A single square of dark chocolate on a white saucer beside a stack of scientific journals and a cup of black coffee on a plain desk, in cool daylight

Is Dark Chocolate Good for You? What the Studies Actually Show

Few foods have had a better publicist than dark chocolate. For twenty years a steady supply of studies, press releases and headlines has told people that the bar they wanted anyway is good for their heart, their brain and their mood, and a fine food shop has every commercial reason to repeat it. This page does not. The health case for dark chocolate is real, small, and routinely oversold, and a reader deserves to know which parts are which. Here is what the better studies show, what they do not, the heavy-metal question that surfaced in 2022, and an honest verdict. Our percentage piece explains what "70 percent" is measuring; that matters here, because the health claims are about the cacao, and the bar is not all cacao.

What is in the bar that might matter

Cocoa beans are unusually rich in flavanols, a family of plant compounds, particularly epicatechin, that have measurable effects on blood vessels: they increase the production of nitric oxide, which relaxes the vessel walls, which modestly lowers blood pressure and improves blood flow in the hours after eating. That is the mechanism behind nearly every health claim. Two things about it matter before the studies. First, flavanol content varies enormously between chocolates, because fermentation, roasting and especially Dutch-processing destroy flavanols, so two 70 percent bars can differ several-fold, and the number on the wrapper tells you nothing about it; our fermentation piece covers the trade-off between flavor and flavanols. Second, the bar also contains sugar, fat and calories, roughly 170 per ounce for a dark bar, and any honest accounting has to include them.

What the studies show

Claim Best evidence What it found Strength
Lowers blood pressure Cochrane review, 2017: 35 trials, 1,800 people, flavanol-rich cocoa or chocolate for two to eighteen weeks A small, statistically significant drop: about 1.8 mmHg systolic, 1.8 diastolic, on average. Real, and smaller than one medication or a daily walk. Moderate
Reduces heart disease and death COSMOS trial, 2022: 21,000 older adults randomized to a cocoa flavanol supplement (500 mg a day) or placebo for a median of 3.6 years No significant reduction in total cardiovascular events (the primary outcome). A 27 percent reduction in cardiovascular deaths, a secondary outcome the authors called promising and in need of confirmation. The dose was a purified extract, not chocolate. Suggestive, unconfirmed
Improves memory or cognition Small trials of flavanol drinks; one COSMOS sub-study on cognition, 2023 Short-term improvements in blood flow to the brain and some test performance in small studies; the COSMOS cognition study found no overall benefit, with a possible effect in people whose diets were poor to begin with. Weak
Improves mood Small, short trials; observational surveys People report feeling better after eating chocolate, which is not a finding that needed a study. Any specific pharmacological effect is unproven. Weak
Helps with weight or metabolism Observational associations; small trials Some studies find dark chocolate eaters weigh less; that is correlation among people who eat small amounts of good chocolate rather than large amounts of candy. The bar itself is calorie-dense. Weak
Antioxidants prevent disease Laboratory antioxidant measures (ORAC), since abandoned by the USDA Chocolate scores high on tests that do not predict anything in the body; the USDA removed its ORAC database in 2012 because the numbers were being misused in marketing. Not evidence

Read together: cocoa flavanols do something real and modest to blood vessels; the one large trial designed to see whether that turns into fewer heart attacks did not confirm it, though a secondary result was encouraging; and the rest of the claims range from unproven to meaningless. That is a more interesting answer than the headline, and it is the honest one.

The dose problem

The trials that found effects used flavanol amounts that are hard to get from a bar. COSMOS gave 500 milligrams of flavanols a day in capsules; a good dark bar might deliver 50 to 250 milligrams per ounce depending on processing, which would mean two to ten ounces of chocolate a day to match, and two to ten ounces of chocolate is 350 to 1,700 calories and a great deal of sugar. The studies are, in effect, about cocoa extract, and the bar is a pleasant, inefficient, calorically expensive way to take a small dose of it.

The heavy-metal question

In late 2022 Consumer Reports tested 28 dark chocolate bars and found that most contained measurable cadmium or lead, and that a one-ounce daily serving of about two-thirds of them would exceed California's very strict Proposition 65 thresholds for at least one of the two. The finding was real and worth taking seriously, with three qualifications the headlines dropped. The California thresholds are set far below the levels regulators elsewhere consider harmful, by design. Cadmium comes from the soil and is taken up by the cacao tree, so it varies by origin. Lead arrives mostly after harvest, from dust during drying and transport, and is lower where beans are dried off the ground. For craft bars this means one thing: makers who source carefully and dry beans off the ground, as our fermentation piece and the producer guides describe, tend to test lower. A square a day is not a concern for most adults on any published assessment; children and people who are pregnant are the groups for whom moderation is advised, and the same is true for every dark chocolate, craft or not.

What the wellness copy leaves out

That the bar is a third to a half sugar by weight below 70 percent, and still a quarter at 70. That an ounce is about 170 calories, which is not nothing. That milk chocolate and Dutch-processed cocoa have most of the flavanols removed, so the health case, such as it is, applies to dark chocolate and natural cocoa only. That the flavanol content is invisible on the label. And that a person who eats a large bar because it is "healthy" has made a worse decision than a person who eats a square because it is delicious.

The honest verdict

A square or two of good dark chocolate a day is a pleasure with modest evidence behind it: a small, real effect on blood pressure, a plausible but unconfirmed effect on the heart, and no meaningful downside for most adults at that amount. It is not a medicine, a supplement or a reason to eat more chocolate than you would have. Choose it for the flavor, which is the thing our craft chocolate guide exists to explain; choose a maker who publishes where the beans come from and how they are handled, which addresses the metal question better than any percentage; and eat it slowly, which is the only health advice about chocolate that has never been contradicted. The bars are at 263 Josephine Street in Cherry Creek, Denver, and ship nationwide from theambrosianpantry.com. None of this is medical advice; the sources below are the studies themselves.

Sources & Further Reading

Back to blog