One thing I have been pointing out to people for years now is that claims like "saturated fat is bad" or "cholesterol causes heart attacks" do not make sense on many levels.  Obviously, some people find it difficult to understand the molecular-level evidence.  One thing that is clear to everyone I've talked to is that when you look at common recipes, you find that cholesterol and saturated fatty acid contents are similar.  In fact, "vegetarian" and "Mediterranean diet" dishes are clearly higher in saturated fatty acid (SFAs) content than beef and pork dishes, if the sample of recipes I clipped out of my local newspaper over a period of several months is any indication.  The vegetarian dishes averaged 4.875 grams of SFAs per serving, while the beef and pork ones averaged 3.222.  Below are some examples taken from this local, but major newspaper, New York's "Newsday:"

Roast Pork With Grapes: 10 grams of fat, 3 of it saturated (SFAs); 75 mg of cholesterol (1/2/2004, page B27).

Pasta With Chicken And Broccoli: 9 g fat,  3 SFAs;  76 mg cholesterol (11/5/2003, page B31).

Beef Picadillo With Toasted Corn Tortillas: 9 g fat, 2 SFAs; 60 mg cholesterol (8/6/2003).

Chicken Pizza With Vegetables: 9 g fat, 4 SFAs; 50 mg cholesterol (12/5/2003, page B39).

Pasta With Tomatoes, Chickpeas, and Feta (vegetarian): 23 g fat, 7 SFAs; 33 mg cholesterol (12/17/2003, page B23).

Greek Tomato And Green Bean Soup (vegetarian): 9 g fat, 3 SFAs; 17 mg cholesterol (10/14/2003, page B17).

Curried Beef Kabobs: 7 g fat, 2 SFAs; 71 mg cholesterol.

Bean and Corn Burritos With Salsa Fresca (vegetarian): 12 g fat, 3 SFAs; 9 mg cholesterol (8/4/2003). 

Individual Greek Pitzas (vegetarian): 16 g fat, 6 SFAs; 33 mg cholesterol (8/13/2003).

Meatless Pad Thai (vegetarian): 23 g fat, 4 SFAs; 106 mg cholesterol (2/12/2004).

Chinese Style Beef In Barbeque Sauce: 10 g fat, 3 SFAs; 78 mg cholesterol (1/26/2004, B16).

Mediterranean Tortellini (vegetarian): 15 g fat, 5 SFAs; 28 mg cholesterol (1/30/2004, page B31). 

Pork Medallions With Pears In Port Wine Sauce: 8 g fat, 2 SFAs; 74 mg cholesterol (2/3/2004, page B19).

Pork Tenderloin With Orange Balsamic Glaze: 4 g fat, 1 SFAs; 74 mg cholesterol (10/6/2003).

Tex-Mex Sausage And Rice: 13 g fat, 5 SFAs; 95 mg cholesterol (1/8/2004, page B19).

Pink Pasta Primavera (vegetarian): 15 g fat, 5 SFAs; 15 mg cholesterol (9/19/2003, page B39).

Chicken Po' Boys: 6 g fat, 1 SFAs; 60 mg cholesterol (7/29/2003, page B17).

Inside Out Cheeseburger: 17 g fat, 7 SFAs; 86 mg cholesterol (7/22/2003, page B19).

Shrimp And Sausage Jambalaya: 6 g fat, 1 SFAs; 106 mg cholesterol (9/26/2003, page B35).

Buffalo Turkey "Wing" Sandwiches: 16 g fat, 3 SFAs; 74 mg cholesterol (10/3/2003, page B37).

Lemon Garbanzo Beans Over Polenta (vegetarian): 16 g fat, 6 SFAs; 33 mg cholesterol (10/31/2003 page B35).

Chilean Beef And Peppers: 13 g fat, 4 SFAs; 59 mg cholesterol (10/28/2003, page B19).

 

Most of these dishes are in the 300-500 calorie range, per serving.  Of course, what matters is how satisfying the dish is, because if it is not, one will likely eat more than one serving, or eat a big dessert.  The only recipes that are significantly different from the ones above are the ones that use a lot of coconut oil, chocolate, or butter.  For example. a chocolate mousse recipe I found on the internet was as follows:

Per serving: 122 calories, 3 g, 17 g carbohydrate, 3.5 g fat (3 g saturated fat), 3 mg cholesterol, <1 g fiber, 351 mg sodium. Calories from fat: 26%.

Source: http://onhealth.webmd.com/script/main/art.asp?articlekey=59751

This recipe, if tripled to the amount of calories found in many "main dishes," would be about 10.5 g fat, 9 SFAs, and 9 mg cholesterol.  If we quadruple it, to 488 calories, which is also similar to many main dishes, the numbers are: 14 g fat, 12 SFAa, 12 mg cholesterol.  I use this example because it is similar to a typical meal of mine, and also because it shows what kinds of food you would have to consume large amounts of in order to be within a range many US "nutritional experts" claim is unhealthy.  It is common for me to eat a lot of whole milk cheese and yogurt, dark chocolate, and some butter or coconut (either in the oil or shredded form) with each meal.  This is not a diet that has been found to be unhealthy, actually.  The evidence suggests it is healthy.  The only stuides that appear not to involve those that do not discriminate among the sources of saturated fatty acids and cholesterol, do not take cooking techniques and "freshness" into account, do not take iron content into account, do not take all possibly relevant "cultural" or socio-economic factors into account, etc., but instead estimate the amount of SFAs consumed and then try to see if there is a "statistically significant" rise in the incidence of one or more "diseases" among those who eat more SFAs overall.  This makes little sense, since one can now easily devise several different diets that are tasty (so that many people will continue to eat this way if they know it is healthy) and consistent with the molecular-level evidence.  These diets can then be fed to species like dogs, which live significantly shorter lives than humans, but are fairly good animal models for overall health, and then we would have very strong evidence indicating which diet is healthiest.  Such experiments were done many decades ago, but are now out of favor, since it is much easier to do a short term study to determine if a particular "marker" (such as LDL) rises or falls due to a dietary change, even though it is known that LDL must be oxidized in order to cause "disease," and so is only a potentially useful marker if people are eating diets rich in oxidized cholesterol (or diets that oxidize cholesterol inside the body).

Below is the most typical kind of study that is conducted these days in nations like the USA concerning possible dietary connections to “disease.�?I ask the reader to think of the information above while reading though this abstract (summary), and then I will comment on it below it.

Arch Neurol 2003 Feb;60(2):194-200.

Dietary fats and the risk of incident Alzheimer disease.

Morris MC, Evans DA, Bienias JL, Tangney CC, Bennett DA, Aggarwal N, Schneider J, Wilson RS.

Rush Institute for Healthy Aging, 1645 W Jackson, Suite 675, Chicago, IL 60612, USA. [email protected]

BACKGROUND: Few studies have investigated the effects of dietary fats on the development of Alzheimer disease. We examined the associations between intake of specific types of fat and incident Alzheimer disease in a biracial community study. METHODS: We performed clinical evaluations on a stratified random sample of 815 community residents aged 65 years and older who were unaffected by Alzheimer disease at baseline and who completed a food-frequency questionnaire a mean of 2.3 years before clinical evaluation. RESULTS: After a mean follow-up of 3.9 years, 131 persons developed Alzheimer disease. Intakes of saturated fat and trans-unsaturated fat were positively associated with risk of Alzheimer disease, whereas intakes of omega-6 polyunsaturated fat and monounsaturated fat were inversely associated. Persons in the upper fifth of saturated-fat intake had 2.2 times the risk of incident Alzheimer disease compared with persons in the lowest fifth in a multivariable model adjusted for age, sex, race, education, and apolipoprotein E epsilon4 allele status (95% confidence interval, 1.1-4.7). Risk also increased with consumption of trans-unsaturated fats, beginning with the second fifth of intake (relative risk, 2.4 compared with the lowest fifth; 95% confidence interval, 1.1-5.3). We observed linear inverse associations between Alzheimer disease and vegetable fat (P =.002), and, after further adjustment for other types of fat, marginally significant associations with intake of omega-6 polyunsaturated fat (P =.10 for trend) and monounsaturated fat (P =.10 for trend). Intakes of total fat, animal fat, and dietary cholesterol were not associated with Alzheimer disease. CONCLUSION: High intake of unsaturated, unhydrogenated fats may be protective against Alzheimer disease, whereas intake of saturated or trans-unsaturated (hydrogenated) fats may increase risk.

The first thing I will say is that they are incorrect about what they are studying!  They say "dietary fats," but then talk about the actual molecules, which are fatty acids in this case.  A "dietary fat" would be something like butter, coconut oil, or corn oil.  Thus, if they had "investigated the effects of dietary fats," they would have examined the difference between people who used butter versus those who used margarine as their primary sources of fatty acids, for example.  This would be much better, though cooking techniques would certainly be a relevant factor that would need to be controlled for, unless hardly anyone cooked their primary dietary fat.  Another point that is worth noting is that they assumed that the questionnaires were accurate, even though 131 of the 815 people studied here developed Alzheimer’s Disease (AD) within four years. Since this is a disease that involves mental impairment, especially memory loss, one would think that the authors would want to address this point. They also do not mention that AD is an inflammatory disorder, and that SFAs cannot be made into inflammatory molecules. In fact, if anything, SFAs act as a buffer against inflammation, due to their biochemical inertness. Among 131 cases of AD, to say that those who consumed the most SFAs and TFAs were 2.2 times more likely to develop AD really only means that a small number of people developed it above the other fifths.

In any case, I don’t necessarily think that their findings are incorrect, but the notion that there is likely causation here is a clear case of “jumping the gun.�?We do not know if those who consumed the most SFAs and TFAs also consumed the most overall calories per day, or the most iron per day, or any number of other items that could act to “fuel�?inflammation, unlike SFAs, which do not. The most glaring mistake made here, from what I can tell, is that they did not control for SFAs and TFAs separately. That is, there was no group of people who consumed a large amount of SFAs, but not TFAs, and vice versa, as far as we know. The scientific method demands that this be done. If it is not done, the result can be a highly misleading study such as this one seems to be. After all, if a high relative consumption of SFAs meant a much greater risk of AD (and if the risk was a tiny one, who would care at all?), then Asians on diet rich in coconut products should have very high rates of AD, when in fact the opposite is the case. It certainly may be the case that foods rich in TFAs also tend to be rich in SFAs, and so they both look like they may be “guilty.�?In light of what is known now, the researchers also should have mentioned that the double bonds of the TFAs may be to blame, in terms of dangerous lipid peroxidation reactions, and so this should have been controlled for as well. Those using high quality oils such as sunflower would at least be provided with some natural antioxidant protection, whereas TFAs may be provided with synthetic antioxidants that are only used to prevent rancidity before the expiration date is reached.  Moreover, someone using lard would also not get any of the antioxidant protection found in something like a good quality olive oil or sesame oil.

However, considering the fat content of the dishes listed above, how would one go about trying to follow the researchers�?tentative conclusions about the meaning of their findings? Many of the vegetarian dishes, including “Mediterranean�?ones, have more SFAs than beef and pork dishes, for example (per serving). Thus, assuming that people avoid TFAs completely, these researchers would have us believe that at least some beef and pork dishes are among the healthiest things one could eat! Notice that they also don’t take cooking techniques into account in their study, which is one of the main reasons a diet rich in high quality beef, for example, may not be healthy. And why do they not address the evidence that suggests that only meat is unhealthy in “diseases�?like AD? It is as if they are totally unaware of the relevant literature. My examination of the evidence leads me to agree with researchers who claim that meat in general is dangerous, at least the way it is usually cooked in nations like the USA. On the other hand, uncooked dairy and coconut products are among the healthiest things one can eat these days. The SFA content is irrelevant. Only in a highly saturated food, such as coconut, will the SFA content act to prevent rancidity, which is the main problem. And other than coconut, there is really nothing else that is close (except for the very similar palm kernel oil, which is rarely used in the kitchens of Western peoples), in terms of the SFA content. This means that except for coconut, one should be very careful about eating anything that contains more than trace amounts of fatty acids (or cholesterol). Foods that are fairly rich in SFAs, such as whole milk yogurt, are generally good, unless one cooks it at high temperature and while exposed to air, or if you can taste that it is “going bad.�?