 How else can we make decisions for ourselves and our families, but by the best available balance of evidence? The latest meta-analysis, pooling data from more than a dozen studies involving more than 300,000 people, indicates that there's a dose-dependent association between egg consumption and the risk of cardiovascular disease and diabetes. But that doesn't mean every single individual study showed evidence of harm. Even though the totality of evidence points to harm, the egg industry can certainly cherry-pick out studies that show no apparent effects. If eggs are harmful, though, why don't all studies on heart disease and egg consumption show significant harm? It may have to do with Rose's concept of a sick population. If an entire population is sick, then the range of quote-unquote health may not be sufficiently broad to establish a significant association. Let me explain. This is one of the most famous papers ever written in preventive medicine. Should be required reading for all medical students. Imagine if everyone smoked 20 cigarettes a day. If everybody smoked, then clinical studies, gauge control studies, cohort studies would all lead us to conclude that lung cancer was a genetic disease. In one sense, that would be true if some smokers get cancer, others smoke their whole lives, and don't. But if everybody smoked, we'd never know that smoking was a risk factor. Thankfully, in the case of cigarettes and lung cancer, it so happened that the original study populations contained about equal numbers of smokers and non-smokers, and in that situation, studies are able to identify the main risk factor. But take cholesterol. Here's the cholesterol levels of the people without heart disease in the famous Framingham Heart Study. And here's the cholesterol levels of those that did develop heart disease. It's hardly any different, because practically everybody's cholesterol was too high. It's like everyone was a smoker. The painful truth is that even someone at quote-unquote low risk for heart disease is likely to die of heart disease. Everyone who eats the standard Western diet, in fact, is a high risk individual when it comes to heart disease. In a sick population like ours, where nearly everyone is eating lots of saturated fat and cholesterol, adding some more saturated fat and cholesterol in the form of eggs may just take us from one sorry state, probably dying from heart disease, to another sorry state, still probably dying from heart disease. So in the federal guidelines, say we need to really restrict dietary cholesterol, especially if we're at high risk for heart disease, we need to realize that nearly all Americans that live past middle age are at high risk of dying from heart disease. It's our number one cause of death. A 20-year-old man, it might feel it's safe to smoke and eat egg yolks, because his heart attack is 45 or so years in the future, but why would he want to accelerate the progression of the atherosclerotic plaque to bring it on sooner? Stopping egg yolks after a heart attack would be like quitting smoking after lung cancer is diagnosed. There may in fact be a plateau of risk for smoking, too. Whether we smoke for 25 years or 35 years, our risk of lung cancer may be about the same, really high, but about the same. So the tobacco industry could truthfully tell someone who smoked for most of their lives for the last 25 years, keep smoking. Don't worry, you can keep smoking and your risk of lung cancer won't go up. Conveniently, just failing to mention that you're already really at high risk, and if you quit completely, your risk would drop dramatically. And the same way the egg industry can design a study showing that adding eggs doesn't make much of a difference, but make the same convenient omission. Imagine if you took a raging drunk and had them take a shot of whiskey. And someone who's hammered it might not make much of a difference, but to a tea-toler, a couple shots could have quite an effect. So it's like the alcohol industry with a group of drunks saying, see, a couple shots, no big deal. But that doesn't mean it's not better to be sober. Instead of going from high risk to high risk, better to go to low risk or no risk.