Dr. Fuhrman Changed My Life. I'm Not Buying His Supplement.

Last night I ate a 200-gram jar of lentils. It was not a heroic act of discipline. It was dinner, and it delivered somewhere north of a quarter of my day's fiber at roughly 230 calories.

Twenty years ago I would not have understood that sentence. The person who taught me to read it was a family physician from New Jersey with a book called Eat to Live.

The equation that did the damage

Fuhrman's book, published in 2003, rests on a single line: H = N/C. Health equals nutrients divided by calories.

It is an almost insultingly simple idea, which is why it works. It turns a plate into a ratio. A handful of kale and a handful of crackers stop being "food" and become two very different fractions. Once you see the denominator, you can't unsee it.

He built the rest of the program on that ratio. G-BOMBS — greens, beans, onions, mushrooms, berries, seeds — as the daily core. The salad is the main dish. Beans every day. Nuts and seeds, not bottled oil, as the fat source.

I followed it, more or less, and the results were the kind that make you evangelical at dinner parties. That's how an on-ramp works. It doesn't have to be the whole highway. It just has to get you moving in the right direction.

What I kept, and what I measured

The trouble with "nutrients" as a numerator is that nobody can count them at the grocery store. Fuhrman's answer was the ANDI score, a proprietary index. Useful, but it asks you to trust the scorekeeper.

Fibertarian keeps his ratio and swaps the numerator for something printed on every label: grams of fiber per 100 calories.

Fiber is an imperfect proxy for nutrient density, but a surprisingly faithful one. It travels with the things you want — intact plant cells, slower absorption, the gut-hormone signals that tell you to stop eating — and it is almost perfectly absent from the things you don't. Lentils run about 6.8 grams per 100 calories. Broccoli, around 9. White bread, about 1. Cheese, oil, and sugar, zero.

That gives you a target you can actually do arithmetic with. Fibertarian aims for 50 grams a day, which at a typical calorie intake works out to about 2.5 grams per 100 calories — nearly double the density behind official guidelines.

The number is a fence, not a goal. At 50 grams, every fiber-free calorie has to be paid for by something else on the plate. You can't reach it on white bread and oil. The refined stuff gets crowded out without anyone having to ban it.

From there, the vocabulary sorts itself:

• Wrapped — foods that arrive with their own fiber: legumes, vegetables, fruit, whole grains, nuts, seeds, olives.
• Unwrapped — whole foods that simply have none of their own. They're fine when they share a plate with something wrapped.
• Stripped — foods that had their fiber removed: white flour, sugar, juice, and yes, olive oil. If you want olive fat, eat an olive.

And from Fuhrman's "beans every day," I went one further: legumes at every meal. That's where the jar of lentils came from.

Where the road forked

In recent years Fuhrman has argued, with increasing urgency, that vegans who don't take long-chain omega-3s — DHA and EPA — are risking dementia. The recommended remedy is an algae-based DHA/EPA supplement. His company sells one.

I don't doubt his sincerity. I do notice the shape of the argument, because it's the same shape I object to elsewhere.

Fibertarian has a running complaint: there is no Big Broccoli. Nobody earns billions from lentils, so nobody funds 2,000-person trials of whole-food diets, while drug companies fund exactly those trials for their products. If I'm going to hold that asymmetry against pharmaceutical companies, I have to hold it against a physician whose clinical conclusion ends in a checkout page.

So I applied the same arithmetic.

The mechanism is plausible. Humans convert plant omega-3 (ALA, from flax, chia, walnuts) into DHA poorly. Fish don't make DHA either — they get it from algae, which is a nice point in favor of skipping the fish and going straight to the source.

The observational hints are real but modest. In a study of 1,111 older women, those with higher blood levels of EPA and DHA had slightly larger brains — about 0.7 percent overall, and 2.7 percent in the hippocampus, the brain's memory center. That's an association, not a verdict.

The trials are not kind. The VITAL trial gave 4,218 healthy older adults a daily dose of marine omega-3s for two to three years. Cognitive decline was no different from placebo. Trials in people who already have Alzheimer's have mostly come up empty as well.

And the population data point the other way. In a Taiwanese cohort of more than 12,000 people, vegetarians had 38 percent lower dementia risk than non-vegetarians, after adjusting for age, heart disease, diabetes, and blood pressure. A 2026 study of older Chinese adults found each step up a fiber-based vegetarian diet score tracked with 17 percent lower risk.

The dementia problem, in other words, isn't showing up where the supplement theory says it should.

Keeping the good parts

None of this makes Eat to Live wrong. H = N/C is still right. The salad is still the main dish. Beans are still the backbone.

It just means that a person can be right about the big picture and less right about the product. That happens to doctors, to scientists, and — I'm confident — to people who run fiber websites.

For the record, Fibertarian's position on omega-3s is modest. Eat ground flax, chia, or walnuts daily; they're wrapped, and they cover the ALA requirement. Grind the flax, or it passes through you with most of its oil intact. Algae DHA is inexpensive insurance if you want it. If you're worried, a finger-prick omega-3 test will tell you more than any marketing copy. What nobody should claim — including me — is that a capsule prevents dementia.

Every on-ramp is built by someone with a destination in mind. The courtesy we owe the builder is to credit the road. The courtesy we owe ourselves is to keep reading the signs after we've merged.

Which of your own on-ramps have you checked lately?