Food and diet

Study finds link between saturated fats and early death

"Eating more saturated fats raises risk of early death, says US study," The Guardian reports.

A major study involving more than 80,000 women would seem to contradict recent high-profile reports that a diet rich in saturated fat is safe.

The latest – a long-term study conducted in the US including more than 120,000 people – found that swapping saturated fat and / or trans fats for polyunsaturated fat such as olive oil could reduce the risk of dying by 27%.

The case for and against saturated fats, including butter and cheese, has changed with different scientific studies in recent years. A recent report, published by the Public Health Collaboration, argued that the official advice on low-fat diets was actually making the obesity epidemic worse; though the report was far from systematic, as we discussed earlier this year.

A more rigorous summary of research published in 2015 found no link between saturated fats and death.

The difference in the conclusions, say the latest researchers, could be because the previous summary of research could not say what people who ate less saturated fat ate instead. In a Western diet, they say, people who eat less saturated fat might eat more sugar or refined carbohydrates, which are known to be linked to type 2 diabetes and cardiovascular disease. This study, they say, allows researchers to calculate the effects of swapping one type of fat for another.

Overall, the study supports official dietary advice to replace saturated and trans fat with unsaturated fat. 

Where did the story come from?

The study was carried out by researchers from Harvard TH Chan School of Public Health and Brigham and Women's Hospital and Harvard Medical School and was funded by the US National Institutes of Health.

The study was published in the peer-reviewed journal JAMA Internal Medicine and has been reported accurately by the UK media.

The UK media highlights the recent confusion and controversy regarding the place of saturated fats in a healthy diet; without acknowledging that much of the confusion has been driven by its own coverage.

For example, The Sun says: "Scientists have been so split in recent years over the dairy food's effect on health that consumers often don't know who to believe." But the so-called "split" is actually a small number of researchers who have argued against official advice.

There is nothing wrong with questioning received wisdom. If nobody ever did that then we would still think that the Earth is flat and at the centre of the universe.

What is misleading is when the media presents a minority opinion as a sudden sea-change in scientific consensus. Official guidelines, whether from the Department of Health, the World Health Organization, or the US Food and Drug Administration, have remained consistent about the dangers of saturated fats.

What kind of research was this?

This study combines results from two ongoing cohort studies that started in the US in the 1980s, with a combined total of 126,236 participants. The authors wanted to compare what people reported about their diets (measured around every four years) with their health records, during the three decades since the studies began.

Large, prospective cohort studies with long follow-up periods are the best ways to look at links between lifestyle factors such as diet and health, because it's impractical to do a long-term randomised controlled trial of something like diet (it would be hard to ensure someone eats the same thing every day for 30 years). However, observational studies can never prove that one factor causes another.

What did the research involve?

Researchers followed two big groups of health professionals for around 30 years. They assessed their health, lifestyle and diet at the start of the study and every two to four years. After adjusting their figures to take account of known risk factors, the researchers calculated the effect on the chances of having died from any cause, or from specific causes, of consuming different types of dietary fat.

Data came from the Nurses' Health Study (83,349 women, starting 1980) and the Health Professionals Follow-Up Study (42,884 men, starting 1986), up to 2012. Researchers calculated the effect of the total dietary fat (compared to total carbohydrate in the diet), then looked at the effect of specific dietary fats compared to carbohydrates.

Specific fats included:

  • saturated fat (from red meat and dairy products)
  • polyunsaturated fat (for example sunflower oil or soyabean oil) or monounsaturated fat (for example olive oil and peanut oil)
  • trans fats (chemically converted fats) – most people in the UK don't eat a lot of trans fats as in recent years many UK food manufacturers have removed trans fats from their products
  • omega 3 and 6 fatty acids

Finally, they calculated the effect of swapping 5% of dietary energy intake from saturated fats or trans fats with polyunsaturated or monounsaturated fats. The researchers took account of a wide range of confounding factors, including:

  • age
  • ethnicity
  • body mass index (BMI)
  • physical activity
  • smoking
  • medical history

They also carried out a number of sensitivity analyses to check whether it might affect the results if people changed their diet as a result of being diagnosed with an illness.

What were the basic results?

The study found:

  • People who ate more saturated fats (compared to carbohydrates) were 8% more likely to have died during the study than those who ate least saturated fats (hazard ratio (HR) 1.08, 95% confidence interval (CI) 1.03 to 1.14).
  • People who ate more polyunsaturated fats or monounsaturated fats were slightly less likely to have died during the study than those who ate least unsaturated fats (HR polyunsaturated fats 0.81, 95% CI 0.78 to 0.84; HR monounsaturated fats 0.89, 95% CI 0.84 to 0.94).
  • People who ate more trans fats were 13% more likely to have died during the study than those who ate least trans fats (HR 1.13, 95% CI 1.07 to 1.18).

In addition, they found that swapping 5% of energy from saturated fats with the equivalent energy from polyunsaturated fats would reduce death rates by 27% (HR 0.73, 95% CI 0.7 to 0.77). Swapping saturated fat for monounsaturated fat would also have an effect, but not as big, they found.

How did the researchers interpret the results?

Researchers said their findings showed different types of fat had different effects on health, and that "these findings support current dietary recommendations to replace saturated fat and trans-fat with unsaturated fat".

They say previous studies might have produced different findings because they didn't look at what people in studies were eating instead of saturated fat, and did not calculate the effects of swapping one type of fat for another.


It's confusing when advice about healthy eating seems to change with each study published, and the experts say different things. However, when you look closely, the two studies mentioned here do not necessarily contradict each other.

The researchers who carried out last year's summary of research warned that their conclusions might change based on future studies, and said they had "very low" confidence in the results, because of the quality of the studies that had previously been carried out. We concluded last year that the summary did not rule out the possibility that saturated fat might be harmful, and that we need to know what people should eat instead of saturated fat.

This study has several strengths. It is very big, carried out over several decades, and checked people's diet every few years, so the researchers can assess the effects of changing diet over time. The researchers also adjusted their figures to take account of confounding factors. By calculating the possible effects of saturated and unsaturated fat compared to carbohydrate, it allows the researchers to calculate the effects of swapping one type of fat for another.

The findings about trans-fats are unsurprising and not controversial. These artificially created fats, used in baked goods, are being phased out because of their effects on health.

Cohort studies can never prove causality beyond doubt, so we cannot say that this shows saturated fat causes earlier deaths. However, the study does provide evidence that different types of fats have different effects on health, and swapping to healthier fats may be preferable.  

NHS Attribution