Food study didn’t show that later breakfasts raise diabetes risk

A new study suggested that eating breakfast after 9 am increased diabetes risk

An article from MailOnline repeated claims in the researchers’ press release that a study shows that eating a later breakfast increased people’s risk of diabetes.

The MailOnline article says the study “suggested that eating breakfast after 9 am increased diabetes risk”, and goes on to say: “the Spanish and French researchers noted that those who opted for breakfast before 8 am reduced the risk of developing type 2 diabetes by 59 percent. What's more, eating a later dinner also appeared to increase the likelihood of the condition.” 

But a study author confirmed with Full Fact that this isn’t what the scientific article said.

It actually looked at data on people’s health and eating habits and suggested an association between later average breakfast times and diabetes—but didn’t say that one was causing the other.

We have written before about issues with health research reporting.

Mistaken reporting about research can lead to mistrust of science from the public, or lead to people making decisions based on incorrect information. It is important that research institutions don’t exaggerate their findings, and that media outlets correct errors once they are discovered.

What was the study?

Researchers compared the meal timings and frequency of over 100,000 people (79% of them female), using data from the NutriNet-Santé study (a long term study in France looking at the diet, lifestyle and health of participants who complete online follow up questionnaires) and looked at whether or not they developed type 2 diabetes. 

The researchers split people into groups, according to what time they ate their first and last meals of the day. They found that people who on average ate breakfast before 8am were 59% less likely to develop type 2 diabetes than people who ate breakfast after 9am. This is the headline finding that the press release and MailOnline article are based on.

However, the study is the work of two groups of researchers—one from the Barcelona Institute for Global Health (ISGlobal), and one from the Nutritional Epidemiology Research Team in Paris. The press release came from ISGlobal.

Full Fact understands that the press release does not necessarily reflect the view of the scientists investigating the NutriNet-Santé cohort. 

Causation or correlation

The English press release contains the statements "Eating breakfast after 9 a.m. increases the risk of developing type 2 diabetes by 59% compared to people who eat breakfast before 8 a.m" and "The results show that we can reduce the risk of diabetes not only by changing what we eat, but also when we eat it".

By using the terms “increases the risk” and “we can reduce the risk”, this clearly suggests that the time people eat can change their risk of developing diabetes.

This isn’t what the study showed. 

The fact that people who ate breakfast earlier in the day went on to develop type 2 diabetes less often doesn’t mean that their early breakfasts were the reason. For example, people eating breakfast later in the day might also have other differences in their lifestyle that affect their risk of diabetes.

A researcher on the study confirmed to us that results from a single epidemiological study cannot infer that eating a late breakfast causes type 2 diabetes, and that causal language shouldn’t be used to describe the findings as other studies are needed to replicate these findings.

A spokesperson for ISGlobal told Full Fact that the “published paper [...] in no way suggests a causal relationship” and accepted that a sentence from its press release taken out of context “could lead to confusion” but maintained that the “overall tone” of the press release was “correct”.

We have contacted MailOnline about these issues but have not received a response at time of writing.

Featured image courtesy of Arjan Richter