An important part of the conventional wisdom on diet and health is that saturated fats are bad. The basic idea is that saturated fats raise blood cholesterol and thereby clog arteries, which can lead to heart attacks and strokes, among other bad events.

The article shown above, “Reduction in saturated fat intake for cardiovascular disease,” by Lee Hooper, Nicole martin, Asmaa Abdelhamid and George Davey Smith, is a meta-analysis of 15 randomized clinical trials that meet criteria for being fairly solid evidence. They show convincingly that whatever it is that happens when people are advised to reduce saturated fat intake by a dietition, nutritionist or nurse is helpful in reducing bad cardiovascular events—perhaps by around 10% to 15% if one had to guess (though substantially smaller or bigger effects are within the confidence intervals).

The idea in most of these randomized trials was to get people to substitute polyunsatured or monounsatured fats for saturated fats that were cut out. Sometimes the experimental subjects were given polyunsaturated or monounsatured fats to eat. The big blind spot in the studies underlying this meta-analysis is that they don’t seem to distinguish between cutting out animal protein and cutting out saturated fat. To see if it is cutting out animal protein that is beneficial rather than saturated fat per se, it would be helpful to have an experiment in which experimental subjects were encouraged to cut out meat and dairy but increase plant-based saturated fat—by consuming more coconut milk, for example.

Alternatively, trials that compare drinking skim milk to whole milk or compare eating lean meat to eating full-fat meat from the same animal would hold the amount and type of animal protein reasonably constant while reducing saturated fat. (One might also want to increase the amount of polyunsaturated fat or monounsatured fat to keep total amount of fat constant.) Googling “randomized trial skim milk whole milk” I didn’t find much other than this very small study of 18 people which claims that the previous evidence was unclear about skim milk vs. whole milk and found little effect itself:

Why is it important to control for animal protein? First of all, a large share of the milk people drink contains A1 casein. See “Exorcising the Devil in the Milk.” If nutritional advice to avoid saturated fats causes some people to avoid milk rather than switch to skim milk (perhaps because they hate the taste of skim milk), it would protect them from A1 milk protein. (Another way to avoid A1 casein is to buy A2 Milk.) Second, at least some types of animal protein stimulates insulin production in the body. See “Forget Calorie Counting; It's the Insulin Index, Stupid” and “Why a Low-Insulin-Index Diet Isn't Exactly a 'Lowcarb' Diet.”

The idea that animal products are bad because of saturated fat has often caused evidence that consuming too much in the way of animal products has health risks to be interpreted as evidence that saturated fat is bad. This idea is encouraged by the fact that saturated fat does seem to make blood cholesterol look worse by the standards of the conventional wisdom about blood cholesterol. But the conventional wisdom is based on blood lipid measurements on only three dimensions of what is probably more like a ten-dimensional vector of important subspecies of cholesterol bearing objects. See “What is the Evidence on Dietary Fat?” (On the subject here, another relevant post is “How Unhealthy are Red and Processed Meat?”)

The bottom line is that a lot of research remains to be done. Way too little research has been done to answer the questions that a skeptic of the dangers of saturated fat would pose. Saturated fat may be terrible. But despite the conventional wisdom, we really don’t know for sure yet.

It matters to get things right. My view is that the health would improve markedly if it became conventional wisdom that dietary sugar is much, much more harmful than dietary fat (even with no particular mention of the different kinds of dietary fat other than transfats being bad). As it is, an anti-dietary-fat message is crowding out the clear anti-sugar message that I think is much more on target and much more important. (The study discussed in “Why a Low-Insulin-Index Diet Isn't Exactly a 'Lowcarb' Diet” for example, should make one quite worried about sugar consumption and processed food consumption—which currently are reasonably close to being the same thing, since almost all highly processed food contains a fair amount of sugar.)

For annotated links to other posts on diet and health, see:

• Miles Kimball on Diet and Health: A Reader's Guide