Study linking gum disease and mental health shows correlation not causation

20 December 2021
What was claimed

Poor oral health raises the risk of mental illness by a third.

Our verdict

A study found people with diagnosed gum disease were more likely to go on to develop a mental health condition, but did not say whether one factor caused the other, as suggested.

“Poor oral health ‘raises risk of mental illness by a third’”

The print version of the Telegraph has reported a new study on the links between gum disease and the risk of chronic health conditions, claiming “gum disease increases the risk of developing mental health problems by more than a third.”

The study, led by researchers at the University of Birmingham, found a link between the two, but did not claim gum disease was responsible for this increase in risk, as the Telegraph’s wording suggests.

The university’s own news report described the findings with the headline: “Gum disease increases risk of other illness such as mental health and heart conditions, study finds”, which may have given the misleading impression that the study examined causal relationships. 

The research, published in BMJ Open, compared the risk of various chronic health issues between a cohort of people with periodontal disease on their GP record and a cohort without confirmed periodontal disease.

Lead author Dr Joht Singh Chandan told Full Fact that one limit to the study is the fact that some people with a record of periodontal disease may not have had that confirmed by a medical professional, and vice-versa, some people in the other cohort may have undiagnosed periodontal disease. 

The cohorts were matched for age and level of deprivation, which removed the impact of those two factors on the results. The two cohorts were also similar in terms of their sex, weight, smoking status and ethnicity distribution.  

It found people with periodontal disease were 79% more likely to have a mental health condition at the time they were included in the study, and 37% more likely to develop a mental health condition over the course of the study, than the other cohort. 

The authors mention that outcomes of gum disease, such as halitosis and teeth loss may negatively impact someone’s social life and therefore their mood, possibly leading to depression.

But they also acknowledge the potential for a causal relationship in the other direction, saying: “Individuals under increased stress may reduce health promoting behaviours (eg, optimal oral hygiene practices).”

Typically, a study trying to ascertain a causal relationship between two factors (for example gum disease and mental illness) would control for more than basic demographics, and Dr Singh Chandan told Full Fact the research “can’t be causal because there will be lots of other factors we haven’t accounted for.”

The Telegraph’s article is just the latest in a long line of media reports which overstate the findings of research as showing causal relationships when they in fact show correlation. 

Over the past six months, Full Fact has written about the issue in relation to a study linking sleep time and heart disease, tea consumption and dementia, and flavonoids and Alzheimer’s.

The Science Media Centre, which helps support scientists and journalists to improve science reporting, says in its best practice guidelines for journalists: “When reporting a link between two things, indicate whether or not there is evidence that one causes the other.” 

We deserve better than bad information.

After we published this fact check, we contacted The Daily Telegraph to request a correction regarding this claim.

They disputed our fact check and did not issue a correction.

We also contacted the University of Birmingham about this. They amended the headline of their press release after we made them aware of our fact check.

It’s not good enough.

Will you add your name for better standards in public debate?

Full Fact fights bad information

Bad information ruins lives. It promotes hate, damages people’s health, and hurts democracy. You deserve better.