Breaking Apart the Component Model

In trying to eat a healthy diet, many focus not on foods, but on food components–avoid carbohydrates, eat more fiber, reduce saturated fats, get lots of antioxidants. An engineer friend built a diet by designing a basic meal plan that defined the things he wanted and didn’t want, and constructed a (highly artificial) block of food that had the recommended components to cut into portions for breakfast, lunch, and dinner for optimal health. While his anecdotal story doesn’t provide enough data to draw any conclusions (he survived the diet, despite the taste), the component model has been repeatedly tested by scientists and companies trying to deliver just a component (fiber drinks, antioxidant pills) without finding a benefit.

Let’s look at the example of pecan nuts. 300 g (11 oz.) of pecans has 28 g of fiber, which is 100% of the USDA recommendation for fiber (good!), but also 100% of the maximum USDA recommendation of 21 g of saturated fat (bad!). So what would happen if you ate this many pecans? From a component model of food, you would think that it would depend which effect is stronger, the benefit of the fiber or the negative effects of saturated fats, and the more you ate, the stronger the effect.

But, that’s not how the effects of pecans reveal themself. A meta-study of dietary habits and lifespan encompassing 16 studies and over 80,000 deaths showed that a handful of up to 20 g (⅔ oz.) of nuts per day is good for you, reducing all-cause mortality by nearly 20%1. After that initial amount of nuts in a day, it appears there is no further benefit or harm from eating more. It’s basically empty calories after that point that neither harm nor hurt you.

Many books talk about “superfoods” and the benefit of this this or that food over other foods of the same kind or avoiding carbs or eating fiber, but longevity research shows food categories (fruits, vegetables, processed meat, etc.) captures most of the good or bad impact, and even for foods that increase health and longevity, there is a threshold, like 20g of nuts, past which there is little benefit from eating more of that food, even of a different type.

Through this blog, I’m going to work through the diet, exercise, and habits for living a long, healthy life. Using the latest research and using all-cause mortality and a form of “ground truth” to choose which changes have the most benefit, I’ll cover areas where the common wisdom is right, areas where it’s right, but you don’t need as much as you might think, and areas where the common wisdom is just wrong.

Please subscribe for more posts!


  1. Food groups and risk of all-cause mortality: a systematic review and meta-analysis of prospective studies