Browse evidence-based analysis of health-related claims and assertions
Eating more saturated fat doesn't appear to raise your risk of having a stroke, according to a review of 8 long-term studies of adults.
Correlational
Eating more saturated fat, like butter or fatty meat, doesn't seem to raise your risk of heart disease, according to a big review of 16 long-term studies of adults.
We can’t be sure if saturated fat is good or bad for you because the studies are all over the place—some say one thing, others say the opposite, and the measurements aren’t very accurate.
Descriptive
Some studies say trans fats raise diabetes risk, others don’t—this might be because some studies account for healthy foods like whole grains and nuts, and others don’t.
Some types of saturated fat in your blood—especially from processed carbs—are linked to higher diabetes risk, while others from dairy are linked to lower risk.
If you swap out trans fats (like in fried foods) for healthy fats like those in fish or nuts, your risk of heart disease drops a lot—what you replace it with matters more than just cutting fat.
There’s a hint that eating more saturated fat might raise the risk of dying from heart disease, but the data is too messy and uncertain to say for sure.
Eating trans fats doesn’t clearly raise or lower the risk of a certain type of stroke called ischemic stroke—the results are mixed and not reliable.
People who eat more trans fats—especially from processed foods—are more likely to die from any cause, including heart disease and cancer.
A specific type of fat found in milk and cheese is linked to a much lower chance of getting type 2 diabetes, possibly because it helps the body use insulin better.
Trans fats from cows and other animals (like butter) don’t seem to raise heart disease risk, but trans fats made in factories (like in margarine) do—so where the fat comes from matters.
Eating more saturated fat from meat, butter, or cheese doesn’t clearly make people more likely to die early or get heart disease or diabetes, but the science isn’t strong enough to be sure.
People who eat more artificial trans fats—like those in fried and baked goods—have a higher chance of having a heart attack or dying from heart disease.
Over the years from 2000 to 2006, both 'bad' and 'good' cholesterol levels in heart disease patients admitted to the hospital got lower on average.
Women with heart disease tend to have higher 'bad' cholesterol levels than men when they are first admitted to the hospital.
Very few heart disease patients—less than 1 in 70—have both their 'bad' cholesterol low enough and their 'good' cholesterol high enough to meet ideal health targets.
More than half of heart disease patients admitted to the hospital have low levels of 'good' cholesterol, which is considered unhealthy.
When people with heart disease are first admitted to the hospital, their 'bad' cholesterol is usually around 105, which is higher than what doctors now recommend.
The model suggests that giving a population a little shock now and then might help it recover better than keeping the stress on all the time—but this is just math, not proven in real life.
The model can tell the difference between when the whole group bounces back after stress, versus just one part of it (like healthy individuals surviving while sick ones die).
You don’t need to believe that low doses of poison help you—this math model can make the same 'helpful low dose' curve just by using how fast populations grow.
Mechanistic
The math model can find the exact point where a little stress helps a population, but too much hurts it—like a tipping point between recovery and collapse.
Quantitative
For a population to bounce back stronger after a small stress, it needs to be really good at healing or reproducing fast—otherwise, it just stays down.
If there aren’t enough individuals left after a shock, the population can’t recover—even if the stress was small. The model says starting size matters a lot.