Eleven days ago, I posted an article about one of the most appalling pieces of epidemiological hogwash I’ve seen for some time. Published in the journal Atherosclerosis, the study claimed that eating three eggs per week was almost as bad for your arteries as – wait for it – smoking cigarettes!
While this ludicrous claim was readily soaked up by a scientifically illiterate media and embraced by those militant purveyors of “plant-based” pseudoscience – vegans – it has attracted widespread criticism pretty much everywhere else.
And rightly so. The study contains so many flaws and discrepancies it should be used in science courses as an example of how not to craft a journal paper.
I’ll expand on this a little later, but for now, let’s take a closer look at the researchers who authored the atrocious Atherosclerosis paper. These researchers have a history of anti-egg crusading and, in their own minds, strongly believe the egg industry exerts a powerful and improper influence on Canadian and US dietary recommendations.
Researchers in Glasshouses Shouldn’t Throw Stones
The Atherosclerosis paper was authored by Canadian researchers J. David Spence, David Jenkins, and Jean Davignon. Before we learn more about who funds these men and which questionable organizations they are aligned with, it bears reiterating they all appear to harbour a strong belief the egg industry is a sinister conglomerate hell-bent on propagandizing away the nasty truth about its products.
Here’s a previous anti-egg paper by these researchers. You’ll see the text of the paper begins by making repeated and dismissive references to the “propaganda” of the Canadian egg industry, but then goes on to discuss the cholesterol recommendations of the NCEP and American Heart Association as if they are of unquestioned veracity.
And if you have a spare 18 minutes and 37 seconds, below is a video interview with Spence. If you don’t have this amount of free time, or don’t care to listen to all the off-topic stuff that has nothing to do with Spence’s egg research, then fast forward to 3.12, 5.50 and 17.48. These segments reveal quite a lot about Spence’s mindset.
Spence just can’t help himself when it comes to making snide remarks about the egg industry, deriding their PR efforts every chance he gets.
Reality check: For better or worse, virtually every sector of the food industry has its own lobby groups, whose primary roles are to portray their products in the most favourable light, and garner favour from media, government, health organizations and public alike. While Spence, Jenkins, and Davignon would like you to believe that lobbying, PR campaigns, and funding of research are unique to the egg industry, the truth is these activities are conducted by everyone from rice producers to peanut farmers to citrus fruit growers. To single out the egg industry for doing exactly what every other food industry sector does, and cite this as evidence of some kind of uniquely devious impropriety, is most disingenuous.
More than disingenuous, it’s downright hypocritical considering Spence and Davignon have themselves received funding from manufacturers of cholesterol-lowering drugs:
“Dr Spence and Dr Davignon have received honoraria and speaker’s fees from several pharmaceutical companies manufacturing lipid-lowering drugs, and Dr Davignon has received support from Pfizer Canada for an annual atherosclerosis symposium; his research has been funded in part by Pfizer Canada, AstraZeneca Canada Inc and Merck Frosst Canada Ltd.”
Spence has also received grants from the Heart & Stroke Foundation of Canada (Ontario), which in turn receives millions of dollars annually from manufacturers of cholesterol-lowering drugs and foods. Especially generous ‘donors’ include:
Pfizer Canada Inc. (manufacturer of Lipitor)
$500,000 – $999,999
Unilever Canada Inc. (manufacturer of cholesterol-lowering margarines)
$100,000 – $499,999
AstraZeneca Canada Inc. (manufacturer of Crestor).
Spence is firmly entrenched in a world, awash with money from vested interests, where cholesterol is uncritically viewed as a primary cause of CVD and where the alleged life-saving benefits of cholesterol reduction are considered a given. Little surprise then, that while Spence derides the PR activities of the egg industry every chance he gets, he makes no mention of the blatant conflict of interest inherent in the NCEP panels that set the cholesterol guidelines he cites. These panels have repeatedly come under fire for being stacked with researchers who receive money from lipid-lowering drug manufacturers. These same researchers have consistently lowered the ‘therapeutic’ cholesterol targets over the years, effectively creating literally millions of potential new customers for their drug company masters.
And while he belittles the egg industry for having the temerity to try and create a favourable impression of its wares, he has absolutely nothing to say about the vast sums of money received by the American Heart Association from manufacturers of cholesterol drugs, nor its dubious ‘Heart Check’ program which awards official AHA ticks of approval to such questionable foods as Cheerios, sugar-rich cereals, crackers, flavoured soymilks, popcorn and pancake mix, so long as the manufacturers fork over several thousand dollars per year per product.
Yep, as is par for the course among dietary sectarians, financially-induced bias only ever occurs in the enemy’s camp. It’s quite OK for Spence and Jenkins to receive money from companies massively vested in the anti-cholesterol paradigm, and it’s quite OK for the organisations who concoct official cholesterol targets to harbour similar blatant conflicts of interest. But when the egg industry tries to PR its way out of a totally unfair situation it should never have found itself in in the first instance – being the subject of fraudulent claims its products cause heart disease and stroke – Spence, Davignon and Jenkins carry on like it’s some kind of massive conspiracy.
Love the double standard, fellas!
The dubious associations don’t end with money received from cholesterol drug manufacturers. The remaining author, David Jenkins, is the creator of the essentially vegan “Portfolio Diet” which encourages people to discard wholesome meats, eggs, and dairy foods in favour of such troublesome pap as estrogenizing soy foods, anti-nutrient-laden whole grains, and sterol-rich margarines.
Jenkins also has close ties with the Physicians Committee for Responsible Medicine, having been a co-investigator in one of their studies and the recipient of a PCRM award for “Compassion in Medicine”. The membership of this so-called ‘Physicians’ committee in fact consists of only 10% from the medical profession; the remainder are mainly activists and extremists associated with groups such as the Animal Defense League, Animal Liberation Front, Earth Liberation Front, and SHAC. Here’s some interesting reading on just what the PCRM is really about:
Excuse my skepticism, but I can’t help but view a bunch of vegan activists masquerading as medical professionals – with ties to terrorist groups – with anything but the utmost suspicion. And I have to seriously question why any researcher who even begins to take himself seriously would have anything to do with such a dubious outfit.
Spence, Davignon and Jenkins: Sponsored by Big Pharma and, in the latter instance, aligned with at least one militant vegan activist group. This, folks, is the research trio that would have us believe the egg cartel is a heinous, heartless profiteering juggernaut and that the humble egg is almost as deadly as cigarette smoking.
Thanks to consumer demand, you can now eat free range eggs, organic eggs, omega-3 eggs, barn-laid eggs and RSPCA-approved eggs, but my personal favourite is Naked Babe Eggs. No additives & 100% organic, just like nature intended!
Never Mind the Bollocks, Here’s the Science
OK, enough of the blatant hypocrisy displayed by the authors, let’s take a look at the actual science. My recent article highlighted a string of glaring flaws inherent in the Spence et al study.
— Ignoring all other foods and looking at egg consumption in isolation, despite the fact that probably no human on this planet eats a 100% egg diet. Yes, you read that right – no other aspect of the subjects’ diets were assessed!
–Ignoring a number of other important confounding factors that epidemiologists would normally consider, including exercise, waist circumference, and alcohol consumption.
–Shunning a raft of other critical factors routinely ignored by epidemiologists in general, including but by no means limited to psychosocial stress, bodily iron stores, n-3:n-6 ratios, antioxidant status, exposure to environmental pollutants (something that can vary widely according to past and present occupation), and illicit and prescription drug use (past and present). All these factors have been shown in previous research to be associated with CVD risk but, unless the explicit subject of the study, are rarely included in the multivariate analyses performed by epidemiologists.
–Seriously believing that aging folk who’ve recently suffered a stroke are able to accurately recall the exact number of eggs they averaged every week throughout their entire life. This data was the foundation upon which the study was built, and it was obtained from a questionnaire given to people who just suffered an event in which brain damage of varying severity can and often does occur. This factor alone should have seen the study tossed in the garbage and the researchers laughed out of the research arena. Instead it passed peer review, was published in a prominent journal, and made headlines around the world.
Bless the rampant stupidity of my fellow humans.
–Failing to include a healthy control group. Normally when researchers conduct this kind of retrospective slop, they at least make the effort to include a healthy control group (these types of studies are referred to as “Case-Control”). Spence et al evidently saw no need for such trifling things as control subjects, and simply gathered the highly selective data they were after from the cases alone.
–Studying risk factors while ignoring clinical endpoints. Spence and his co-authors chose to measure carotid total plaque area because, as he notes in this letter, some researchers have found it “is a strong predictor of the risk of heart attacks and strokes”. This point bears repeating because it’s extremely important: The only reason anyone give’s a rat’s ass about carotid plaque volume is because it may help to predict the likelihood of heart attacks and strokes. Well guess what? Every single patient in Spence et al’s study had already suffered either a transient ischaemic attack or stroke! Heck, that’s how they came to be in the study in the first place! So, all incessant wanking on about egg yolk years and plaque volume aside, the reality is every patient in this study had suffered a stroke of varying severity, regardless of whether they ate zero eggs or fourteen eggs per week. So if we’re concerned not with secondary markers like plaque volume but primary outcomes like actual vascular events, this study in fact tells us bugger all about the impact of eggs on stroke risk. In fact, the only valid conclusion we could draw from this study is that the number of eggs one claims to eat each week x the number of years one claims to have been eating eggs has no relationship with stroke risk. If the authors object to this then, again, they need to drop the epidemiological bollocks and organize themselves a randomized controlled clinical trial involving individuals assigned to egg-free and egg-containing diets, who are then followed for several years, and whose incidence of cardiovascular events and mortality is then tallied up and compared. My forewarning to the researchers: So long as this trial is conducted properly and analyzed impartially with the proper blinding in place, be prepared to have your anti-egg hypothesis smashed to pieces.
–Ignoring the indisputable fact that association does not equal causation. This is one of the most basic precepts of good science, yet despite the fact that by their very nature epidemiological studies cannot prove causation (this is the province of controlled clinical trials), the authors went ahead and confidently proclaimed a causal role for eggs in the development of cardiovascular disease.
You’d think Spence and his colleagues had already made up their mind about the deadliness of eggs.
And you’d be correct.
The Spence, Jenkins and Davignon paper I linked to earlier was published back in 2010 and was titled “Dietary cholesterol and egg yolks: not for patients at risk of vascular disease”
They claim in the abstract of this paper:
“Dietary cholesterol, including egg yolks, is harmful to the arteries. Patients at risk of cardiovascular disease should limit their intake of cholesterol. Stopping the consumption of egg yolks after a stroke or myocardial infarction would be like quitting smoking after a diagnosis of lung cancer: a necessary action, but late.”
It’s pretty clear the authors had already decided back in 2010 that dietary cholesterol and eggs in particular were bad news. Is it really any surprise their recent paper was so blatantly one-sided?
Analysing an Absolutely Asinine Assortment of Half-Baked Hogwash
Spence, Jenkins and Davignon repeat a number of key arguments in their assortment of anti-egg/anti-cholesterol material. Namely:
Let’s dismantle these allegations one by one.
The Animal Non-Argument
Argument #1 is a dismally poor one, on several counts. Monkeys, pigs and other species routinely cited by the anti-cholesterol crowd such as rodents and rabbits all have one thing in common – they are primarily herbivorous species. Feeding high-cholesterol or high-animal fat diets to these species has indeed produced ‘fatty deposits’ in their arteries. In stark contrast, studies with carnivorous animals such as cats, dogs and foxes have failed to reproduce the same results. Cholesterol-feeding experiments with dogs completely fail to induce arterial deposits unless their thyroid glands are either removed or neutralized with thyroid-suppressing drugs beforehand. High amounts of cholesterol appear to be readily metabolized by carnivorous animals, whereas herbivorous animals may not be equipped to metabolize large amounts of dietary cholesterol or animal fat, both of which are absent from plant foods.
What’s more, the fatty deposits seen in susceptible animals’ arteries differ significantly from atherosclerotic lesions seen in human CHD patients. Rabbit lesions resemble those of early stage human arteriosclerosis, but they do not develop into more advanced types of plaques. The distribution pattern of these lesions also differs considerably from humans, occurring at locations in the aorta uncharacteristic of human arteriosclerosis (human atherosclerosis tends to develop in locations subject to the highest sheer stress as blood pumps through).
Perhaps most telling of all is the fact that when rabbits are force-fed cholesterol and fat-rich animal foods, their blood cholesterol values skyrocket to astronomical levels, far higher than those ever seen in human beings. Cholesterol does not just accumulate in their arteries but permeates the organs, causing fatty build-up in the kidneys and liver, yellowing of the eyes, and loss of fur. Unless sacrificed beforehand, these animals die from a loss of appetite that leads to emaciation and starvation, not from a heart attack.
Read those last 5 words again, in case you missed their overriding significance. Feeding cholesterol to animals does not cause heart attacks. It might clutter the arteries of herbivorous animals with wads of cholesterol they never evolved to metabolize, but it does not cause myocardial infarctions. Citing these types of experiments as evidence that cholesterol causes heart attacks in humans really is quite dumb.
Especially considering the fallacy of extrapolating the results of such studies to humans was extensively highlighted way back in 1935. Nevertheless, in the proud, worldwide tradition of staunch dogmatists impervious to conflicting evidence, many researchers – including Spence, Davignon and Jenkins – continue to cite these irrelevant animal studies as ‘proof’ that saturated fat and cholesterol are damaging to humans.
In members of the omnivorous Homo sapiens species (a.k.a. “human beings”), cholesterol consumption does not cause the astronomical blood cholesterol elevations seen in rabbits. In fact, dietary cholesterol has hardly any effect on blood cholesterol at all because the body readily accommodates changes in dietary cholesterol intake simply by increasing or decreasing its own output as required. An extensive review of 167 cholesterol-feeding studies confirms that the increase in blood cholesterol induced by dietary cholesterol is negligible and has no association with CHD risk.
More Monkey Business
Monkeys further highlight the futility of extrapolating animal cholesterol-feeding responses to humans. Because monkeys are primates, and primates are our closest animal cousins, the lipid hypothesists believe any agent that causes ‘atherosclerosis’ in monkeys must surely do the same in humans.
In experiments with African Green monkeys, researchers have repeatedly found that omega-6-polyunsaturated fats cause less atherosclerosis than saturated and monounsaturated fats[4-6]. If monkey experiments were a valid proxy for humans, this would mean that replacing monounsaturated and saturated fats with polyunsaturated omega-6 fats reduces atherosclerosis and hence heart disease. As anyone who’s read The Great Cholesterol Con will know, both autopsy studies and decades of dietary intervention trials show that nothing could be further from the truth.
The Los Angeles Veterans Administration Study compared the effect of a diet high in polyunsaturated fats (from soybean oil) and low in saturated fat with that of a control saturate-rich diet among 846 institutionalized veterans. The researchers observed a reduction in CHD deaths amongst those eating the polyunsaturated-rich diet. However, a significant increase in cancer deaths amongst the diet group entirely negated the reduction in CHD mortality. After eight years, the overall death rate in both the diet and control groups was almost identical. And this was despite the fact that there were more non-smokers in the soybean oil group, and a significantly higher number of heavy smokers in the control group.
When autopsies were performed on the deceased subjects the researchers found that there was very little difference in the degree of atherosclerosis between the two groups. If anything, those in the soybean oil diet group, despite having lower serum cholesterol levels, had slightly more plaque build-up in the aorta, the main artery carrying blood away from the heart.
Polyunsaturated omega-6 fats might prevent ‘atherosclerosis’ in monkeys, but they do no such thing in humans. The only thing they do achieve is to counter the increased CVD risk from smoking with an increased risk of cancer. Most unawesome. And a most ridiculous scenario for those who cite monkey cholesterol-feeding experiments to find themselves in. If they are going to claim monkey studies are relevant to humans, then for the sake of consistency they should start recommending humans not only eat less cholesterol but more harmful omega-6-rich vegetable fats.
Oh, and one last thing to really blow Spence et al’s animal argument to smithereens: When researchers pull cholesterol from the diets of rhesus monkeys, but do not reduce the high level of saturates they are consuming, they observe a significant drop in cholesterol and regression of atherosclerosis in the animals. This indicates that the cholesterol-raising and atherogenic factor is cholesterol itself, not saturated fat. In your typical human, however, dietary cholesterol has almost no effect on serum cholesterol, so why should we believe it has any effect on atherosclerosis – especially when dietary cholesterol restriction, just like saturated fat restriction, has completely failed to lower CHD mortality in humans?
Bottom line: The animal argument is both self-contradictory and utterly irrelevant.
Dodgey Arguments #2 and #4: Epidemiological Studies Show Dietary Cholesterol Increases CHD in Healthy Folks and Eggs Increase Mortality in Diabetics
According to Spence et al, epidemiological studies show that dietary cholesterol increases CHD risk in humans. In attempt to confirm this, they cite two studies (the Western Electric and Ireland-Boston Diet-Heart studies) published back in the early 80s. And in an attempt to discredit the egg industry’s argument that the totality of evidence shows no harmful effect of egg consumption, they cite two studies in which egg consumption reportedly raised mortality among diabetics.
What Spence, Davignon and Jenkins would no doubt prefer you did not know is there have been a heck of a lot more than a mere four epidemiological studies examining this topic. And not only have there been many more such studies, there have also been a number of systematic reviews summarising the findings of these studies.
Here’s a quote from one such review by fellow Canadian Peter Jones, from the University of Manitoba, Winnipeg:
“For many years, both the medical community and the general public have incorrectly associated eggs with high serum cholesterol and being deleterious to health, even though cholesterol is an essential component of cells and organisms. It is now acknowledged that the original studies [my note: this includes the Western Electric and Ireland-Boston Diet-Heart studies eagerly cited by Spence et al] purporting to show a linear relation between cholesterol intake and coronary heart disease (CHD) may have contained fundamental study design flaws, including conflated cholesterol and saturated fat consumption rates and inaccurately assessed actual dietary intake of fats by study subjects. Newer and more accurate trials, such as that conducted by Frank B. Hu of the Harvard School of Public Health (1999), have shown that consumption of up to seven eggs per week is harmonious with a healthful diet, except in male patients with diabetes for whom an association in higher egg intake and CHD was shown. “
In a review specifically addressing eggs, Kritchevsky states:
“For much of the past 40 years, the public has been warned away from eggs because of a concern over coronary heart disease risk. This concern is based on three observations: 1. eggs are a rich source of dietary cholesterol; 2. when fed experimentally, dietary cholesterol increases serum cholesterol and; 3. high serum cholesterol predicts the onset of coronary heart disease. However, data from free-living populations show that egg consumption is not associated with higher cholesterol levels. Furthermore, as a whole, the epidemiologic literature does not support the idea that egg consumption is a risk factor for coronary disease.”
As Krischevsky notes, many of the earlier studies claiming an association between eggs and CVD didn’t even bother examining let alone adjusting for other confounding dietary factors.
Not to be deterred by this total lack of reputable epidemiological support for their thesis, Spence et al make much ado about the fact that, while eggs show no relationship with CVD in the general population, they have been associated with increased CVD risk among diabetics.
This claim is derived from the findings of The Harvard Egg Study headed by Hu and a Greek study by Trichopoulou et al.
Stupid Question (Except, Evidently, to Epidemiologists): Do Eggs Kill Diabetics – or is it the Diet and Lifestyle that Predisposes to Diabetes that Kills Diabetics?
The Harvard Egg Study followed 37,851 men aged 40 to 75 years and 80,082 women aged 34 to 59 years at study outset, initially free of CVD, diabetes, hypercholesterolemia, or cancer. The men were followed up for 8 years, and women for 14 years.
After adjustment for age, BMI, cigarette smoking, parental history of myocardial infarction, multivitamin supplement use, vitamin E supplement use, alcohol consumption, menopausal status/postmenopausal hormone use, history of hypertension, physical activity, and total energy intake, the authors found no evidence of an overall significant association between egg consumption and risk of CHD or stroke in either men or women.
The relative risks (RR) of CHD across categories of egg intake were:
Less than 1 per week (1.0);
1 per week (1.06);
2 to 4 per week (1.12);
5 to 6 per week (0.90).
(P for trend = .75).
Less than 1 per week (1.0);
1 per week (0.82);
2 to 4 per week (0.99);
5 to 6 per week (0.95).
(P for trend = .95).
And despite Spence et al’s claim that egg consumption accelerates carotid atherosclerosis, the results were similar for stroke. As in the large Hiroshima study, no association was observed between egg consumption and risk of ischemic or hemorrhagic stroke.
Spence et al’s case is looking weaker than an anemic vegan’s handshake. But what about the increased risk in diabetics that they wank on and on about? Does this salvage their theory in the face of a virtual avalanche of conflicting evidence?
Let’s take a look at just what Hu et al had to say about this relationship:
“In subgroup analyses, higher egg consumption appeared to be associated with increased risk of CHD only among diabetic subjects (RR of CHD comparing more than 1 egg per day with less than 1 egg per week among diabetic men, 2.02 […trend = .04], and among diabetic women, 1.49 […trend = .008]).”
Notice how the authors use conservative language and, unlike Spence et al, are careful not to triumphantly pronounce a causal association. Indeed, Hu et al further note:
“This result should be interpreted cautiously due to numerous subgroup analyses, but the consistency of the association in the 2 cohorts argues against a chance finding.”
Speculating as to the reason behind this finding, the authors write:
“The increased risk may be related to abnormal cholesterol transport due to decreased levels of apolipoprotein E37 and increased levels of apolipoprotein CIII38 among patients with diabetes.”
Rather than speculate on the role of exotic apolipoproteins that have never been shown to play a causal role in the development of atherosclerosis in diabetics (or non-diabetics for that matter), I’ll propose an alternative explanation drawing, not upon esoteric lipoprotein wankology, but the vastly under-deployed trait of commonsense.
Using commonsense, we acknowledge that diabetes is a disease of disordered blood sugar metabolism. We acknowledge that in the case of Type 2 diabetes (the most common form), such factors as physical inactivity, overweight/obesity (which in turn is caused by excess caloric consumption and/or insufficient activity), high refined carbohydrate consumption, and insufficient intake of key minerals such as magnesium and chromium may all contribute to the development and progression of the disease. We also acknowledge that high iron stores are quite common in diabetics and that clinical trials have shown reduction of these iron stores via phlebotomy or deferoxamine often brings about significant improvements in glycemic control, blood pressure and arterial function.
Unlike esoteric lipoprotein subfractions, many of these same factors have also been documented to play a key role in the pathogenesis of CHD.
Now, before I tie all this together, I want to point out one more thing. Over the last several decades, the industrialized world has been bombarded with so much anti-cholesterol, anti-saturated fat, anti-red meat propaganda that most health-conscious people make a concerted effort to limit or even totally avoid intake of these dietary elements.
While the benefits of reducing or avoiding consumption of these foods are highly questionable, health-conscious people are more likely to also display a number of other traits about which the abundant benefits have been well established. These include avoidance of cigarettes and other recreational drugs, increased physical activity, increased consumption of fruits and vegetables, and a much lower incidence of overweight.
On the flipside, individuals with little regard for their health tend to remain largely indifferent to mainstream health recommendations, be they pertaining to smoking, exercise, meat/egg intake, fruit and vegetable consumption, and cholesterol/saturated fat ingestion.
So let’s see what the Harvard researchers found when they looked at this very issue:
“Egg consumption was positively associated with smoking, lower physical activity, and a generally unhealthy eating pattern (ie, more whole milk, red meat, and bacon and less skim milk, vegetables, and fruits) in men. Confounding due to these factors would artifactually produce an elevated risk for egg consumption. As expected, an apparent positive association with higher egg consumption in the age-adjusted analysis in men was attenuated after adjustment for smoking and other covariates. After further adjusting for bacon intake, which was positively associated with risk of CHD in our cohorts, the RRs became weakly inverse. This speaks to the importance of considering overall eating patterns when examining the effects of egg consumption.”
Now remember, this comment was made about the men in the study as a whole. While details of the diabetic sub-analyses were not included in the paper, we know that the main analyses only factored in age, BMI, smoking, parental history of myocardial infarction, multivitamin and vitamin E supplement use, alcohol, menopausal status/postmenopausal hormone use, hypertension, physical activity, and total energy intake.
Factors such as refined carbohydrate intake, low fruit and vegetable intake, bodily iron stores, deficiency of critical minerals such as magnesium, and successful adherence to glycemic control protocols among diabetics do not appear to have been included in the multivariate analyses (and I’m quite confident the last three were never even considered).
So…using the all-too-often dormant faculty of commonsense, do you think someone who ends up diabetic just might be more likely to have a history of eating not just more eggs, but of eating a poorer diet overall and living an unhealthier lifestyle? And that the inability of epidemiological analyses to fully account for such factors may have swayed the results against eggs and produced an association with eggs and CHD in diabetics that was actually an artefact of other confounding factor/s?
Unless your brain has been totally dulled by the B12 deficiency that inevitably arises from years of meat avoidance, you no doubt answered in the affirmative to that question.
I will repeat: In all likelihood, the relationship between eggs and CHD in diabetics in the Hu study was an artefact of other confounding factors. To claim a causal role for eggs in diabetic CHD using this confounder-prone statistical association – as Spence et al do – is both premature and plain poor science. Such a causal claim can only be made from clinical trials; more on that later.
First, let’s take a look at the Greek study by Trichopoulou et al. This is also repeatedly cited by the Spence-led trio as ‘proof’ that eggs cause all manner of cardiovascular grief for diabetics.
From 1993 to mid-2004, using 28,572 volunteers in the Greek arm of the huge EPIC study, researchers isolated 1,013 subjects initially taking drugs for T2D and followed them for a mean 4.5 years (the researchers excluded 509 subjects who indicated their diabetes was managed through diet alone “because they were considered as having unconfirmed diagnosis”). All the included subjects were also taking cholesterol or hypertensive medications.
So this was a much smaller project than the Harvard Egg Study. As for the methodology employed in the study, dietary intake during the year preceding enrollment in the study was assessed via a questionnaire.
And it is here that we run into the first problem with this study (and pretty much all epidemiological nutrition studies): The highly questionable quality of self-reported data.
When you look at Table 2, you can see that the claimed daily energy intakes for men and women in the study were 1,842 and 1,528, respectively. Needless to say, these are very low energy intakes and, if correct, would endow this population with a significantly lower BMI than the general population. However, if you scroll back up to Table 1, you can see that 885, or a full 87%, of the participants in the study were officially overweight (i.e. BMI of 25 or greater)!
Given there exists no evidence whatsoever that individuals of Greek heritage are prone to higher than normal levels of overweight despite consuming far lower than normal daily caloric intakes, we can quite confidently assume the daily caloric intakes reported by the participants were wildly inaccurate. And given that study after study shows dietary misreporting is par for the course among self-reporting participants, this assumption is not at all unreasonable – it is in full accord with the available science.
The irreconcilable disparity between the self-reported daily caloric intakes and the BMI data becomes even more pronounced when one considers the vast majority of this population claimed to be physically active to the tune of at least 30 MET-hours daily (approximately equal to 1 hour of brisk walking or 30 minutes of jogging or biking every day).
So if the caloric data were bollocks, why should we trust the rest of the dietary data?
Especially when it returned data claiming that every extra 10 grams of daily egg intake increased the relative risk of death among diabetics by 34%, while sugar and soft drinks returned non-significant reductions in death risk.
And every 60 grams of cereal products, which included “flour, flakes, starches, pasta, rice, other grain, bread, crisp bread, rusks, breakfast cereals, biscuits, dough and pastry, etc.” was associated with a 29% relative risk reduction in mortality – and in this instance the p level was significant at 0.04.
If you’re face isn’t contorting into that “Huh?!” look right now…it bloody well should be.
Let’s be perfectly clear – diabetics have a greatly attenuated lifespan and suffer higher rates of everything from infection to blindness to cancer to heart disease because their blood sugar metabolism is royally screwed. Diabetics die from poorly controlled blood sugar and its complications, not from cholesterol.
Everyone except bloody-minded epidemiologists knows that diabetes is a disease of dysfunctional blood sugar metabolism, and that reducing the GI and/or glycemic load of the diabetic diet often greatly improves markers of glycemic control and even allows for reductions in medication. Yet if we are to believe this study, the humble egg – which contains almost no carbohydrate – is more harmful to diabetics than the refined sugars and cereals that helped cause their condition in the first instance.
Let’s reflect again on the study methodology. Remember how I said the respondents filled out diet questionnaires at the start of the study? Well, those questionnaires were administered by interviewers. Now, before I explain the possible significance of this, keep in mind that diabetics are routinely admonished to limit their intake of sugars, sweets, soft drinks and often, in this day and age of whole-grain political correctness, white flour products. Eggs, on the other hand, are not subject to anywhere near the same kind of attention in diabetic dietary advice because their GI isn’t just low, it’s pretty much non-existent.
Also keep in mind that when people answer questions administered by a live interviewer, it’s no secret their answers will often be far less honest than when they fill out a survey anonymously. Social desirability bias is the tendency of people to answer questions in a manner that will be viewed favourably by others. It can take the form of over-reporting good behaviour or under-reporting bad behaviour. So…imagine a diabetic who has issues with resisting the lure of the very high-GI foods that they’ve been repeatedly admonished by friends, family and doctors to avoid, and now some researcher is sitting in front of them quizzing them on their consumption of these same foods. Knowing what you know about human nature in general, and the existence of such phenomena as social desirability bias, do you think it’s highly likely that our diabetic interviewee will under-report his/her consumption of sugar- and flour-rich foods in order to avoid feelings of embarrassment, shame and guilt?
To paraphrase Sam Kekovich: “Of course it bloody well is!”
I’ve said it again, and I’ll no doubt say it many more times before my time on this mixed-up planet is over: Epidemiology is a vastly inferior and often totally inadequate mode of research for studying relationships between diet and disease. It’s a sad, sad testament to the woeful state of nutritional science when so many researchers and so-called ‘experts’ hold epidemiology in the same esteem as data from randomized controlled clinical trials, when the typically inaccurate and confounder-prone data of the former cannot even begin to compare with the far more reliable data of the latter.
If preventive nutrition science is to ever have any hope of pulling itself out of its current morass of impotence, and exert efficacy in actually reducing the incidence of diseases like diabetes, cancer and heart disease – something it has so far failed miserably to do and in all likelihood has actually worsened (obesity epidemic, anyone?) – then it needs to drop its terribly misguided obsession with epidemiology and start insisting on testing dietary theories in RCTs before unleashing them on an unsuspecting and all-too-gullible public.
Oh, and paying attention to the decades of already existing RCT data showing a theory to be complete bollocks would also be a huge step forward…
So yes, egg intake has been linked to CHD and overall mortality in diabetics – in two epidemiological studies highly susceptible to confounding factors. And in other epidemiological studies, eggs have roundly failed to show any connection with CHD or mortality in non-diabetic folk except in a small number of older studies where the attempts to adjust for confounding factors were either poor or non-existent.
Needless to say, that’s really not a good foundation for holier-than-thou researchers to attack the egg industry for its alleged unscientific actions…
RCTs are the only appropriate type of study from which we can infer causality about eggs and disease. Before we discuss those trials, let’s take a quick look at Spence et al’s claim that eggs cause all their alleged devastation in the post-prandial period (the several hours after a meal).
Shoddy Argument #3: Post-Prandial Schmandial
The authors repeatedly claim that cholesterol does the most damage to arteries in the post-prandial period. In an attempt to support this claim, they cite a study by Vogel et al, in which healthy volunteers were fed a high-fat meal (900 calories, 50g fat, 14g saturated fat, 255 mg of cholesterol) and an isocaloric low-fat meal (0g fat, 13mg cholesterol) on 2 separate occasions. Before and after each meal, Vogel and his team measured brachial artery diameter, brachial arterial blood flow, and also the amount of brachial blood flow and vasoactivity after a high-pressure upper arm cuff was deflated. These measurements were taken prior to the meals and at hourly intervals for up to 6 hours after each meal.
According to Spence et al, it’s not so much fasting cholesterol we should be concerned with (something they really need to tell their drug company benefactors and the folks at the AHA/NCEP); rather, it’s elevated post-prandial levels we should be fretting about instead. Dietary cholesterol, they claim, causes its damage in the post-prandial period by raising post-meal blood levels of cholesterol.
Most ironic then, that they cite a study showing allegedly negative changes in arterial function after a high-fat meal when that high-fat meal produced similar lack of change in post-prandial blood levels of total and LDL cholesterol as the low-fat meal! Yep, at 2 and 4 hours (the only time-points blood lipid levels were measured) both diets caused little to no change in blood levels of these lipids.
That right there undermines whatever basis Spence et al had for citing the study. Whatever caused the arterial changes in this study, it sure as heck wasn’t a massive dietary cholesterol-induced surge in blood cholesterol.
So what else may have influenced the results? Well, take a look at the two test meals:
Egg McMuffin, Sausage McMuffin, 2 hash brown patties, and a non-caffeinated beverage courtesy of the McDonald’s Corporation.
Kellogg’s Frosted Flakes, skimmed milk, and orange juice.
Geezus. This was a comparison of two vastly different junk food meals of extremely doubtful nutritional quality that tells us absolutely nothing about the health effects of fresh eggs – you know, real eggs not accompanied by several hundred calories of highly-processed, high-GI, refined oil-rich slop.
Before we discard this irrelevant study for good, let’s quickly take a closer look at the arterial findings. This is not the only study that has reported allegedly harmful effects on arteries from high-fat meals, but what exactly are the arterial responses all these researchers are reporting on?
While these researchers (and an eternally-moronic media) would have you believe these studies show high-fat meals start producing atherosclerosis in your arteries as soon as you finish your meal, the reality is they show absolutely nothing of the sort.
All they show is that high fat meals produce different arterial responses when a blood pressure cuff is inflated then deflated at various time points following a meal. But what exactly has this got to do with heart disease? What kind of a tosser sits at his dinner table every night repeatedly inflating then deflating a blood pressure cuff on the hour, every hour, for several hours after a meal?
Most coronary events are caused by the rupture of atherosclerotic plaques and/or arterial spasm, not by miniscule changes in the blood flow response after deflation of a blood pressure cuff that is only ever applied in clinical studies.
Perhaps most tellingly, baseline blood flow and baseline artery diameter – i.e the only two of the tested measurements that would be seen in your average person who never applies a blood pressure cuff in the post-prandial period – were similar after both the high-fat and low-fat meals.
The irrelevancy of these extremely short-term studies to the pathogenesis of CHD, which typically develops over decades and not in the few hours after a single meal, is reinforced by the fact that both dietary cholesterol and dietary fat restriction have proved themselves dismal failures in lowering real-life CHD mortality in dietary intervention trials all around the world (something I discuss at length, trial by trial, in Chapter 8 of The Great Cholesterol Con).
Oh, I almost forgot the second study Spence et al cite in support of their post-prandial thesis. This is the study by Tannock et al that reported “Feeding of egg yolks induces inflammation characterized by elevations of C-reactive protein and serum amyloid A, but this effect is more pronounced in lean individuals without insulin resistance”. Not only does this again rely on acute post-meal changes that likely have no impact on long-term CHD outcomes, and changes whose alleged detrimental effects have not been borne out in cholesterol-lowering dietary intervention trials, but it also constitutes a whopping self-contradiction.
Remember how Spence at al keep harping on about the fact that while eggs have not shown any harmful association with CHD or mortality in normal folks, there are 2 studies showing such a relationship with diabetic folks? Well, now they are citing a study claiming harmful post-meal changes from egg consumption, but primarily in lean, non-diabetic individuals!
Lawdy, lawdy…Spence, Davignon and Jenkins don’t know whether they’re coming or going!
One More Time: Get an RCT Up Ya!
Folks, before I expel the atrociously unscientific work of Spence et al from my consciousness, I want to reiterate one more time the only way to infer causality between a particular food item and a certain health outcome is to test them in a randomized, controlled, clinical trial.
As I stated in last week’s article, there is not a single long-term trial in existence that examined the effect of egg restriction in isolation on subsequent CVD events (nor any other clinical health outcome).
There have, however, been two shorter-term studies examining the effect of egg consumption on arterial function, blood pressure and glycemic control.
In the first study, Yale researchers took forty-nine healthy males and females (mean age 56 years) and assigned them to eat either two eggs or oats daily for 6 weeks in random crossover fashion with a 4-week washout (this means that all participants followed both diets for six weeks each, four weeks apart). Brachial artery function was tested prior to dietary assignment and at the end of each six-week diet phase.
What did they find?
The ability of the arteries to dilate in response to heightened blood flow after having a high pressure arm cuff deflated was stable in both egg and oat groups, and between-treatment differences were not significant (egg 0.96%, oatmeal 0.79%; p > 0.05). There were also no differences in BMI or systolic blood pressure levels after the egg and oat treatments.
In the second study, researchers here in Adelaide recruited sixty-five diabetic folks – yep, diabetic folks, the very population Spence et al would have us believe based on dodgey epidemiology are most susceptible to the allegedly harmful effects of eggs. They assigned these diabetics to consume similar weight loss diets, the only difference being that participants in one group consumed two eggs per day, whereas the other group replaced the eggs with 100 grams of lean animal protein. After 12 weeks, there was no significant difference in weight loss, and all the subjects reduced systolic and diastolic blood pressure. Markers of glycemic control such as HbA1c and insulin also dropped in both groups, while plasma folate and lutein increased more on the egg diet. As the researchers noted:
“These results suggest that a high-protein energy-restricted diet high in cholesterol from eggs improved glycaemic and lipid profiles, blood pressure and apo-B in individuals with type 2 diabetes.” [Bold emphasis mine]
As I pointed out in my previous article, in both of the above studies the egg phases involved consumption of fourteen eggs per week, and produced absolutely no adverse consequences. Yet Spence et al are claiming, in all earnestness, that a mere 3 eggs per week is going to clog your arteries with nearly the same efficiency as cigarette smoking.
What utter rubbish.
I don’t normally devote a second entire article to a study I’ve already recently debunked, but there is something especially odiferous about the recent Atherosclerosis paper. Several things, actually. One is the blatantly hypocritical piety of the authors, who routinely deride the egg industry for its self-serving “propaganda” while they themselves gladly receive money from drug companies and heart associations massively vested in the anti-cholesterol propaganda they themselves propagate.
The second and most alarming thing is the incredibly poor quality of their research, and the fact that this pseudoscientific rot not only passes peer review but is allowed to trigger alarmist anti-egg headlines around the world.
It’s moments like this where I truly wonder how the human race got to where it is, and where it’s going to end up. We are the only species clever enough to create things like airplanes, buildings, computers and Ramones CDs, yet so many other aspects of our behaviour and knowledge are still stuck in the Dark Ages. As I sit typing this on the morning of September 15, a good twelve years into the 21st millennium, millions of my fellow humans are unnecessarily popping muscle-damaging CoQ10-depleting cholesterol pills and smothering their tasteless “multi-grain” breads with sitosterol-rich maragarines in the earnest belief they are preventing heart disease. Never mind that after almost sixty years of this idiotic cholesterol paranoia the incidence of CHD remains largely unaffected and CHD remains the number one killer in most industrialized countries.
We can routinely save people from the clutches of death when they are frantically wheeled into trauma units, but thanks to the corrupting and all-too-pervasive influence of money, we are nowhere near finding a cure for heart disease. And why would we be? There’s simply too much at stake for the big earners in the CHD arena. Who would need the statin drugs that constitute the most profitable class of drugs and reel in tens of billions of dollars each and every year for the drug companies?
And the various Heart “Associations” and “Foundations” that owe their very existence to the ubiquitous prevalence of CHD would literally be out of business overnight. No more billion-dollar annual turnovers, no more Fortune 500-level salaries for their CEOs, no more all-expenses paid travel to plush conferences for upper level staff, no more highly lucrative thinly-disguised kickback schemes that award “Heart Healthy” ticks to foods of questionable nutritional quality, and no more donations and lavish bequeathments from a well-meaning but sadly ignorant public.
Yep, if we discarded the ineffective cholesterol paradigm for the counterproductive sham that it is, a whole lot of deeply vested interests would suffer big time.
Stuff ‘em. That money is more important than the truth, and more important than making ruthlessly objective and earnest efforts to alleviate human suffering and premature mortality is an absolute disgrace. We might be able to build things that other species can’t, but no other species shares our combined exaltation of material wealth and worship of authority at such disastrous expense to their own wellbeing.
And so here we are…members of a species that actively collaborates to perpetuate false beliefs about one of our biggest killers. And while a select group of individuals profit handsomely from these false beliefs, the greater picture is that the incidence of this killer disease continues unabated. Lives that could have been saved if researchers turned their attention away from the cholesterol sham and onto other far more plausible contributors continue to be lost.
Don’t be like Spence, Davignon and Jenkins, and the gullible journalists who took their absurd claims at face value. Instead, engage your faculties of reason and impartiality and think for yourself.
It might just save your life.
FINANCIAL DISCLAIMER: The author of this article has absolutely no ties with the egg industry, financial or otherwise. Nor has he ever received a cent from the beef, dairy, or any other food-related industries. And (unlike Spence and Davignon) he sure as heck doesn’t receive funding from Big Pharma.
1. Steiner A, Kendall FE. Atherosclerosis and arteriosclerosis in dogs following ingestion of cholesterol and thiouracil. Archives of Pathology, 42: 433-444. 1946.
2. Duff L. Experimental cholesterol arteriosclerosis and its relationship to human arteriosclerosis. Archives of Pathology, 1935; 20: 81-123, 259-304.
3. McNamara DJ. The impact of egg limitations on coronary heart disease risk: Do the numbers add up? Journal of the American College of Nutrition, 2000; 19 (90005): 540S-548S.
4. Rudel LL, et al. Compared with dietary monounsaturated and saturated fat, polyunsaturated fat protects African green monkeys from coronary artery atherosclerosis. Arteriosclerosis, Thrombosis, and Vascular Biology, Dec, 1995; 15 (12): 2101-2110.
5. Rudel LL, et al. Dietary polyunsaturated fat modifies low-density lipoproteins and reduces atherosclerosis of nonhuman primates with high and low diet responsiveness. American Journal of Clinical Nutrition, Aug, 1995; 62 (2): 463S-470S.
6. Wolfe MS, et al. Dietary polyunsaturated fat decreases coronary artery atherosclerosis in a pediatric-aged population of African green monkeys. Arteriosclerosis and Thrombosis, Apr, 1994; 14 (4): 587-597.
7. Dayton S, et al. A controlled clinical trial of a diet high in unsaturated fat in preventing complications of atherosclerosis. Circulation, 1969; 40 (Suppl. II): 1-63.
8. Eggen DA, et al. Regression of experimental atherosclerotic lesions in rhesus monkeys consuming a high saturated fat diet. Arteriosclerosis, Mar-Apr, 1987; 7 (2): 125-134.
Anthony Colpo is an independent researcher, physical conditioning specialist, and author of The Fat Loss Bible and The Great Cholesterol Con. For more information, visit TheFatLossBible.net or TheGreatCholesterolCon.com
Copyright © Anthony Colpo.
Disclaimer: All content on this web site is provided for information and education purposes only. Individuals wishing to make changes to their dietary, lifestyle, exercise or medication regimens should do so in conjunction with a competent, knowledgeable and empathetic medical professional. Anyone who chooses to apply the information on this web site does so of their own volition and their own risk. The owner and contributors to this site accept no responsibility or liability whatsoever for any harm, real or imagined, from the use or dissemination of information contained on this site. If these conditions are not agreeable to the reader, he/she is advised to leave this site immediately.