'Stop demonising butter,' say researchers

"Butter has been wrongly 'demonised' as unhealthy," reports the Daily Express following the publication of a study that found eating butter did not increase the risk of heart disease, stroke and diabetes.

Researchers analysed the findings from nine studies published since 2005 involving more than 600,000 participants from 15 countries, including the UK.

They found eating 14g a day of butter – about a tablespoon – had little to no effect on overall risk of death, heart disease and stroke. Butter even seemed to protect, slightly, against diabetes.

This was a well-designed review, but it is only as good as the studies included, and in this case no relevant randomised controlled trials were found.

The studies involved were all cohorts, which do not show whether any other factors may be at play.

We also do not know if the participants were able to accurately remember their butter intake, which is a common issue with data collected by questionnaire.

When it comes to diet, the best approach is to consume high-energy products, such as fat and sugar, in moderation.

Maintaining a healthy weight and having a balanced diet with regular exercise is the best way to reduce your risk of cardiovascular disease and diabetes.

Where did the story come from?

The study was carried out by researchers from Tufts University and Stanford School of Medicine, both in the US, and the University of Sydney in Australia.

Funding was provided by a grant for Circulating Dietary and Metabolic Fatty Acids, Major CVD Outcomes and Healthy Aging.

It was published in the peer-reviewed journal, PLOS One and is free to read online.

This review was picked up by several UK media outlets, who mostly overstated the importance of the findings.

The Daily Mail provided a useful quote from one of the study authors, Dr Dariush Mozaffarian, who said: "Our results suggest that butter should neither be demonised nor considered 'back' as a route to good health."

What kind of research was this?

This systematic review and meta-analysis aimed to pull together information on the long-term association of butter consumption with major health conditions, such as cardiovascular disease, diabetes and death.

A systematic review is the best way of gathering the available evidence on a topic. However, they can be limited by the quality of the studies included.

All of the studies included here were observational, and may be of variable methods, outcomes and quality.

The results of a meta-analysis may be unreliable if there was a great deal of difference between the individual studies.

What did the research involve?

The research team searched nine medical databases for prospective studies (cohorts or trials) providing estimates of the effects of butter intake on the following outcomes:

death
cardiovascular disease, including heart disease and stroke
diabetes

The researchers excluded studies that were retrospective in design; those with less than three months follow-up; where the population had a major disease, such as cancer; or if it was not possible to distinguish consumption of butter from other dairy products or fats.

Titles and abstracts of search results were reviewed by one researcher, while two were responsible for reviewing potentially relevant full text articles.

All the studies were assessed using a risk of bias tool and statistical methods used to combine the findings of individual studies.

What were the basic results?

The database searches found 5,770 potentially relevant studies, of which only nine met their inclusion criteria, presenting data from more than 636,000 participants.

All the studies were considered high quality, although they were all observational cohort studies.

No randomised controlled trials, often considered the gold standard of scientific research, were identified.

The average age of the participants ranged from 44 to 71 years.

All the studies were published between 2005 and 2015, including populations from the UK and a number of other European countries.

Most used food questionnaires to obtain dietary information, while one study used interviews.

The average butter consumption across studies ranged from 4.5g to 46g per day.

Pooling of results found no clear evidence of any link between butter consumption and death or cardiovascular disease.

Pooling the results of nine groups of participants (covered by two large studies) found each additional 14g serving of butter was very weakly associated with a 1% increase in risk of death.

However, this finding was only of borderline statistical significance, meaning it doesn't provide good evidence that there's any link at all (relative risk [RR] 1.01, 95% confidence interval [CI] 1.00 to 1.03).

Butter consumption was not significantly associated with risk of cardiovascular disease (based on four studies), heart disease (three studies), and stroke (three studies) did not appear to increase.

Butter appeared to have a protective effect against type 2 diabetes (four studies), with a 4% decrease in risk with each 14g serving (RR 0.96, 95% CI 0.93 to 0.99).

There was no evidence to suggest bias as a result of differences in the way the results were collected in each study, known as heterogeneity.

There was also no evidence of publication bias – that is, studies with positive findings being more likely to be published.

How did the researchers interpret the results?

The researchers concluded: "This systematic review and meta-analysis suggests relatively small or neutral overall associations of butter with mortality, cardiovascular disease, and diabetes.

"These findings do not support a need for major emphasis in dietary guidelines on either increasing or decreasing butter consumption, in comparison to other better established dietary priorities; while also highlighting the need for additional investigation of health and metabolic effects of butter and dairy fat."

Conclusion

This systematic review and meta-analysis assessed the long-term association between butter consumption and major health conditions, such as cardiovascular disease, diabetes and death.

The nine identified studies found little to no evidence that butter consumption increased the risk of death, cardiovascular disease, heart disease or stroke.

Perhaps surprisingly, though, there did seem to be a protective effect against the risk of type 2 diabetes.

This review has both strengths and limitations that may affect the reliability of the findings.

The researchers used careful search methods that aimed to identify only studies relevant to the general population and the large number of people included.

The researchers also did their best to provide the most reliable estimate by performing a quality assessment of studies, assessing heterogeneity and risk of publication bias.

In terms of limitations, the review did not include any randomised controlled trials, which could have compared health outcomes with people who did not eat butter.

Other health and lifestyle factors may also be at play – like age, smoking, other dietary habits and physical activity – that the studies included, by nature of their design, are unable to account for.

The studies included mainly collected data using questionnaires, which are subject to various types of bias, and people may inaccurately report their butter consumption.

For example, we do not know if participants were asked about their consumption of products that contain butter, such as biscuits or cakes, and this may alter results.

It is also unclear from this report whether the studies included had good response rates. The studies mainly involved an older population, so we do not know if the same risk would apply to children or younger adults.