level, but was not statistically significant (i.e. p > 0.05). Also, enforcing a diet by replacing animal fats with unsaturated oils is not a multiple intervention in the normal sense, and this is exactly what you want to evaluate (as Rose did). If this is a multiple intervention then all trials should be excluded from the analysis, because all of the other trials also replaced animal/saturated fat with unsaturated oils. Including changes in animal fat AND other factors (e.g. increases in fruit and vegetable intake, reductions in trans fat and processed foods etc.) is employing multiple interventions. Oslo and STARS employed multiple interventions, Rose did not. Get it?
Casey: “You say “saturated fat” in the beginning when referring to the dairy part but deny saturated fats have anything to do with meat.”
NO. Dairy is not associated with risk so there is no evidence that saturated fat is harmful there. Meat increased risk but there was insufficient evidence to conclude that this was due to saturated fat.
I’m looking for evidence that saturated fat increases risk, not meat. After all, I hope you realize that the claim is specific, that saturated fat increases risk.
wrong again. Lets try again.
In rose’s trial the asked one arm to do nothing (no intervention). Each of the other two arms were asked to do two things 1st they were asked to avoid fried foods, and asked to restrict milk eggs among other things. If they left it at that it would be a “Single intervention trial” the intervention being the diet, the one intervention not included in the placebo arm. Here is where you get confused. They then asked each of the intervention arms to consume a certain amount of oil each, the oil as a substitution is NOT considered a food replacement it is a separate intervention. It is not the same as saying eat more fish and less red meat. It is like saying eat less meat (one intervention) and take this horrible tasting pill. There were 2 interventions neither of which were followed very vigilantly .
The results become clouded because it isn’t clear whether the harm or benefit came from addition of oil or restriction of certain foods. Do you see the two interventions? In star they did not do this. There was ONE intervention. If that doesn’t help you might need a class on research. The author of the research and everyone else seems to understand why a paper from 1965 that had multiple interventions, poor compliance and did not reach statistical significance was not included, but KM does not.
Wow, you doubled down on wrong.
“NO. Dairy is not associated with risk so there is no evidence that saturated fat is harmful there. Meat increased risk but there was insufficient evidence to conclude that this was due to saturated fat.
I’m looking for evidence that saturated fat increases risk, not meat. After all, I hope you realize that the claim is specific, that saturated fat increases risk.” Let me rephrase your own sentence, you will see the folly.
“I am looking for evidence that saturated fat increases risk, not dairy. After all I hope you realize that claim is specific, that saturated fat increases risk” Pretty silly isn’t it?
Lets try again. The researcher who performed the research thought there might be something besides saturated fat when they came to the conclusion regarding both dairy and meat. You don’t get to pick one (maybe in your mind you do, but I will once again quote the author when it comes to meat and will even add the dairy part. You turn milk cheese yougur and butter into “saturated fat” and then pretend like “meat and processed meat” has nothing to do with saturated fat. I don’t care whether you call it saturated fat or not, but you can’t have it both ways.
“Pooled relative risk estimates demonstrated that high intakes of milk, cheese, yogurt, and butter were not associated with a significantly increased risk of mortality compared with low intakes. High intakes of meat and processed meat were significantly associated with an increased risk of mortality”
Casey: ” Each of the other two arms were asked to do two things 1st they were asked to avoid fried foods, and asked to restrict milk eggs among other things. If they left it at that it would be a “Single intervention trial” the intervention being the diet, the one intervention not included in the placebo arm. ”
If they left it at that it would be called a “dietary fat reduction trial”, not a “fat modification trial”, and if not compensated by other foods weight loss would occur and confound the comparison.
Rose is a “fat modification trial”, comparing animal/saturated fats to unsaturated fats (in this case from vegetable oils). The diets of both groups were similar with the exception of the fat sources. Given this, Rose’s trial had the ability to evaluate polyunsaturated fat vs. animal/saturated fats. STARS was unable to do this because of multiple confounders (numerous differences other than dietary fat between the groups). Calling oil a “substitution” or “food replacement” is irrelevant, since the point of the trial was to compare different sources of fats, which it did.
By your definition every trial (not involving weight loss) should be excluded because every trial involves replacing something with something else to maintain energy balance. So every trial would involve at least two interventions.
Casey: “The author of the research and everyone else seems to understand why a paper from 1965 that had multiple interventions”
Actually, I’m pretty sure that you are the only one who thinks that Rose employed “multiple interventions”.
Casey: ” The researcher who performed the research thought there might be something besides saturated fat when they came to the conclusion regarding both dairy and meat.”
You continue to miss the point. No evidence was found that saturated fat was harmful, regardless of food source. Period, end of story. I am not the one turning milk, cheese etc. into saturated fat. People who demonize saturated fat point to these foods and specifically blame saturated fat. These are the people that turn whole foods into specific nutrients, not me. I am simply refuting these claims, showing that demonizing a specific nutrient is misguided. Don’t you understand that? If you claim that saturated fat is harmful then you have to show specifically that saturated fat is harmful.
Also, I don’t know why you are referring to me as “KM”.
You are hopeless, going back to your clear lack of knowledge about trials and about these trials specifically.
Lets go back to square one. The 1965 trial was excluded by the author because of “multiple interventions’ so your statement
“Actually, I’m pretty sure that you are the only one who thinks that Rose employed “multiple interventions”
demonstrates just how little you know about both works.
You also comment
“If they left it at that it would be called a “dietary fat reduction trial”, not a “fat modification trial”, and if not compensated by other foods weight loss would occur and confound the comparison.”
wrong, please reread the information already provided by me. Also, since it is clear you did not read Rose’s work. IT WAS COMPLICATED BY WEIGHT LOSS. They dismissed it as unexplained. Wow, what great research. There are many reason to exclude this multiple interventions trial. So it seems that you are the only one NOT calling it a multiple interventions trial (i suspect you have never actually performed research).
You state that the research demonstrated that saturated fat did no harm and used dairy as an example you can’t do that unless you also claim that saturated fat from animals does harm. You can say, dairy and such don’t seem to increase mortality but meat does. It is disingenuous to say “dairy doesn’t seem to cause harm so saturated fats don’t cause harm but meat causes harm but I don’t agree that it is from saturated fat so i wont include it”. (to use your terms) if you want to show saturated fat is not harmful then you have to show that it is the saturated fat that is not harmful not the dairy. It works both ways. I don’t care whether you say the paper didn’t show anything or that it showed that the specifics on saturated fats might matter (subtype) but you can’t say that since dairy didn’t show harm then we know saturated fats are harmless but because meat showed harm it is meaningless. Did you read the paper? Did you read what the author wrote. So once again you stand alone in your interpretation of the information. Your “agenda” as dobson says is clear.
sorry bout the KM thing
PP Link. I have a hard time swallowing information from a cite so angled toward a plant based diet. with that said. I do believe a plant based diet is better and after reading the statistics presented by your site and some of the things said by casey and zm I think I need to read more before I comment as an informed consumer. So far I am with ya! Go plants, but can I still have bacon?
Burt: The information provided can be difficult to understand. When reading research papers you have to understand a lot about research itself. There are flaws in all research and some research can’t be applied to the general population. For instance early in this blog Chris provide as evidence to support the Paleo style diet a trial that had about 10 people in it and he reports it as fact. That is laughable. A trial that small might, at best, be a “hypothesis generating” trial. Meaning if it finds something it might design a larger trial to determine if it applies to the general population or even a specific population.
In general the RCT’s are considered the best. Meta analysis are always fraught with errors, but still provide information. Sometimes we have to rely on observational, cohort or other studies to give us information because the data can’t be obtained any other way. For instance, with tobacco smoke we don’t have any real RCT’s to prove that they cause cancer (some would argue that without a good RCT you can’t prove anything, like they do with saturated fats) but it is unethical to make a group of people smoke, which is what you would have to do with a RCT so one will never get done.
I would like to say that we are lucky to have such a large group of people who are trying to follow the different diets in our current time. They will provide data and evidence to either support their usefulness or they will fall away and disappear like so many do.
In our hospital we recently had a young woman come into the hospital following the Paleo diet who was in ketosis (no big deal) because she wanted to be in ketosis. She was in great health and had been for many years because she exercises often (runs and lifts weights). However, she ended up in our hospital because after a workout her abdomen was hurting her urine was darkening and her abdomen began to swell. When I saw her she was being treated for Rhabdomyolysis (for those who don’t know this is when the muscle starts to break down and can damage the kidneys. It happens in severe accidents, because of medications, and because of exercise (these are the main ones, there are other causes). However, dehydration lowers the threshold for getting rhabdo and this person because of her diet and exercise was in chronic mild dehydration. The most likely cause (the straw that broke the camels back so to speak) of her rhabdomyolysis was the diet. She had not changed her exercise routine in years, but her diet was new (a couple of months). She didn’t believe that diet played a role went home and drank more water daily and was back in the hospital within a month with rhabdomyolysis. I would like to add that rhabdomyolysis has a high mortality rate.
Thanks casey and zm, I read your blog regarding the mente study. I think you (ZM) are trying to do what you think is right, but I came away from reading that trial to understand that the mediterranean style diet is one thing they found to meet the strong criteria. This is the style of diet that casey seems to recommend. It is low in saturated fats and high in vegetables/fruits. I do think the two of you are closer in agreement than it would seem on here. I don’t know how you find the time to understand this material in such detail, it seems confusing and it seems like the jury is still out on most things. The one thing that seems to ring true is that the mediterranean diet is healthy.
B.Dobson, you’re right that the mediterranean style diet has shown benefits, but I think this has nothing to do with low saturated fat and don’t see them as incompatible.
If you can stick to a mediterranean style diet then by all means go for it. Here is an article you may be interested in – http://journals.cambridge.org/action/displayAbstract?fromPage=online&aid=8453189&fileId=S136898001100259X
So much writing since my last visit. I have finished the blue zone and read more of the research the two of you (casey/zm) are quibbling about.
Casey I see your point on the research either showing the trend toward harm with saturated fatty acids or benefit from substituting. I also see that ZM does appear to look the other way when data isn’t giving him what he wants. I still think he has an agenda.
ZM: it looks like there is something wrong with most of the research when it comes to diet. Almost impossible to do a good randomized controlled trial. You seem to accept some but reject others when they all seem to have fault. If you think the only thing wrong with that meta analysis is one paper they included and one they didn’t I would say that you are sweating the small stuff. I don’t know if casey is being clear or not, but when I looked at the 2 studies I understood what he and the authors meant by interventions.
You two should stop with the small stuff.
I believe based on what I have read is that the studies are incomplete. I also believe that the proof is in the pudding and everything I have read says cancer and heart disease are lowest in those who follow plant based diets. The rest is scientific conjecture based on impossible to design trials.
Bingo! You have hit the nail on the head. In general we in the world of science (some, not all) want “Evidence based” therapies. Before we believe something you need a “randomized controlled trial” (the gold standard) of trials to see if something works or not. With a medicine it is fairly easy. You blind everybody. Divide the groups into two similar groups in most respects then you give one the intervention and one a placebo or other current therapy. The outcomes provide reasonable evidence. HOWEVER>>>>>>>>
When it comes to diet it becomes almost impossible to design a randomized control trial that is meaningful because it is difficult to “blind” (can you make half of people to eat tofu chicken and make them think it is actually chicken?) it is also difficult to get compliance. The reporting system isn’t the best (a questionnaire, and sometimes only one a year. What did you eat the day before to reflect what you have eaten all year). Then the problem is so many foods have such a diverse composition. “I ate two eggs yesterday” we can get a general idea of the protein, but what if i fried it or what if it is organic. What about yogurt, the variations in carbs, sugar, protein and fat are tremendous. So the questionnaires while most have past validity tests reliability and generalizability to exact macro and micro content are poor. Just as KM points out (but then manipulates the findings) just because meat demonstrates risk does it mean it is from saturated fat? Just because dairy didn’t show risk how do we know the nutrient value of dairy masked the harmful effect of saturated fats. We don’t.
When it comes to diet a good RCT if practically impossible except in terms of generalizability. When it comes to specific diets the data is a little different, the 7th day adventists follow a pretty strict diet, mediterranean (crete) style is fairly generalizable. This is why epidemiological data is so important when it comes to diet because when you perform a RCT there are always questions and doubt even more so with diet (as you see from the quibbling). To complicate matters saturated fats are mostly myristic, lauric, palmitic, and stearic acids and most foods with saturated fats have a mixture of them. These fats have a mix of how they are digested whether through the portal vein or through incorporation into chylomicrons. And each has a very specific effect on inflammatory mediators, we know this from specific isolation of each fat and infusion studies. The problem is that we lose information as we move away from the microscope because most foods have a mix of them. There are some exceptions, like coconut oil has a lot of lauric acid where most saturated fatty acid foods don’t. So you see the confusion when we try to generalize the effect of saturated fatty acids. While RCTs are still helpful they can be confusing to interpret,which is why in any discussion on an RCT the authors mention trends and don’t just focus on the P value, which is important but isn’t an “end all be all” sort of thing. Let me explain that a little.
We normally use a P value of 0.05, it can be any value (P of 0.1 can be used or 0.01, in the selection we call this an “alpha”, no need to worry about the difference) basically this means there is a 5% chance that the study is wrong. So even if you get a p value of 0.05 there is still a 1 in 20 chance that the study is wrong. What if you did a study and your p value was 0.051. Some would draw a line in the sand and say you didn’t prove anything others would say that it didn’t meet the 0.05 level but come on with a p of 0.051 there is a 5.1% chance that the study is wrong versus a 5% is it really that big a difference. It can be, but because diet trials are so difficult to actually randomize the “trend” really matters. Your “proof in the pudding” is exactly right. What we know with certainty is …..
1) longevity seems to follow persons/people who have diets high in plants low in meats (as the “blue zones” points out)
2) evidence that replacing saturated fatty acids with PUFA’s is pretty good (especially if they are omega-3’s)
3) don’t smoke (you won’t find an RCT on smoking but that doesn’t mean cigarettes are good)
4) Diets high in meats tend to have lower life expectancy. Even ZM seems to agree with this (he doesn’t agree that it is from SAFA).
5) We will never have a well designed RCT that puts the nail in the coffin for a macro dietary nutrient like “saturated fatty acids”, it would just be too difficult to perform.
B.Dobson: “Almost impossible to do a good randomized controlled trial. You seem to accept some but reject others when they all seem to have fault.”
No one ever said that there are perfect trials. However, there are degrees of quality based on objective principles. Different trials differ in what they can claim based on quality and confounders.
The trials finding benefits were rife with confounders to anyone who has read them, since there was no attempt to isolate the pufa safa change. Most of the older trials attempted to isolate this change and so were far better (not perfect) controlled. If you are familiar with the concept of a scientific experiment you would know what I’m talking about.
Remember that the claim is that replacing safa with pufa is beneficial, and if a trial replaces safa with pufa AND decreases trans fat, decreases processed foods, increases fruits and vegetables etc., then the ability of the trial to evaluate the pufa safa change is considerably diminished. This is basic science here, not an agenda.
B.Dobson: “You two should stop with the small stuff.”
I tried to end the discussion earlier, but hopefully it can truly end now because there is no resolution in sight!
Casey: “You are hopeless, going back to your clear lack of knowledge about trials and about these trials specifically.”
OK Casey, enough. I’m not going to reply anymore, because you would keep carrying on and on to no end. I’ve also endured your snide remarks that you have inserted throughout your replies. If you think your arguments are so strong then let it rest, please.
Because of the confusion on this matter I spoke with a colleague of a friend of mine. He is an epidemiologist who worked with bill harris (of omega-3 fame). He put things in context for me and helped me understand why so many studies were done and why the debate still rages. He gave me useful insight into the battle, the war and the evidence. I will try to write what he told me.
* your body can make all the saturated fat it needs and your body would be ok if it never got saturated fats
*your body can not make all the PUFA’s it needs and therefore they are “essential” in the diet
*with trials most people want an obvious winner and loser so a generic threshold for winning is determined. That is the P value. Lower than the P value and you have a winner, higher and you don’t
*current science has generally accepted a P value of 0.05 in almost all cases, but they shouldn’t. The smoking analogy makes sense now. We know from so many sources that smoking is bad (BUT NOT ALWAYS!) that setting a p value of 0.05 isn’t necessary. You could design a trial with a p value of 0.08 or even 0.1 if you wanted. The p value is arbitrary
*the meta analysis and trials in general have people who “poke holes” in the trial to discredit the trial or just demonstrate the weaknesses in the trial.
* with saturated fatty acids many trials never reach the p value that is set for the trial so some say that saturated fats aren’t harmful and they are sort of right, but because even though it tends to not reach the p value saturated fats almost always lose the race.
*one of the problems with being able to reach that p value in the trials as Z.M. points out is there are so many confounders in the trials that it makes them impossible to perform and any trial performed will have faults. The things that complicate the trials can be called noise.
*another problem with saturated fats as Casey points out is not all saturated fats are equal some are likely bad and that is why saturated fats never win the race.
*because of the inability, except in a lab, of isolating the effects of each type of saturated fat we won’t have good RCT’s on the matter.
*duration of the trials matter. Will 2 years of eating a low saturated fat diet make up for 50 years of not eating a low saturated fat diet?
*calories probably matter more. If you are eating too much that is the first problem.
*do you already have cardiovascular disease
*what about genetic issues? he mentioned that some persons have genetic disorders in processing of saturated fats making them more at risk for developing disease with saturated fats.
*he also said that data from societies, cultures and people strongly suggest that consumption of fish is likely good, lower consumption of meat is likely good, higher consumption of fruits and vegetables is good.
*despite the noise from many of the trials it still seems prudent to limit foods high in saturated fat since they lose most of the races and diets high in them have higher rates of cancer and cardiovascular disease, eat more fruits and vegetables, eat more fish.
*There was something interesting he said about the questionnaires that are filled out by people in many of these trials. He said they were tested to see if they accurately assess what people ate over time and he said they were pretty good but had faults. One of the faults is that people tend to fill them out and underestimate how much saturated fat loaded foods they eat and overestimate how much veges and fruit they eat. He said that the trials that use them start out with a certain amount of “noise” before the trial even gets started. He said this was one reason they set some of the trials up to measure quartiles or Tertiles because they want to assume that everyone will exaggerate in the same way, which is also not true.
*I am still confused, but I will aim for a mediterranean style diet with bacon on sunday!
B.Dobson: well said.
It took me awhile to understand lipids but I persisted…
1. Threw out all cheap oils in plastic containers (canola, safflower etc) and replaced with non fake coldpressed virgin coconut oil and olive oil (for hot/cold foods).
2. I eat raw almonds although I do eat limited amounts of other raw or roasted non salt nuts and seeds.
3. I eat grass fed beef once a week, fish 2-3 times a week.
4. I love avocados and was fearful to include in my diet, but I did.
These changes led me to question ALL processed foods which I have limited over 4 years.
I lost weight and for the first time in my life I feel well fed.
My fears are that food manufacturers will make a sham of what is thought to be healthy in present day sense and that could mean it becomes a mess to sort out for most of us.
The right fats, whole foods veggies, fruits, nuts, seeds and legumes I what I eat without calorie counting.
Ohh, othe health benefits: AiC is normal, cholesterol numbers are all in range except slight elevation of LDL. The main thing is I feel better.
It takes a leap of faith to reject what doctors tell you…mine went in a mini doc rage when she asked what fats I eat and I said coconut oil….. but she is old school which did nothing for me nutritionally.
One more thing…. I do not eat bt corn, bt soy etc. I really believe we are totally duped by chemical companies like Monsanto. This was a true eye opener.
How has the American public been duped Chris? The Standard American Diet (SAD) includes so much pizza that cheese consumption has doubled from 1970 to 2012.
http://www.ers.usda.gov/amber-waves/2014-june/trends-in-us-per-capita-consumption-of-dairy-products,-1970-2012.aspx
From 1970 to 2000 per capita meat & poultry consumption is up 19 pounds a year or about 60%.
http://books.google.com.ph/books?id=wvZIXf2azKkC&pg=PA42&lpg=PA42&dq=US+beef+consumption+per+capita+since+1970&source=bl&ots=8Tp2XunTrZ&sig=QfBZkmgrTM81_Dde29L8TAXNkpY&hl=en&sa=X&ei=lVypU8S5O8W_kQXXq4DgAg&ved=0CFAQ6AEwCg#v=onepage&q=US%20beef%20consumption%20per%20capita%20since%201970&f=false
I am not saying it is only or even primarily the saturated fat but since Ornish and Esselystyn were able to reverse heart disease with a particular diet that limited these items please do not be so quick to discard the thought that maybe limiting meat and cheese consumption is a good idea.
We need much more study about saturated fats and the contents of meat and cheese.
Hi everyone, Im 23. As someone here whos encountered years of horrific dietary SAD in my early years and a past sufferer if manic depresssion\PTSD anxiety as well, I can easily say symptoms of ill health, both mentally and physically have reversed in a 7 month time span due to a High- Carb, Low fat Vegan diet. Symptoms improved in 2 weeks into 80\10\10, and has incredibly changed my life for the bettrr. I no longer in need of Anti depressants and Anti-pyschotics or any form if extra stimulants as i eat this way, primarily Fruits and Vegetables. Its liberating. I come from a meat dominated and processed food background which I can personally say had an impact on my health. Most people, from my perspective and personal exoerience are missing out on the emotional tole fat plays our health, abd well being and i realized it first hand when went High carb lowfat vegan. I even can go a full week without an over-fat and still cycle for 2-3 hours with no fatigue daily. Even eating this way, created so much distance from eating fats at all. I find the taste of any fat more and more repulsive as i eat more carbs. All i know is my body and brain thrive on carbs and its absolutely fixed ny emotional\mental health. Just a thought
Brendan,
Your symptoms sound more like leaky gut. Do you have any tummy issues like constipation, diahrea, IBS, etc? Eliminating processed food always result in improved digestive issues. Even processed meats. Fat really does not take it’s toll on the brain unless it is rancid or transfat, afterall, the brain is made of fat. SO fats from fresh grassfed meats, coconut oil, butter etc is healthy for the brain. What causes depression when leaky gut is involved is tiny amounts of the intestine leak into the bloodstream and make its way to the brain, causing mental issues.
Forget this sensationalist hack. He’s ignoring two studies which disprove his claim:
http://www.plosmedicine.org/article/info%3Adoi%2F10.1371%2Fjournal.pmed.1000252#pmed-1000252-g003
http://ajcn.nutrition.org/content/93/4/684.short
When will we put the nail in the coffin of shoddy journalism?
Lewis, I think things get complicated for the average person. Many of the recent meta analysis suggest that diets higher in saturated fats may not be statistically worse than other fats. However, as is true with most meta analysis, there are people who question the methods and so on. If you read my comments above you will see that even if we let the data they present as is…..even that data “suggests” (even though not statistically significant) that saturated fats are worse than any other fats but trans fats.
You are also very correct with the replacement story. Almost all data to date suggest if you replace saturated fats with mono’s or Poly’s you have much better cardiovascular outcomes.
Perhaps you have read the latest on the inuits? They have always been touted for having a diet high in fats (they do, but not saturated fats so much) and low in heart disease. We have known for a while their risk of strokes was higher than average. New evidence suggests they also have rates of heart disease as high as ours (we already knew they died about 13 years before we do).
http://www.sciencedirect.com/science/article/pii/S0828282X14002372
Lewis, who are you talking about. Please clarify.
really?
This young man has speaking engagements throughout the country and has the blog that I mentioned earlier. Here are some comments from his saturated fat blog
http://www.cavinbounce.com/semis-and-road-closures/
It’s the animal protein that contributes to coronary heart disease and not fat. Check the NIH research done by Dr Colin Campbell .
It is sugar/carbs…the mother of all inflammation.
I would rather eat high fat than high carb. I agree with the other commenters who say that they feel negative effects when eating sugar and other high-glycemic carbs.
I also cook with coconut oil every day which is very beneficial to our health and is yet a saturated fat.
Melody, I agree. A diet high in saturated fats…and other fats, are healthy and they rev up the metabolism.
Lights camera Action. want to introduce you to Cavin “Bounce” Ballaster:
http://www.cavinbounce.com/fat-chance-part-four/
Here is another observational link:
http://en.wikipedia.org/wiki/Inuit_diet#Perceived_benefits_of_the_diet
ANother cancer study
http://www.donnieyance.com/can-a-ketogenic-diet-cure-cancer/
Importance of Cholesterol:
http://articles.mercola.com/sites/articles/archive/2014/05/11/cholesterol-trans-fats.aspx?e_cid=20140511Z1_SNL_Art_1&utm_source=snl&utm_medium=email&utm_content=art1&utm_campaign=20140511Z1&et_cid=DM45851&et_rid=517925072
LCHF could replace dialysis…really?
http://www.nbcnews.com/id/42689095/ns/health/t/low-carb-high-fat-diet-could-replace-dialysis/#.U3HWiMIo_4g
http://www.mountsinai.org/about-us/newsroom/press-releases/diabetic-kidney-failure-may-be-reversed-with-low-carbohydrate-diet
Here is another link describing the diet and health of Aborigines.
http://www.livestrong.com/article/291690-the-aboriginal-diet/
Brain injuries, neurological disorders such as epilepsy, parkinsons and dystonia have been successfully treated with a diet that is rooted in saturated and other fats. Better known as the ketogenic diet, there is an organization , “The Charlie Foundation” which was established on the importance of the ketogenic diet in treating epilepsy. Here is a link discussing “Your Brain on Ketones”.
http://www.psychologytoday.com/blog/evolutionary-psychiatry/201104/your-brain-ketones
I really am glad that found you on here and respect the fact that you post criticism to your blog because I know you could easily just delete it or not allow it. I want reference studies that say that some saturated fats have negative effects on brain function but at the same time all of these studies could be total hogwash, but I don’t know..
The problem is that we label saturated fat in the diet as controversial. Saturated fat has and will always be a healthy staple in our diets rather if we admit to it or not. Controversy lies in the fats that were made in factories and labled as healthy. Then people were more than happy to consume canola oil and margarine. ALso Asians have a very high incidence of insulin resistance so poor lipid balance must travel side by side with that. I’ve doubled if not tripled my saturated fat consumption over that past 14 years. My kidneys did not explode, my hair didn’t fall out and my cholesterol is exemplary. Back to the Asian culture with high incidence of insulin resistance/ diabetes, I would like to share this link.
http://aadi.joslin.org/content/asian/why-are-asians-higher-risk-diabetes