"Butter isn't bad for you after all: Major study says 80s advice on dairy fats was flawed," is the headline on the front of the Daily Mail as a new study argues dietary fat guidelines introduced in the 1980s lacked a rigorous evidence base…
"Butter isn't bad for you after all: Major study says 80s advice on dairy fats was flawed," is the headline on the front of the Daily Mail as a new study argues dietary fat guidelines introduced in the 1980s lacked a rigorous evidence base.
The study in question looked at guideline advice on saturated fat published in 1983 in the UK and in 1977 in the US. The researchers wanted to see if the evidence available at the time – specifically, the results of randomised controlled trials (RCTs) – supported the recommendations made.
The researchers identified six RCTs available at the time. The pooled results showed that specific advice to control saturated fat intake did not have a significant effect on deaths from heart disease or other causes.
But it is very important that these findings are interpreted in the correct context – this means we cannot conclude the recommendations were "incorrect".
We do not know what evidence was used to back up the official guidelines in the late 70s and early 80s. They could have looked at studies other than RCTs, such as observational studies (where health outcomes are studied over time).
This new review considered just six RCTs published before 1983, and all of them were conducted in men, most of whom already had heart disease.
Current dietary advice is not stuck in the 1980s, wearing shoulder pads and sporting a bubble perm. It has evolved as new evidence has emerged. In fact, a small amount of saturated fat is recommended as part of a balanced, Mediterranean-style diet.
But it would be a mistake to conclude from this evidence that you can eat as much saturated fat as you like without damaging your health.
The eatwell plate
The government's healthy eating advice is set out in the eatwell plate, which shows the foods we need for a healthy, balanced diet.
The eatwell plate shows how much you should eat of each food group. This includes everything you eat during the day, including snacks. So, try to eat:
- plenty of fruit and vegetables
- plenty of bread, rice, potatoes, pasta, and other starchy foods (choose wholegrain varieties whenever you can)
- some milk and dairy foods
- some meat, fish, eggs, beans, and other non-dairy sources of protein
- just a small amount of foods and drinks high in fat and sugar
Where did the story come from?
The study was carried out by researchers from the University of the West of Scotland, Cardiff Metropolitan University, and the University of South Wales in the UK, and Saint Luke's Mid America Heart Institute in the US.
No sources of funding are reported and the authors declare no conflicts of interest. But the lead author of the study, Zoë Harcombe, does run a commercial diet plan called The Harcombe Diet®, which promotes "eating real food", including dairy products.
The study was published in the peer-reviewed medical journal, Open Heart. This is an open-access journal, so the study can be read online for free or downloaded as a PDF.
Overall, the media reporting was poor and potentially quite dangerous for several reasons. Much of the reporting gives the impression that the claim "saturated fats are not bad for you" represents a change in official dietary advice. This is not the case. The claim is the opinion of a small group of researchers.
The headline writing was particularly sensationalist. Journalists took the findings at face value, writing potentially scaremongering headlines, and may leave their readers questioning the evidence base for the current guidelines. Debate on national guidelines is always welcome, but much of the debate in the media was ill-informed.
Even if you did accept the claim that saturated fat is not especially harmful, it certainly does not follow that eating more of it would be good for you, as the Daily Express bizarrely claimed: "Fat is key to living longer". A diet high in saturated fat could in fact lead to obesity.
A more balanced account of the relevance of this study was given in the accompanying Open Heart editorial, which is also open access.
Various dietary experts have considered this research in a broader context. The consensus from the experts is that only focusing on evidence from RCTs for dietary guidelines is somewhat unrealistic and narrow, and missed useful evidence of other types.
Some went a lot further. Professor Christine Williams, professor of human nutrition at the University of Reading, said: "The claim that guidelines on dietary fat introduced in the 1970s and 80s were not based on good scientific evidence is misguided and potentially dangerous."
What kind of research was this?
This was a systematic review and meta-analysis of research published 30 or more years ago.
It aimed to investigate whether national dietary advice introduced in the 1970s and 80s in the US and the UK to reduce coronary heart disease (CHD) by reducing saturated fat intake was backed up by contemporaneous evidence from randomised controlled trials (RCT), which are seen as the "gold standard" in evidence-based medicine.
In the US, public health dietary advice was issued by the Select Committee on Nutrition and Human Needs in 1977. This was followed in 1983 by UK public health dietary advice from the National Advisory Committee on Nutritional Education.
The authors state these recommendations advised that people reduce their overall fat consumption to 30% of their total energy intake, and reduce their saturated fat consumption to 10% of their total energy intake.
The researchers discuss several possible limitations of these publications, saying they included wording that was far from conclusive, such as "tended to be related", and how neither publication made reference to any RCTs available at the time.
The authors of this review therefore aimed to look for RCTs available when the dietary guidance was published to see whether the available evidence supported the recommendations.
What did the research involve?
The authors searched the literature databases Medline and the Cochrane Library to identify studies published up to 1983. They restricted their search to these two databases because others did not provide adequate coverage of the early publications this review was interested in.
They looked for RCTs in adults of at least one year's duration where:
- people were randomised to a dietary intervention (a programme that attempted to control or modify specific elements of their diet)
- the aim of the study was to look at whether a reduction or modification in dietary fat or cholesterol had an effect
- health outcome data on all-cause mortality, CHD mortality and cholesterol measurements was available
Six RCTs met their inclusion criteria:
- Rose Corn Oil Trial
- Research Committee Low-fat Diet
- MRC Soya-bean Oil
- LA Veterans Study
- Oslo Diet Heart Study
- The Sydney Diet Heart Study
The researchers extracted data from these studies and considered their quality and risk of bias. They pooled the results of these trials in a meta-analysis.
What were the basic results?
The six RCTs included a total of 2,467 men, and all but one of the studies looked at secondary prevention. This means the participants already had cardiovascular disease.
The researchers involved in these RCTs looked at whether dietary intervention could reduce the risk of further disease events, such as a heart attack. The average study duration was five to six years.
Of the six studies, four looked at giving vegetable oil (three of which assessed it as a replacement for saturated fat), one looked at a roughly 20% fat diet, and one looked at a 10% saturated fat diet.
As this new study points out, five of the six RCTs did not look at either a total fat consumption of 30% or saturated fat as 10% of energy intake, as given in the offical recommendations made in the 70s and 80s.
Across the studies, 30.2% of the intervention groups and 29.8% of the control groups died. The pooled results of all the studies found no statistically significant effect of the dietary interventions on deaths from all causes (relative risk [RR] 0.996, 95% confidence interval [CI] 0.865 to 1.147).
The pooled results did not find that the dietary interventions had any significant effect on coronary heart disease mortality specifically (RR 0.989, 95% CI 0.784 to 1.247).
Cholesterol levels fell in both intervention and control groups, though there was a greater reduction in the intervention groups. The pooled reduction in the intervention groups was a reduction of 12.6% (give or take 6.7%), while the reduction in the control groups was 6.5% (give or take 5.1%).
How did the researchers interpret the results?
The researchers say that: "No randomised controlled trial had tested government dietary fat recommendations before their introduction.
"Dietary recommendations were introduced for US and UK citizens by 1983, in the absence of supporting evidence from RCTs."
They state that: "The present review concludes that dietary advice not merely needs review; it should not have been introduced."
This research found the pooled results of six RCTs available prior to 1983, which all looked at interventions to moderate saturated fat intake, did not find this had an effect on deaths from heart disease or any other cause.
But it is very important that the specific purpose of this review is considered, and the findings are interpreted in the right context.
This review specifically looked at nutrition guidance given in the US in 1977 and in 1983 in the UK. In particular, the researchers looked at two recommendations:
- reduce overall fat consumption to 30% of total energy intake
- reduce saturated fat consumption to 10% of total energy intake
The researchers specifically wanted to see whether RCTs available at that time supported those recommendations. But there are some specific points to consider from the results of this analysis.
Evidence considered by the 1977 and 1983 guidelines
We are not able to review the methods used by the US and UK government bodies in forming their nutrition recommendations. We also do not know what evidence they considered.
The authors of the current review state that: "Both documents acknowledged that the evidence was not conclusive ... the Dietary Goals for the US noted 'there will undoubtedly be many people who will say we have not proven our point.' The UK publication referred to 'a strong consensus of opinion'."
We cannot comment further on how this nutrition guidance may have been produced, or how they may have considered their evidence and formed their recommendations.
It is possible the methods used by these organisations more than 30 years ago may have differed from those used in producing the most robust evidence-based guidelines today.
But we cannot conclude that the recommendations were unreasonable, or not backed by any supporting evidence, just by looking at the information included in this systematic review.
It is likely the guidance from more than 30 years ago may have considered observational evidence looking at how saturated fat intake was related to mortality and heart disease.
While the lack of RCTs at the time may potentially be of concern, it is also not particularly surprising. RCTs involving diet are notoriously difficult to run because of compliance issues: researchers can never be sure that participants are sticking to their recommended diet plans. Also, exposing participants to an intervention you think may be harmful is unethical.
It is also not fair to say observational study designs are of no value. In fact, when looking at issues such as dietary patterns, there is often more information available from observational studies. These types of studies can review a person's longer-term lifetime dietary patterns and see how this is related to health outcomes.
We therefore cannot conclude that the recommendations made over 30 years ago were "incorrect". This review has kept a narrow focus, only looking at RCTs available at the time.
Other types of study, such as long-term population-based studies, can provide rich and useful information, and may have been considered when the guideline recommendations were being formed.
Issues with the RCTs included in the review
The finding there is no link between saturated fat intake and deaths from heart disease and other causes is based on six very specific RCTs. These studies are all likely to have differences in terms of their design, the methods used, their duration, and their quality.
The studies only included men, so the results may not be applicable to women, for a start. And five out of six included men who already had heart disease.
The studies also looked at the relatively short-term effects of specific interventions (such as substituting vegetable oil) and whether this influenced outcomes. They did not look at lifetime dietary patterns. Overall, they found the interventions did not affect outcomes.
This review's findings do not mean that current government nutritional advice and recommendations (the eatwell plate) are wrong.
National dietary guidance is based on all the relevant evidence that has accumulated to date, and is updated regularly to consider important new evidence. Current guidance will have considered a much larger body of research than the guidance issued in 1983.
There's nothing wrong with the occasional buttered scone. But, based on the current body of evidence, it would be potentially dangerous to think you can eat as much saturated fat as you want without it having an effect on your health.
Analysis by Bazian. Edited by NHS Choices. Follow Behind the Headlines on Twitter. Join the Healthy Evidence forum.