"Trans fat ban could save 7,200 lives by 2020, says study," The Guardian reports. This is the conclusion of a new modelling study looking at whether banning trans fatty acids – associated with "bad" cholesterol and heart disease – would improve public health outcomes.
Trans fats make up around 0.8% of the estimated energy consumption of the average UK diet. There are two types of trans fat: naturally occurring trans fatty acids found at low levels in meat and dairy products, and artificially made trans fatty acids labelled as hydrogenated fats or oils in some processed foods.
The artificial type became a popular ingredient for the food industry as they help prolong shelf-life while also improving taste. But research has shown a link between trans fats and heart disease. This led to several countries banning the use of artificial trans fats in food products.
In the UK there is no ban, but in 2012 most supermarkets and the bigger fast food chains agreed to sign up to a voluntary agreement not to use artificial trans fats. It is unclear how many products still contain trans fats.
The researchers calculated how many deaths they think could be avoided if a total ban was imposed, and what savings could be made in health and other costs.
While the figures are interesting, they are all based on assumptions fed into a mathematical model. It's hard to know how accurate these predictions are.
The study was carried out by researchers from the University of Lancaster, the University of Liverpool and the University of Oxford, and was funded by the National Institute for Health Research.
Most of the UK media covered the study accurately, although few questions were asked about how the figure of 7,200 prevented deaths had been reached.
Oddly, the Daily Mirror claimed a ban on trans fats "could prevent at least 10,000 deaths". They appear to have added a calculated 3,000 reduction in unequal deaths to the total 7,200 deaths prevented, when actually the 3,000 figure is part of the 7,200.
This was an epidemiological modelling study, which means it used data gathered about populations to create mathematical models to estimate the effect of possible changes in policy.
This type of study is a useful way to calculate the possible future effect of change, but it cannot be seen as a precise prediction of exactly what will happen.
Researchers used several big data sets and the results of previous studies to construct mathematical models about the possible effects of three policies over the next five years:
They then calculated the effects in terms of deaths avoided or delayed, healthcare costs, costs to the economy, and the effect on health inequalities.
The researchers used the findings of a 2006 meta-analysis, which estimated the effect of how many trans fats we eat as a proportion of total energy intake. The meta-analysis found there are 23% more new cases of heart disease for every 2% of total energy that comes from trans fats.
The researchers then combined this figure with information from questionnaires from the National Diet and Nutrition Survey (an ongoing government project to monitor dietary trends) to find out what proportion of people's diets consisted of trans fats.
They also used data about the socioeconomic status of people in England, which assigned people into five groups depending on their wealth and levels of deprivation.
They used a mathematical model to calculate the different effect various policies might have on these groups – for example, people in the lowest socioeconomic group eat the most trans fats as percentage of diet, so any policy that affected this group more would have a bigger effect on health overall.
The researchers made assumptions for their models. For example, they assumed changing labelling would have a bigger effect on people in higher socioeconomic groups than on lower groups, and people from lower socioeconomic groups were more likely to eat at fast food outlets and less likely to eat at restaurants.
They did multiple calculations using this data to work out the possible reduction in deaths from heart disease, the savings to the taxpayer, the effect on health inequalities, and the savings to the economy overall.
The researchers calculated an outright ban on the use of trans fats in food products would cut the amount of trans fats eaten by half, from 0.8% to 0.4% of total energy – the remainder would be the amount still consumed from naturally occurring trans fats in meat and dairy.
Their models found improved labelling or bans in restaurants and fast food outlets would, at best, achieve half that reduction, lowering trans fat consumption to around 0.6% of total energy.
They say most of the benefit from improved labelling or restaurant bans in terms of trans fat reduction would be seen among higher socioeconomic groups, so the policies would widen health inequalities.
In contrast, they say a total ban would affect lower socioeconomic groups more because they eat more trans fats, so it would narrow health inequalities. The researchers suggest the "gap" between the numbers of people from upper and lower groups who died of heart disease would narrow by about 3,000 people with a total ban.
They used the figures from the previous analysis to calculate the effect of this reduction in trans fat in the diet. They assumed deaths from heart disease would fall at the same rate as numbers of new cases of heart disease, giving a total figure of 7,200 deaths delayed or avoided over five years from a total ban (95% confidence interval [CI] 3,200 to 12,500).
They said improved labelling or restaurant bans might delay or avoid 1,800 to 3,500 deaths, depending on the model used. They claim a total ban would save £297 million (95% CI £131 to £466 million). These savings mainly represent savings in "informal care" – the care given to people with heart disease by friends and family.
They also included productivity at work and healthcare costs. Estimated direct healthcare savings are relatively small, at around £42 million, while estimated informal care savings are £196 million.
The researchers say their findings show that, "elimination of trans fatty acids from processed foods is an achievable target" and "would lead to health benefits at least twice as large as other policy options".
They warn trans fats "could creep back into processed foods" if action is not taken now to ban them completely.
Trans fats are already at low levels in the UK diet compared with 10 or 20 years ago. However, this study suggests lowering them even further could reduce the number of people getting and dying from heart disease over the next five years.
This study does have limitations, however, which means we cannot rely on the findings to be precise. Any study that uses a mathematical model relies on the researchers making correct assumptions when they feed in the data.
The researchers say they have had to make assumptions based on little data in some cases. For example, there is no information about what proportion of the diet is made up of trans fats for those in the top socioeconomic class. We also don't know what proportion of trans fats are consumed in restaurants or fast food outlets.
More importantly, it is possible reducing consumption of trans fats will not have the effect on heart disease the researchers think it will. They used a study from 2006 that combined the results of previous trials to come up with their figure. But this study's finding that trans fats are linked to an increased chance of heart disease does not automatically mean reducing trans fats will reduce the chance of heart disease by the same amount.
However, it does seem likely reducing trans fats will reduce the numbers of people getting heart disease and dying from it. Whether or not banning trans fats will have exactly the effect the researchers predict is less certain.