Fried food study does not reflect UK diet
The idea that regularly eating fried foods causes heart attacks is a ‘myth’, The Daily Telegraph has reported.
The newspaper has panned conventional wisdom based a large Spanish study investigating how people’s fried food consumption was linked to their risk of events such as heart attacks and episodes of heart pain. To study the relationship researchers surveyed over 40,000 people on how much fried food they ate during the previous year, looking at any episodes related to coronary heart disease over an average period of 11 years.
In this particular setting the researchers found no link between consuming food fried in olive or sunflower oils and the risk of heart disease (the UK’s biggest killer) or death from any cause. However, while such a phenomenon could exist in the context of this specific group of people eating a specific type of diet, the research does not support the idea that fried food is generally harmless, or that it represents diets among the UK population. For example, most participants reported eating a relatively small amount of fried food each day, much less that might be typically found in a fry-up or a fast food meal. In short, it is not clear whether the same study conducted in the UK would have the same results.
Many news sources have reported the results in an accurate, measured way. However, The Daily Telegraph’s account has been somewhat overcooked, as the study did not look at the impact of factors such as frying with other types of fats, reusing oils several times (as is the case in most fast food outlets), or consuming fried snacks high in salt, sugar or calories.
Where did the story come from?
The study was carried out by a collaboration of researchers from Spanish universities and other health organisations based in Spain. It was funded by Spain’s FIS Fund for Health Research and published in the peer-reviewed British Medical Journal.
Media reports were generally balanced, with many making the point that the study was primarily concerned with a Mediterranean diet containing olive oil, which is typically very different from the dietary patterns in the UK. This limits how relevant this study is to the UK population. Many news sources also rightly warn against assuming that the study results suggested that consuming large amounts of fried foods is not harmful. The Telegraph included a picture of a full English breakfast, which is not a recognised part of a Mediterranean dietary pattern.
What kind of research was this?
This prospective cohort study investigated the association between people’s consumption of fried foods and their risk of experiencing coronary heart disease events such as heart attacks and heart pain (angina) which require surgery. It was conducted in a Spanish population.
Previous research has shown fried foods have been associated with high blood pressure, obesity and low levels of ‘good’ cholesterol (high-density lipoproteins or HDL cholesterol). However, existing studies into the association of fried food and coronary heart disease provided mixed results, the authors report.
When food is fried the nutritional content changes; water is lost and fat is absorbed, increasing the calorie content of the food. During frying, the oils also deteriorate and change into other forms, especially when reused, as is often the case in fried fast food outlets. This process leads to a loss of unsaturated fats and an increase in harmful trans fats.
A prospective cohort study is a good way of investigating whether fried foods are linked to heart disease because you can be sure the consumption of the food occurred before the development of the disease. However, the limitation is that you cannot be certain that fried foods caused a disease because its development may be influenced by numerous other factors, some measured in the research and some not.
The best type of study to definitively assess this link would be a large randomised control trial, providing it could be performed ethically. However, given the cost and complexity of such a task, this approach may ultimately be impractical.
What did the research involve?
The participants were Spaniards enrolled in a research project called the European Prospective Investigation into Cancer and Nutrition (the EPIC-Spain cohort). The analysed cohort consisted of 40,757 adults aged between 29 and 69 who were recruited between 1992 and 1996. They were followed up for an average of 11 years to see what diseases they developed and what they died of.
When enrolled, participants were asked by trained interviewers to complete a questionnaire about the food they consumed over a typical week during the previous 12 months. Only foods consumed at least twice a month were recorded.
The study’s authors report that this method had previously been shown to be accurate (validated) at assessing the diet of Spanish people. The type of oil used for frying (olive oil versus sunflower oil or other vegetable oils) was recorded and then checked again after two years. No further information on diet was collected at a later date.
At enrolment, the researchers also recorded non-dietary variables. These included demographic variables such as age and sex as well as educational level, body mass index (BMI), smoking and physical activity levels. Participants were also asked if they had coronary heart disease, diabetes, high blood pressure, cancer or angina or had experienced a heart attack or stroke in the past.
For the analysis, people were divided into four groups depending on their level of fried food consumption. The researchers specifically analysed how consuming fried food was linked to the chance of having a coronary heart disease event and dying from any cause (known as all-cause mortality) and how these varied for the four groups.
The analysis was adjusted for a large number of factors, including energy intake, educational level, smoking and physical activity. This statistical technique aims to minimise the effect these external factors have on the association of interest, in this case between fried food and heart disease.
What were the basic results?
An average of 138g of fried food was consumed daily, including 14g of oil used for frying. About 7% of the total amount of food consumed was fried. Of the total amount of fried food consumed, 24% (34g/day) was fish, 22% (31g/day) meat, 21% (30g/day) potatoes, and 11% (15g/day) eggs. Almost two thirds (62%) of participants used olive oil for frying, the rest used sunflower oil or other vegetable oils.
During the 11-year follow up, 606 definite cases of heart disease were recorded (466 heart attacks and 140 episodes of angina requiring surgery) and 1,135 deaths from all causes.
Fried food consumption was not associated with the risk of coronary heart disease after adjusting for possible confounders. There was also no association between fried fat consumption and death from all causes. These findings were no different in those who fried using olive oil compared to sunflower oil or other vegetable oils.
In the study, coronary heart disease events were defined as definite, probable and possible depending how they were assessed. The researchers first analysed their data using only the definite cases of coronary heart disease and found no association between fried food and these cases. The same result was found when they combined the definite coronary heart disease cases with those that were less certain (probable and possible groups).
How did the researchers interpret the results?
The researchers concluded that within the EPIC-Spanish cohort they found ‘no association between consumption of fried food and risk of coronary heart disease or all-cause mortality.’
They point out that their results are ‘directly applicable only to Mediterranean countries with frying methods similar to those in Spain’ and state that oil (mainly olive and sunflower) rather than solid fat is used for frying in Spain. They also point out that the majority of fried food eaten in Spain is not necessarily fast food.
They also make the important point that ‘frying with other types of fats, reusing oils several times, or consuming fried snacks high in salt may still be harmful’.
This study found no association between how often people ate fried food and their risk of coronary heart disease or death from any cause in a large Spanish cohort.
This study has strengths, including using a valid method of assessing diet, a large sample size and long follow-up time but also has significant limitations. The following limitations should be considered when interpreting the findings of this study:
- The study looked at frying using olive oil or sunflower oil in the context of a Mediterranean diet. The authors make the important point that frying with other types of fats or reusing oils several times may still be harmful. Reusing oils is common in fast food preparation in the UK, and so this study does not show that consuming this type of food is not linked to heart disease.
- Diet was measured only when the participants were first enrolled. Any changes in diet after this will be missed and could lead to an association being masked because of error in the classification of fried food consumption.
- Fried food consumption was self-reported using a computerised questionnaire and so there may have been under-reporting given that many people perceive it to be unhealthy and may want to partially conceal the amount they eat. This could potentially hide an association.
- The study did not assess the extent to which the oils were reused, assuming that this was not common in foods consumed in the home. Hence, the effects of foods fried in reused oils are still unknown and may warrant further research as reusing oils is known to make them more harmful, the authors’ report.
- This study cannot separate the effect of the food from the cooking method. For example, the beneficial effect of omega 3 fatty acids from fish compared with the effect of frying this fish in oil that may be harmful. Therefore, the food itself blurs the link between the cooking method and disease. It is not clear whether the same results would be found if the study was conducted on the UK diet, which is generally considered less health than the Mediterranean diet.
- This study cannot say that consuming fried foods at a higher level than their highest consumption group (249.6g/day) does not increase risk of heart disease.
Bearing in mind the limitations, this study showed that there may not be a link between consuming foods fried in olive oil or vegetable oil and coronary heart disease in Spanish adults consuming a typical Mediterranean diet. The relevance of this to the UK is limited due to differences in diet and the type of fried foods consumed in the two countries.
It is important to note that frying foods with other types of fats, particularly saturated fats such as lard or butter, reusing oils several times, or consuming fried snacks high in salt may still be harmful.
- Guallar-Castillón P, Rodríguez-Artalejo F, Lopez-Garcia E at al. Consumption of fried foods and risk of coronary heart disease: Spanish cohort of the European Prospective Investigation into Cancer and Nutrition study. BMJ 2012;344:e363