Mediterranean diet 'cuts cancer risk'

Wednesday July 2 2008

A study has found that “eating a Mediterranean diet rich in fruit, vegetables, and fish can substantially cut the risk of developing cancer”, The Daily Telegraph reported. It said that the study followed more than 26,000 Greek people over eight years, and found that changing a minimum of two aspects of a person’s diet, such as using olive oil instead of butter, could reduce their overall risk of developing cancer by 12%.

This large study aimed to demonstrate the relationships between the intake of certain food items and cancer incidence. Although the results indicate a relationship between cancer incidence and eating more of the components of the diet, there are several aspects to the study that limit confidence in its findings. For example, how beneficial the diet is to the risk of cancer is subject to how the data is examined. Also, some of the newspaper reports may have over-emphasised the relationship between certain individual components of the Mediterranean diet and cancer. The only individual diet component found to reduce the risk of cancer incidence was the higher consumption of monosaturated fat than saturated fat.

This research contributes to the increasing evidence that a Mediterranean pattern or style of diet can reduce cancer risk. However, future studies will need to provide more evidence before this is conclusive.

Where did the story come from?

The research was carried out by V. Benetou and colleagues from the Department of Hygiene, University of Athens Medical School, Greece; the International Agency for Research on Cancer, Lyon, France; Hellenic Health Foundation, Athens, Greece; and Harvard School of Public Health, US. The study was funded by the Europe against Cancer Program of the European Commission, the Greek Ministries of Health and Education, and a grant to the Hellenic Health Foundation by the Stavros Niarchos Foundation. The study was published in the peer-reviewed British Journal of Cancer .

What kind of scientific study was this?

In this cohort study, the researchers aimed to investigate the relationship between diet and cancer incidence. The Mediterranean diet has previously been associated with improved cardiovascular health and overall mortality.

Between 1994 and 1999, 28,572 people from throughout Greece were entered into the European Prospective Investigation into Cancer and nutrition (EPIC). This was a large cohort study across 10 countries, and investigated the role of biological, lifestyle and environmental influences on cancer and chronic diseases.

A food questionnaire, covering 150 types of food and drink, was administered by researchers in person to assess the participants’ food intake over the past year before they enrolled in the study. The frequency with which foods were consumed and the size of portions were assessed with the aid of photographs to give estimated quantities in number of grams per day. From this, the researchers calculated nutrient and total energy intakes for different food groups such as vegetables, fruit, nuts, dairy, etc.

The researchers also used a 10-point scale to assess the participants’ adherence to the traditional Mediterranean diet. For components of the diet thought to be beneficial, such as vegetables, legumes (peas etc.), fruit, nuts, cereals and fish, the participants scored zero if they consumed less than the average needed for any presumed benefit, and one point if they consumed above the average. Opposite scores were given for the presumed detrimental components of the diet, such as dairy and meat.

The same point scoring was used for the components that had an “ideal range”. Here, the participants scored one point for a daily alcohol intake of between 10-50g for men and 5-25g for women. Fat intake was scored by allotting one point to individuals who had an above average ratio of monosaturated fat (olive oil) to saturated (animal fat) for the group. Final scores ranged from zero (minimal adherence) to nine (maximum adherence). The participants also had their activity levels assessed and calculated as metabolic equivalent hours per day, and body measurements such as BMI were taken.

The researchers excluded people who already had cancer when they enrolled, people who were missing data for analysis, and those for whom there was no follow-up information available. This left 25,623 of the original Greek cohort (10,582 men and 15,041 women) who were followed for an average of 7.9 years. Trained professionals monitored the participants or their next of kin through telephone interviews, and each reported cancer was verified using pathology reports, medical records, discharge diagnoses or death certificates. Cancers were classified using the International Classification of Diseases for Oncology.

The researchers calculated the time between cancer being first diagnosed to the person’s death from cancer, and looked at associations with variations in their diet. The researchers took into account the age, education, BMI, smoking, alcohol, physical activity and supplement-use of the participants in their analysis. They also controlled for dietary components that were not included in the Mediterranean diet score, such as potatoes, eggs, confectionary and non-alcoholic drinks.

What were the results of the study?

From the 25,623 participants, there were 851 new diagnoses of cancer during the follow-up period, 110 of which died of cancer without a date of diagnosis being given. In general, those with the higher Mediterranean diet scores (better adherence) tended to be younger, with a better level of education, and more physically active. For those who developed cancer, lung was the most common in men, followed by prostate, large bowel and stomach. In women, breast cancer was the most common, followed by large bowel, ovary and uterus.

After taking into account other confounding factors, having a higher ratio of monosaturated to saturated fats was found to significantly decrease the risk of new cancer by 9%. There were no other significant associations with any other food group including vegetables, legumes, fruit, nuts, dairy, meat or fish. When the researchers looked at Mediterranean diet score, they found that compared to having a score of 0-3, having a score of 4-5 reduced the risk of new cancer by 14%, and having a score of 6-9 by 22%. Overall, each two-point increment in score was calculated to reduce risk by 12%. When the researchers separated the group by sex, only women had a significant reduction in risk with an increased Mediterranean diet score. Separating the group to smoking-related and smoking-unrelated cancers only demonstrated significant reduction in risk with increased Mediterranean diet score for smoking-unrelated cancers.

What interpretations did the researchers draw from these results?

The researchers concluded they have found evidence “that adherence to the traditional Mediterranean diet is associated with markedly and significantly reduced incidence of overall cancer, which is appreciably larger than predicted from examining individual [diet] components”.

What does the NHS Knowledge Service make of this study?

One of the strengths of this cohort study is the large sample of people who participated in it. Although the study appears to demonstrate a relationship between cancer incidence and eating an increased number of components of the traditional Mediterranean diet, there are several aspects of the study that should be considered when interpreting its results:

  • The participants’ food intake was assessed at only one time at the beginning of the study when they were asked to recall their diet over the previous year. Although a thorough food questionnaire was used, their responses are subject to recall bias and inaccuracies in their reports of the frequency and quantity of different foods eaten. Foods eaten over the past year may also not be representative of a lifetime’s diet prior to this period, or following it over the course of the study.
  • The relationship between certain components of the Mediterranean diet and cancer may have been slightly over-emphasised by the newspaper reports; the only individual diet component found to reduce the risk of cancer incidence was the higher consumption of monosaturated fats than saturated fat. A higher Mediterranean diet score did reduce the incidence of cancer overall, but the scoring system would have been subject to some inaccuracy by grouping people into the broad categories of ‘below’ or ‘at or above average’ consumption of certain food items. It is unclear how the average (median) consumption values calculated for this group relate to average values in non-Mediterranean countries.
  • The follow-up period of an average of 7.9 years is relatively short, and follow-up over a longer period during which there would be a greater number of diagnosed cases of cancer may change this relationship to Mediterranean diet.
  • As the study only examined the relationships between food and overall cancer, no conclusions can be made about the dietary effects on any particular type of cancer.
  • As the authors acknowledged, date of death was substituted for diagnosis of cancer in 12% of the cancer cases, meaning that any estimates of death rates at a particular time point may include some inaccuracies.
  • Although other physical, social and lifestyle factors have been considered and taken into account in the analyses, it is not clear whether these efforts have been sufficient, or whether other medical or genetic risk factors for cancer could be contributing to the difference between the groups.
  • As the population studied were all inhabitants of Greece, the same relationships between diet and cancer may not be seen if the same diet were consumed by other countries. Ethnic, cultural and environmental variations may all have effects upon cancer incidence.

The numerous benefits to health of a diet including high quantities of fruit, vegetables, nuts, oils and fish with lower levels of saturated fats, meat and dairy are well known. This research contributes to the increasing evidence that such a Mediterranean pattern or style of diet can reduce the risk of cancer. However, future studies will need to provide more evidence before this is conclusive.

Sir Muir Gray adds...

Pass the olive oil please.

Analysis by Bazian
Edited by NHS Choices