The "Intermittent Fasting Will Causes Heart Disease" Articles Are Bullshit, But Let's Talk About Why

My wife doesn't follow nutrition research and writing the way I do. She doesn't, for example, have email alerts set up to tell her about new studies and publications. So when she becomes aware of a new study or article, I can usually be pretty sure the item in question is filtering out in to the mainstream consciousness. 

Today the item in question is a report from the American Heart Association that engaging in intermittent fasting, the practice of limiting your food consumption to a certain time period, is associated with a significant increase in heart disease. This shocking headline is being reported by the Washington Post, and CNN, and NBC News, and so on and so on. Mainstream consciousness, indeed. These articles suggest that the researches found “that people who adhered to the eight-hour eating plan had a 91 percent higher risk of dying from heart disease compared to people who followed a more traditional dietary pattern of eating their food across 12 to 16 hours each day” and includes quotes from the researches advising those who practice intermittent fasting to be “extremely cautious.” ¹

The major problem with these results though, is that the researchers did not in any way study people who engaged in intermittent fasting. Nor did they make any attempt to

 

Intermittent Fasting Research

Let’s take a moment here to pause and consider what you as a researcher might do if you wanted to assess the effects of intermittent fasting.

The best strategy would probably be to take a bunch of people, record a lot of health data, have them engage in intermittent fasting for some period of time, and measure those same markers again. Bonus points if you have a second comparison group alongside them to eat the same amount of food spread throughout the day. Then you could see if intermittent fasting had any effects on common markers of metabolic health, inflammation, etc.

You will not be surprised to learn that this has been already been done many, many times. This is good, hands-on research that controls and monitors the subjects’ dietary patterns to be sure the intervention in question is being fairly assessed. And this research, it turns out, consistently returns findings that suggest intermittent fasting can improve common markers of metabolic health, insulin resistance, inflammation, etc.^2–6 Not every trial finds improvements in all the health markers they analyze, but they very often do. And since I’ve yet to find a fasting intervention that demonstrated a worsening of chronic health markers, I feel pretty damn confident in saying that the actual hard science demonstrates a tendency for intermittent fasting to improve chronic health.

 

Intermittent Fasting Surveys

So if we wanted the best results possible, we’d engage in hands-on interventional research. Hard science. But hard science is costly, labor-intensive, and time-consuming. So for those or other reasons, maybe we just want to get a peek at what’s happening to those who engage in intermittent fasting, especially those who have been practicing for a while. What would we do then? We’d ask them! You could ask people how long they’ve been practicing intermittent fasting, how often they engage in the practice, how long their typical eating window is, what foods they typically eat, etc. You’d also want to ask anything else you could imagine related to their health (exercise, alcohol, sleep, lifestyle and socioeconomic factors, etc.) to make sure none of those might influence your results. Then when you have all this data, you could compare the current health of your subjects to a control group or the general population to see if you could identify any differences.

This approach is generally weaker than the interventional trials we described above because they can’t demonstrate any causal effects, but if done properly there can still be reasonable value derived from the findings.

“Intermittent Fasting” Surveys

We’ve covered a couple reasonable approaches for how we might assess the question at hand if we wanted decent to good results. But what if we didn’t give a single flying fuck about good results? What would we do then? What if all we wanted was to attach a soundbite to a trending health topic? What if we just wanted to manufacture a scary headline to make it look like not eating for a little bit could literally cause a heart attack?

Well, we can rule out interventional trials. We know with great certainty that those won’t work, because those have repeatedly demonstrated that intermittent fasting is far more likely to be beneficial than harmful. So we’ll have to go the survey route instead. But, we probably want to be careful about the questions we ask. If we ask a bunch of detailed questions that accurately identify subjects deliberately engaged in the practice of intermittent fasting, we might accidently confirm the hard science if we end up with people who are healthier than the general population.

So I have an idea. We’ll find thousands of people and just ask them what they ate yesterday, and when they ate it. And then later this year we’ll ask them again – what time did you consume food yesterday? Then we’ll take this sample size of two and categorize anyone who ate their food in a single eight hour window as someone engaged in “intermittent fasting,” even if they have no idea what those words even mean. And of course we’ll pretend this sample size of two accurately reflects their long-term dietary patterns while we sit around and wait to see who dies over the next several years.

 

Deliberating Avoiding Actual Intermittent Fasting

Hopefully it’s fairly clear that the sarcastic paragraph above actually just describes the study in question.⁷ It was two 24 hour dietary recalls in the span of a year, and years of tracking deaths. The researchers were not seeking out people engaged in regular intermittent fasting or attempting to understand anybody’s long-term dietary habits, even though every damn news article includes references to “people who practice intermittent fasting for long periods of time.”

But why does this produce terrible results? For the same reasons you can’t draw definitive negative conclusions from survey data, I actually can’t say definitely. Buuuuuuut I have a pretty good guess. Take a look at the chart below, which filters the subjects by their average eating window during those two dietary recalls:

 

A couple things jump out to me right away – those that they characterize as engaged in intermittent fasting (8 hour or less eating window) are quite a bit more likely to smoke than the average person, and are 3 times more likely to be black. This suggests a couple of things to me. One if that a good number of these people are obviously not actively seeking the health benefits of intermittent fasting. And the other is that, unfortunately, this group of subjects is almost certainly at a socioeconomic disadvantage compares to the others. It remains a regrettable reality in this country that black Americans are twice as likely to live in poverty and face food scarcity.^8,9 Meanwhile, smoking is nearly twice as common below the federal poverty line as it is above. ¹⁰ We don’t literally have it from what the researchers presented, but we do actually know it – the “intermittent fasting” group is of far lower socioeconomic status than the others.

I can only think of a couple reasons someone would engage in deliberate intermittent fasting. Weight loss and health benefits are the obvious ones. Or maybe that’s really just one reason. Regardless, people who seek out the deliberate practice fasting are relatively privileged enough to be doing so.

But let’s reframe it – what are some reasons a person might skip a meal? Because to a large extent that’s what is actually happening here. Even just on its face, I would frankly expect there to be more people who skip breakfast because they stayed up too late and tried to maximize every minute of sleep before class or work than there are people engaged in mindful intermittent fasting. That’s obviously not a healthful approach to life, but it’s not being accounted for at all. It’s more than just that though, of course. We know that the “fasting” subjects here are disproportionally likely to be poor, have poor access to food, and so forth.

Some people in this study probably are health-minded individuals engaged in intermittent fasting. But quite a number are poor people who stayed up late working their second job and didn’t eat breakfast in the morning. Or skipped breakfast those two days to save a couple bucks. Or are among the millions of highly food-insecure Americans who can’t find dinner tonight.

Maybe that’s being too dramatic, I really don’t know for sure. But I know with certainty that actual hands-on science tends to show the health benefits of intermittent fasting. And I know with certainty that these researchers made zero effort to find people engaged in intermittent fasting. And I know with certainty that the cohort they found to represent “intermittent fasting” is much more likely to be impoverished and hungry than the general population.

What I don’t know, as someone in a position fortunate enough to engage in deliberate healthful fasting, is how much poverty, food scarcity, and who knows what else affects cardiovascular disease and mortality. Id imagine that the daily stress experienced by the “intermittent fasting” group explains most or all of the measured increased in CVD mortality. In fact, it seems that low socioeconomic status itself is associated with CVD mortality to a far greater degree than this study claims intermittent fasting to be.¹¹ And basically to a greater degree than any dietary habit. Perhaps the AHA might focus on that relationship, if they aspire to a reduction in CVD deaths.

 

Conclusion

So that about sums it up. We know, from legitimate science, that intermittent fasting tends to impart positive health effects. We also know that these researchers took the about the shittiest approach they could to studying “intermittent fasting,” and I really have no problem accusing them of doing it deliberately. And I have no problem noting that this terrible research being spammed all over the internet, which has not yet even been fully published or peer-reviewed, just happens to once again align with the American Heart Association’s persistent opposition to healthful dietary habits. And while we’re at it, I have no problem suggesting that a lot of what I just wrote about should probably be absorbed as social/political commentary, even if it started as a defense of healthy eating habits.

 

But the bottom line is this – intermittent fasting is unequivocally not going to induce heart disease.

1.               The intermittent fasting trend may pose risks to your heart. Accessed March 20, 2024. https://www.msn.com/en-us/health/other/the-intermittent-fasting-trend-may-pose-risks-to-your-heart/ar-BB1k7bsm

2.               Gu L, Fu R, Hong J, Ni H, Yu K, Lou H. Effects of Intermittent Fasting in Human Compared to a Non-intervention Diet and Caloric Restriction: A Meta-Analysis of Randomized Controlled Trials. Front Nutr. 2022;9. doi:10.3389/fnut.2022.871682

3.               Ahmed N, Farooq J, Siddiqi HS, et al. Impact of Intermittent Fasting on Lipid Profile–A Quasi-Randomized Clinical Trial. Front Nutr. 2021;7. doi:10.3389/fnut.2020.596787

4.               Yuan X, Wang J, Yang S, et al. Effect of Intermittent Fasting Diet on Glucose and Lipid Metabolism and Insulin Resistance in Patients with Impaired Glucose and Lipid Metabolism: A Systematic Review and Meta-Analysis. International Journal of Endocrinology. 2022;2022:e6999907. doi:10.1155/2022/6999907

5.               Wang X, Yang Q, Liao Q, et al. Effects of intermittent fasting diets on plasma concentrations of inflammatory biomarkers: A systematic review and meta-analysis of randomized controlled trials. Nutrition. 2020;79-80:110974. doi:10.1016/j.nut.2020.110974

6.               Cho Y, Hong N, Kim K won, et al. The Effectiveness of Intermittent Fasting to Reduce Body Mass Index and Glucose Metabolism: A Systematic Review and Meta-Analysis. Journal of Clinical Medicine. 2019;8(10):1645. doi:10.3390/jcm8101645

7.               Chen M, Xu L, Horn LV, et al. Association of 8-Hour Time-Restricted Eating with All-Cause and Cause-Specific Mortality.

8.               Poverty in the United States: 2022. Accessed March 20, 2024. https://www.census.gov/library/publications/2023/demo/p60-280.html

9.               Nearly Half of Black and Hispanic People in the U.S. Face Food Insecurity, New Study Finds | Johns Hopkins | Bloomberg School of Public Health. Published February 26, 2021. Accessed March 20, 2024. https://publichealth.jhu.edu/2021/nearly-half-of-black-and-hispanic-people-in-the-us-face-food-insecurity-new-study-finds

10.             Garrett BE. Socioeconomic Differences in Cigarette Smoking Among Sociodemographic Groups. Prev Chronic Dis. 2019;16. doi:10.5888/pcd16.180553

11.             Zhang YB, Chen C, Pan XF, et al. Associations of healthy lifestyle and socioeconomic status with mortality and incident cardiovascular disease: two prospective cohort studies. BMJ. 2021;373:n604. doi:10.1136/bmj.n604

 

 

The Study That MIGHT Change Lipidology

Last weekend, UCLA cardiologist Dr. Matthew Budoff presented preliminary baseline data at the World Congress on Insulin Resistance, Diabetes, and Cardiovascular Disease that could theoretically, potentially, eventually, help turn the fields of lipidology and cardiology on their heads. 

The first of its kind prospective (ie. forward-looking) study directly assesses arterial plaque and calcification in a population of individuals with extremely high LDL-C resulting from a very low carbohydrate diet (Why this happens HERE). This population, while not entirely uncommon, is largely unstudied and may offer a number of clues to the true nature of lipid behavior and cardiovascular disease. 

Why is this? Because these individuals have very, very high LDL-C - an average of 272 mg/dl. Remember, conventional medical guidance considers an LDL-C above 100 mg/dl to be "high" while average is something around 120 mg/dl. 272 mg/dl is way past the 99th percentile and is more than 40% higher than the American Heart Association's threshold for automatic, no-questions-asked high-intensity statin therapy. Factor in the subject's average age of 56 years and they should, in theory, seem like a no-brainer for lipid-lowering therapy. 

These people should be at extreme risk for cardiovascular disease, should they not? Well, yes...if the traditional paradigm is correct. But the lipid-heart hypothesis is not nearly the settled science the AHA, pharmaceutical companies, and the medical establishment would have you believe. Instead, there are I believe two principle theories by which atherosclerotic cardiovascular disease may develop: 

•  Theory 1 is that LDL-C and LDL particles themselves are the direct, primary, and uniquely necessary elements of atherosclerotic disease. Elevated levels of LDL particles and LDL-C increase your risk for such disease by increasing the chance that, over time, more and more of these particles will become trapped in the wall of the blood vessel and ultimately lead to the development of plaque, arterial occlusion, etc. This is the common theory advanced by the medical industry. 

• Theory 2 is that LDL particles damaged as a result of oxidative stress or hyperglycemia are the target of immune cells whose function is to trap the damaged particles for recycling and removal. When poor metabolic efficiency, hyperglycemia, and increased oxidative stress lead to an excessive number of damaged LDL particles, this process occurs at a greatly increased rate. The development of plaque, arterial occlusion, etc. is fundamentally an inability of the body to remove and recycle these damaged particles as quickly as they accumulate. I've written many thousands of words about why I favor this theory. 

Now the subjects in this study are excellent candidates to test these theories because they not only have very high LDL-C, but also good metabolic health (HDL = 90, Triglycerides = 64, BP = 117/76, A1C = 5.4%). This means the key sentence in Theory 2 ("poor metabolic efficiency, hyperglycemia, and increased oxidative stress...") is of limited relevance and allows a unique opportunity to compare the two theories. If Theory 1 is correct, these individuals should rapidly develop cardiovascular disease. If Theory 2 is correct, they should not.

So far we only have preliminary, baseline date, which you can watch be presented HERE, but which I'll summarize briefly:

CAC and plaque scores for low-carb and control groups

• The researchers assessed two direct measures of atherosclerotic disease - coronary artery calcification and a "total plaque score" (calculated via CT angiography). These were measured in both the low-carbohydrate group and a control group matched for age, metabolic health, etc. The only major difference between the two groups was LDL-C, which was more than twice as high in the low-carb group. 

• The researchers will assess both of these metrics again in another year to assess any disease progression in each group

• The majority of subjects in each group had coronary artery calcification scores (CACs) of 0, indicating no calcification (CACs of significantly at-risk individuals commonly range from 100-400 or higher).

• Total Plaque Score is calculated from 0 to 45 (0-3 at each of 15 different potential plaque sites). In the low-carb group, the median subject had a TPS of 0 with an interquartile range (the middle 50%) of 0-2. No subject had a TPS greater than 12. The control group had a median TPS of 1 with an interquartile range of 0-4. The difference between the two groups was not statistically significant. 

• There was no difference in CAC or TPS between the low-carb, high LDL-C group and the control group.

• There was no relationship in either group between LDL-C levels and measured plaque.

Now, it should be noted that CACs and TPSs of 0 are not particularly remarkable. In fact a little more than half of 56 year olds have a CAC of 0 (you can play around on this NLHBI site if you'd like to see). Its normal and healthy to be free of coronary calcification. But what makes this study so potentially fascinating is that these subjects aren't considered "healthy" in a traditional sense. Years of extremely high LDL-C should predict the impending doom of atherosclerosis, but so far at least this group shows little and often no disease progression whatsoever. In fact, their cardiovascular disease state is indistinguishable from matched subjects with good metabolic health and no elevated LDL-C. 

We'll have to wait until 2024 for follow-up results, in which actual plaque progression in each group will be compared. But if two groups with similar metabolic health yet vastly different LDL-C levels continue to demonstrate identical disease states (or the lack thereof), it would lend significant credence to the notion that metabolic health and other associated factors, rather than LDL Cholesterol, drive chronic cardiovascular disease. Ideally, this would help launch a number of other studies to further challenge common assumptions and explore underappreciated aspects of lipidology and cardiovascular disease. Will that be the case? TBD...

The Problematic Paradigm of LDL-C, Part 1

The Development of the Lipid-Heart and Diet-Heart Hypotheses

The lipid-heart and diet-heart hypotheses. The average person may not know them by their name, but certainly everybody knows them in some capacity or another. Quite simply, their combined assertion is this: consuming dietary fat and cholesterol causes cholesterol levels in the blood to rise, and these rising levels of cholesterol are the primary factor in the development of heart disease. Put even more simply – eat too much fat, risk a heart attack.

This singular basic message has formed the backbone of nutrition and dietary advice for decades, for as long as most who will read this have been alive. One would be forgiven if, somewhere along a road lined with high-school health classes, heart-healthy checkmarks, and ubiquitous advertisements for low-fat yogurt and statins, that these ideas were settled science existing in a space far beyond any credible doubt. But what makes the longevity of these twin concepts so notable is both the absence of strong evidence to support the assertions, and the incredible simplicity of the model that allows them to persist unabated nonetheless.

The simplicity is apparent and approachable – eating fat and cholesterol causes cholesterol to accumulate in the bloodstream, where it…. gets stuck, basically. The mechanics of WHY cholesterol behaves this way are less apparent and rarely explained, but it seems the basic idea presented to the public is simply that cholesterol is a sticky substance that, in large amounts, inevitably attaches itself to the walls of the arteries and blocks the normal flow of blood. Obviously, things are not this simple. Likely surprising to many, cholesterol does not exist “in” the bloodstream at all, but in carrier “boats” called lipoproteins, and so the basic construct by which it sticks to the blood vessels needs at least some greater level of detail. In fact, it requires a much greater level of detail, which will be explored over the course of this series of posts.

But before exploring the realities of cholesterol and the failures of the lipid-heart and diet-heart hypotheses, one needs to understand where these parallel concepts came from and the factors that allowed them to thrive.

Some 70 years ago, as it is today, heart disease was a common killer in western society. Autopsies of young soldiers killed in action found significant atherosclerosis, artery-clogging deposits that contained, among other things, a waxy substance known as cholesterol. In an era that still saw doctors recommending their preferred brand of cigarette, the heavy rates of smoking among these young soldiers and the population at large didn’t jump off the page as an obvious candidate in the search for the heart-stopping culprit.

Around the same time, a University of Minnesota physiologist named Ancel Keys presented to the World Health Organization his theory that the consumption of fat (saturated fat from animal products in particular) was the primary factor in the development of cardiovascular disease. To support this claim, Keys hand-selected six countries, from a complete dataset of about two dozen, that allowed him to draw the most pronounced connection between increasing fat consumption and increasing heart disease. This extremely biased methodological error, in which Keys dismissed all available data contradicting his beliefs, led him to be publicly refuted in his original attempt at promoting his idea.¹

2

However, other factors were also developing that would eventually assist Keys in his effort to demonize the consumption of saturated fat. In 1911, while attempting to use vegetable oil as a base for soapmaking, the company Proctor and Gamble inadvertently discovered that they could create a partially solid product that reasonably resembled butter at room temperature. Vegetable oils, previously used primarily as industrial lubricants, were suddenly an appealing option for the creation of a cheap, shelf-stable, butter-like compound that could be marketed as a cooking product. Thus was born Crisco, an industrially-produced product with little saturated fat.³

For the next couple of decades, sales grew steadily as P&G marketed their butter-replacement product as a cheaper, easier cooking alternative. However, their big moment came in 1948 with their large donation to a fledgling non-profit known as the American Heart Association. On the back of this cash influx, the AHA, presided over at this time by Dr. Paul Dudley White, would go from an organization with little influence or notoriety to the largest non-profit in the nation in just more than a decade.^4,5

In 1955, during the AHA’s rise to prominence, President Dwight D. Eisenhower, himself a heavy smoker, suffered a heart attack that would be instrumental in shaping the American dietary paradigm. His personal doctor, the same Dr. White who presided over the AHA during the Crisco partnership and who was himself unsurprisingly opposed to saturated fat consumption, entrusted the role of the President’s personal nutrition advisor to the like-minded Ancel Keys. Eisenhower bought in to the “prudent diet” approach being offered by Keys, whose relationship with the popular president helped push him and his ideas forward in the American consciousness.^6,7

Now in lockstep with the President, Keys and the ever-growing AHA led a national charge to erase the menace of heart disease from American society. Within years of President Eisenhower’s heart attack, Keys was featured on the cover of TIME magazine as an expert on diet and heart health, while the AHA formalized guidance that saturated fat consumption should be strictly limited for the prevention of heart disease. These guidelines, based on effectively no clinical evidence, did not yet push for the elimination of all fat from the diet, arguing that the replacement of saturated fat with polyunsaturated fats such as Crisco would do the trick. As we will see in the next section, clinical trials launched after the fact were meant to validate the guidelines already put in place, but failed notably in this endeavor.

Unfortunately for President Eisenhower, the prudent diet approach assigned by Keys was ineffective, as he would go on to suffer multiple additional heart attacks, a stroke, and develop diabetes before his death from obstructive coronary disease. Now, much of this declining health is again likely attributable to smoking, but it is without question that the public face of America’s new low-fat diet experiment fell well short of resounding success.

Meanwhile, scientists such as John Yudkin advocated not for a low-fat, but instead a low-sugar approach to dietary intervention. In the mid-1960s, a series of studies pointing towards sugar as the primary driver of cardiovascular disease helped, briefly at least, keep alive the fat vs sugar debate. In an effort to combat the impact of these studies the research director of the Sugar Research Foundation, John Hickson, paid two Harvard scientists to author editorials in the New England Journal of Medicine that would discredit the research and exonerate sugar as a disease-causing agent. One of these scientists would become an administrator for the USDA, while the other would lead the Kellogg’s-funded nutrition foundation at Harvard.^8,9

Efforts such as these, combined with the growing prominence of Keys and his like-minded followers, the influence of the AHA, and the full resolve of the US government, helped cement the lipid-heart and diet-heart hypotheses as the standard dietary paradigm. While evidence was still extremely limited, one major observational study had now confirmed a general association between cholesterol and heart disease (the Framingham study, which we’ll examine in detail in the next section), with the promise of more and better research to come being promoted by Keys and others. While some of these research efforts would ultimately do more to discredit than to confirm the attack on saturated fat and LDL cholesterol levels, they would not do so until such notions were firmly entrenched in the minds of scientists, health care professionals, and the general public.

The lipid-heart and diet-heart hypotheses, weak and unsupported as they may be, were here to stay.

Part 2 – LDL Studies and the Association Between LDL-C and Heart Disease, pt.1

 

1.               1957_Yerushalmy_Hilleboe_Fat_Diet_Mortality_Heart_Disease.pdf. Accessed December 12, 2022. http://library.crossfit.com/free/pdf/1957_Yerushalmy_Hilleboe_Fat_Diet_Mortality_Heart_Disease.pdf
2.               Deep Nutrition: Why Your Genes Need Traditional Food - Catherine Shanahan, M.D. - Google Books. Accessed December 12, 2022. https://books.google.com/books?hl=en&lr=&id=1fs3DAAAQBAJ&oi=fnd&pg=PP1&dq=deep+nutrition+cate+shanahan&ots=YNQWfUTs9a&sig=tVMLoLV8CQ34Lhsq4Rduo__tQdg#v=onepage&q=deep%20nutrition%20cate%20shanahan&f=false
3.               MSEd LWB. This Is How Crisco Is Really Made. Mashed. Published July 9, 2020. Accessed December 12, 2022. https://www.mashed.com/224919/this-is-how-crisco-is-really-made/
4.               History-of-the-American-Heart-Association.pdf. Accessed December 12, 2022. https://www.heart.org/-/media/Files/About-Us/History/History-of-the-American-Heart-Association.pdf
5.               Teicholz N. The Big Fat Surprise: Why Butter, Meat and Cheese Belong in a Healthy Diet. Simon and Schuster; 2014.
6.               Dwight Eisenhower: Treating his Heart Attack. Accessed December 12, 2022. https://doctorzebra.com/prez/z_x34mirx_g.htm
7.               Paul Dudley White. In: Wikipedia. ; 2022. Accessed December 12, 2022. https://en.wikipedia.org/w/index.php?title=Paul_Dudley_White&oldid=1125479668
8.               Taubes G. The Case against Sugar. First edition. Alfred A. Knopf; 2016.
9.               Fredrick J. Stare. In: Wikipedia. ; 2022. Accessed December 12, 2022. https://en.wikipedia.org/w/index.php?title=Fredrick_J._Stare&oldid=1079782061