Updated October 2019
Apparently Peter and his buddy Gary Taubes just love themselves some crappy pseudoscience. Both of them are funded by the Arnold Foundation, with substantial ties to animal agriculture industry lobbying. Marion Nestle notes the Arnold’s working relationship with a National Restaurant Association and the National Cattlemen’s Beef Association consultant.
So let’s go over the bogus arguments Attia makes in this hour long lecture.
He starts off by citing the Siri-Tarino et al. (2010) study, saying no significant evidence could be found to show dietary saturated fat intake is associated with an increased risk of coronary heart disease (CHD) or cardiovascular disease (CVD).
The Siri-Tarino meta-analysis only looked at prospective cross-sectional epidemiological studies. We’ve known since the 1970s that cross-sectional epidemiological studies don’t have the statistical power to show an association between saturated fat intake and heart disease. The only studies that have the power to show the relationship come from dietary change experiments, comparative studies or studies where they control for baseline cholesterol values.
Cross-sectional epidemiological studies are expected to show a zero-correlation due to the wide variability of baseline cholesterol levels in a given population. This means they do not disprove the cause-and-effect relationship between saturated fat/cholesterol intake and heart disease risk, even if no relationship shows up in their data.
He goes on to cite Chowdhury et al. This study looked at a mix of observational and randomized control trials. Same issues again with the observational data, but the RCTs in this case only looked at supplementation of omega 6 and 3 polyunsaturated fats. So this tells us absolutely nothing about the role saturated fats play in CVD, it only tells us that supplementing more polyunsaturated fats isn’t going to protect us.
Dr. Walter Willett, Chairman of the Department of Nutrition at Harvard School of Public Health, discussed the Chowdhury et al. study here, saying, “this meta-analysis contains multiple serious errors and omissions, the study conclusions are misleading and should be disregarded.” Dr. Michael Greger discusses both the Siri-Tarino and Chowdhury studies and their flaws here. (I highly recommend taking the time to watch that short video by Dr. Greger as it pretty much destroys all of Attia’s arguments from here on out.)
Then Attia goes on a long rambling history lesson about Ancel Keys that makes no real point at all, other than to bad mouth Keys’ ground breaking epidemiological work. Keys’ work at the time was the best science that could be done on the subject. Keys’ findings are what led to the hundreds of dietary change experiments being done, which showed us conclusively that dietary saturated fat and cholesterol are the main culprits behind cardiovascular disease.
Then Attia gives a lecture on the subject of “correlation does not equal causation”, while bad mouthing epidemiological data. I find this hilarious, given that the only studies he’s cited so far in favor of his position are epidemiological studies. 30 minutes in, he’s still rambling about Ancel Keys.
Then he starts attacking the Framingham Heart Study. He notes that the FHS found no association between men with a total cholesterol over 300 and those with a total cholesterol under 170 with the amount or type of fat consumed. Again, this study lacks the statistical power to making a significant finding. This is probably why the data he cites remained unpublished.
When we look at metabolic ward studies, where they lock people in a lab and completely control their diet, we can see that the addition of dietary saturated fat raises serum cholesterol so consistently that you can actually create a mathematical formula that will predict exactly how much saturated fat is required to raise your cholesterol by a given number of points.
There is NO DEBATE among medical professionals that the higher your serum cholesterol values, the greater your risk of heart disease is. We know with total certainty that ingesting saturated fat and cholesterol will raise your serum cholesterol values, as was proven in those 395 independent medical ward dietary trials. No matter how many epidemiological studies Attia may cite, this basic fact of science remains the same.
Then he starts going on about more epidemiological data that counters what Keys found. Again, arguing about epidemiological data as it pertains to CVD and CHD at this point is nothing more than a red herring argument. It’s pointless to argue about since we have randomized control trial dietary change experiments that show conclusively that dietary saturated fat and cholesterol are the primary risk factors for disease.
Then he cites the 1981 paper by Shekelle, saying that even though the paper concluded that eating a high fat diet was a contributing factor for heart disease, we should ignore the paper because they found saturated fat in the diet wasn’t significantly correlated to CHD. Again, another epidemiological study that suffers from statistical power problems, but the authors were smart enough to understand this, which is why they came to a conclusion that was opposite to what their own data suggested.
Now he’s back to talking about 1970s and 60s data from unreferenced sources, claiming that in 1963 Hungarian researchers found a benefit for eating a maximum of 1.5 oz of fat per day, while in 1965, British researchers found no benefit to eating 1.5 oz max per day. It’s impossible to refute since the source isn’t cited. I suspect something funny was going on with the diet in the British study.
Then he cites the 1973 Minnesota Coronary Study. Here’s what that study actually looked at:
The Minnesota Coronary Survey was a 4.5-year, open enrollment, single end-time double-blind, randomized clinical trial that was conducted in six Minnesota state mental hospitals and one nursing home. It involved 4393 institutionalized men and 4664 institutionalized women. The trial compared the effects of a 39% fat control diet (18% saturated fat, 5% polyunsaturated fat, 16% monounsaturated fat, 446 mg dietary cholesterol per day) with a 38% fat treatment diet (9% saturated fat, 15% polyunsaturated fat, 14% monounsaturated fat, 166 mg dietary cholesterol per day) on serum cholesterol levels and the incidence of myocardial infarctions, sudden deaths, and all-cause mortality.
And we’re supposed to be surprised that the treatment group had more deaths? That’s the best he’s got to support his nonsense? The treatment group was still eating a 38% fat diet and 166 mg a day of cholesterol! We know from studies done by Ornish that when total fat is lowered to 10% and saturated fat and cholesterol are lowered to zero, cardiovascular disease is reversed. And we also know that when carbs are consumed along side animal protein the insulinogenic effects of the meal are nearly doubled. Here’s a short video that provides more information on the insulinogenic properties of meat when paired with carbs. Given this information, it makes sense that the treatment group ended up with more incidents.
In the AHS-2 study, composed of 96,000 people followed for 6 years, 8% of which were vegan, and 28% of which were vegetarian, they compared the vegans to the meat eaters and found vegans/vegetarians had far lower incidences of cancer, heart disease, diabetes… basically every bad thing you can think of was lower in the vegan/vegetarian groups compared to the meat eaters. Further, the meat eaters in this study ate much smaller portions of meat and were in better health compared to the general population.
Why is he still rambling on about studies done in the 1970s? Why is he ignoring the hundreds of other dietary change experiments that have been done in the meantime?
Now he’s back to the Framingham study. Still rambling on about the history of cholesterol computation. He’s claiming that based on the study’s findings, LDL is a marginal predictor of heart disease, so we should ignore it and instead focus on HDL, since the study concluded that raising HDL lowered disease risk. So by his logic (not the paper’s), since eating saturated fat raises HDL, we should eat saturated fat. I mean wow. Way to twist the findings.
It’s worth noting that subsequent studies have found that saturated fat raises LDL at a rate that is higher than it raises HDL when consumed with dietary cholesterol. To quote one study, “The ability of saturated fats to raise LDL cholesterol is enhanced by increased intake of dietary cholesterol as well as baseline LDL cholesterol concentrations.” (and that study was written by Ron Krauss, who is funded by the National Dairy Council, National Cattleman’s Beef Association, and the Robert C. and Veronica Atkins Foundation!)
This dietary change study found that adding cholesterol to a diet with saturated fat increased LDL cholesterol levels about 5 times more than just the saturated fat alone. So coconut and palm oils are bad, but it’s those hamburgers and eggs that are really going to kill you. Here’s a short video that covers this subject in much greater detail.
He goes on and on about HDL and triglycerides. It’s worth noting that Ornish found that when he put people on a whole food plant based diet, their HDL dropped and their triglycerides increased, yet they still showed a reversal of their heart disease through angiography. So that basically blows his whole theory about HDL and triglycerides out of the water. Current research suggests that low HDL plays no role in the cause of disease.
Then he covers the MRFIT and LRC trails, which were drug control trials. I don’t find it surprising that there were virtually no differences between them. Again, he ignores conclusions of the studies authors claiming that the data doesn’t support their assertions, while ignoring the reasoning behind the authors claims.
Bad mouths the NIH consensus conference. No evidence presented.
Then he talks about the Cochrane Collaboration’s 2001 study, claiming that they found modified fat intake had no significant effect on longevity or cardiovascular events out of 27 RCTs. Here’s what the study actually found:
Twenty seven studies were included (40 intervention arms, 30,901 person-years). There was no significant effect on total mortality (rate ratio 0.98, 95% CI 0.86 to 1.12), a trend towards protection form cardiovascular mortality (rate ratio 0.91, 95% CI 0.77 to 1.07), and significant protection from cardiovascular events (rate ratio 0.84, 95% CI 0.72 to 0.99). The latter became non-significant on sensitivity analysis. Trials where participants were involved for more than 2 years showed significant reductions in the rate of cardiovascular events and a suggestion of protection from total mortality. The degree of protection from cardiovascular events appeared similar in high and low risk groups, but was statistically significant only in the former.
Of course, he never mentions the bolded part of the findings. Further, who knows what constituted a “low fat” diet. Most trials consider 30% to be a low fat diet, which is still a ridiculously high amount of fat compared to what a plant based diet would provide. And of course, cholesterol intake was not controlled for either.
Then the 2006-11 study that found that risk factor intervention had no effect on mortality. Well this study only looked at using counseling and education aimed at behavior change. Is anyone surprised that Billy Bob didn’t quit smoking and stop eating his ribs when the doctor told him to? This study doesn’t tell us anything about risk factors, it only tells us that counselling has a limited ability to change people’s habits. This isn’t some shocking revelation.
Then he talks about the Women’s Health Initiative study. This study found no reduced risk of CHD or stroke between treatment and control groups. The treatment group lowered their fat consumption by 8.2% and saturated fat by 2.9%. Again, the women in the treatment group were still eating a high fat diet. They were still eating 29% of their diet from fat, and they were still eating loads of cholesterol, so it’s not surprising that there were no differences in disease risk between the two groups. For comparison, the control group averaged a 35% fat diet. The authors wanted the treatment group to hit 20%, so there was very poor compliance.
Then he talks about the A to Z trial in 2007 that compares various diets to each other. He notes that the Atkins diet outperformed the Ornish (plant based) diet on several metrics. It’s worth noting that he’s using erroneous claims about the results of that study. JAMA actually published a retraction on the weight claims for that study saying, “Weight loss was not statistically different among the Zone, LEARN, and Ornish groups. At 12 months, secondary outcomes for the Atkins group were comparable with or more favorable than the other diet groups.”
So at best the Atkins diet is a wash compared to Ornish, but it may actually be worse because the study didn’t run long enough to track mortality or cardiovascular events. Ornish mentions this retraction in his response to The Scientific American, where he refutes many of the arguments Attia is making here. Further, the Ornish comparison group in the A to Z study didn’t even eat the Ornish diet. They were eating a 30% fat diet with 20 grams of fiber a day, which isn’t even close to the kind of diet Ornish recommends eating.
Ornish has proven his diet reverses heart disease, and every time a low carb diet has been tested for the treatment (reversal) of heart disease, it has failed. The only diet proven to reverse heart disease is a low fat, whole food, vegan diet.
Further, we know that a certain subset of the population is at a far greater genetic risk for the development of CVD if they consume a high saturated fat and cholesterol diet. So it is downright dangerous to suggest people eat that way. Some people will do fine, others will drop dead. Biomarkers don’t tell us about disease progression, and the trial didn’t run long enough to look for mortality or cardiac event risk. And it’s not just Ornish who’s reversing heart disease with a plant based diet, there’s a large body of doctors successfully treating various diseases with a plant based diet.
Then he talks about the Diet Trial in 2008, which compared low fat, Mediterranean and low carb diets. The low carb had modestly better outcomes, but again, he’s not giving us all the information. Here’s what the researchers considered “low fat”:
The low-fat, restricted-calorie diet was based on American Heart Association guidelines. We aimed at an energy intake of 1500 kcal per day for women and 1800 kcal per day for men, with 30% of calories from fat, 10% of calories from saturated fat, and an intake of 300 mg of cholesterol per day.
So a severely restricted calorie diet, 30% fat and 300 mg of cholesterol. For comparison, the average American eats a diet of 35% fat. None of the participants lowered their LDL levels after 2 years of eating any of those diets. All of them were still at high risk for cardiovascular disease, with LDL levels well over 100. All this study did was prove that all three diets are a surefire way to die of heart disease, including the worthless AHA diet. I personally think the AHA should be charged with criminal negligence for recommending people eat that way. The people at the AHA know better.
Then he talks about the 2014 diet trail, which again faces the same problems as the first diet trail. People were still eating 30% total fat and 9% saturated fat in the “low fat” group, no limits on cholesterol intake.
And that’s the extent of “proof” in his total bullshit lecture. It takes a seriously sick individual to twist the results of the studies he cites and ignore the MOUNTAINS of studies that run contrary to his position. And finally, if you want to see what happened to my cholesterol levels on a vegan diet, look here. My LDL levels dropped like a rock, and they are still declining.
It’s worth noting that low carb diets have been proven to be just as effective for weight loss as vegan diets. Weight loss by any means, including a bad methamphetamine habit or from chemotherapy, will improve blood lipid profiles and other metabolic disease metrics. However, that does not mean a low carb diet is a good diet for long term health.
When scientists actually measure the blood flow in people eating a diet high in saturated fats, they show signs of worsening atherosclerosis. The same is not true of people eating a low fat high carb diet. To quote one study that compared the two diets:
In this chronic study, we found that SFA [saturated fat] impaired endothelial function and that subjects had a marked increase in TG [triglycerides] and a fall in HDL-C on the low-fat CARB diet without an effect on FMD [arterial function]. PUFA [polyunsaturated fat] and MUFA [monounsaturated fat] diets compared with SFA reduced P-selectin concentrations. These results lend further support that a high-SFA diet is atherogenic through its adverse effect on endothelial function and P-selectin levels.
Ornish found the same in his study results. People on the low fat plant based diet had “worse” HDL and triglyceride levels, but when he actually measured the plaque in their arteries, he could see their atherosclerosis was reversing. Here’s another study that found the same thing with high protein diets.
The evidence against eating animal products and low carb diets is simply overwhelming. Harvard researchers estimate 1 out of 3 early deaths could be prevented by people simply not eating meat.
Studies on children and adults have shown increased arterial stiffness on low carb diets. Saturated fat has been directly observed to cause inflammation, as well as impair endothelial function. A high saturated fat meal also leads to an acute post-meal increase in circulating endotoxins, which is especially terrible for diabetics. Saturated fat promotes lipotoxicity through inflammation which leads to insulin resistance.
We also know that meat, in particular processed meat, is a known carcinogen. The current evidence for the carcinogenicity of processed meat places it in the same category of evidence as cigarettes. Animal protein is known to increase IGF-1, a cancer promoting hormone. Several forms of cancer, such as colon cancer and lymphoma, have been directly linked to meat consumption.
It’s also worth noting that when people are placed on ketogenic diet, their athletic performance suffers. Given that we know high fat diets impair endothelial function, stiffen arteries, reduce blood flow and cause hyperlipidemia, it makes sense that a randomized control trial would show a reduction in athletic performance for people on a ketogenic diet.
Here’s a study showing the impact of an ad libitum (eat as much as you want) whole food vegan diet on disease markers after just 7 days. The study notes that, “The median weight loss was 1.4 kg. The median decrease in total cholesterol was 22 mg/dL. Even though most antihypertensive and antihyperglycemic medications were reduced or discontinued at baseline, systolic blood pressure decreased by a median of 8 mm Hg, diastolic blood pressure by a median of 4 mm Hg, and blood glucose by a median of 3 mg/dL. For patients whose risk of a cardiovascular event within 10 years was >7.5% at baseline, the risk dropped to 5.5% at day 7.” – so that’s a massive improvement in all disease markers after just a week on an unrestricted low fat whole food vegan diet.
Here’s another study showing a vegan diet, where portion sizes, energy intake and carbohydrate intake were unrestricted, did better than a calorie restricted American Diabetes Association diet at getting blood glucose and lipids under control. Dr. Greger explains exactly how saturated fat impairs insulin sensitivity in this short video.
In conclusion, the evidence shows that low carb high fat diets are atherogenic, cancer causing and do not lead to better long term weight loss or disease prevention when compared to other forms of diets, even though they may improve blood lipid profiles due to the weight loss they induce. On the other hand, whole food plant based diets have been proven to reverse atherosclerosis, reverse some forms of cancer, reverse type 2 diabetes and lower chronic disease risk.
Now if you want some real science, watch this:
Oh, one more thing. When this article was brought to Attia’s attention, this is what he had to say:
The title of his post is “Dr. Peter Attia: Hack, Liar and All Round Disgusting Individual.” Really? Wow, tell me how you really feel. When last I checked someone who opens with an ad hominem attack isn’t worth the ether they occupy.
You’re free to judge for yourself if that’s a worthy response to the mountain of scientific criticism I’ve leveled against his garbage. He knows he’s an industry funded shill. How people like that can look themselves in the mirror every morning is something I’ll never understand.