The big health news of the week is the claim that eating a diet high in processed meat can increase the risk of premature death due to cancer and heart disease.
The current media scare stems from a large Europe-wide study looking at diet and mortality, involving just under half a million people who were followed for an average of 12.7 years.
One of the main findings was that people in the study who ate the most processed meat (160g or more per day) had a 44% increased risk of dying during follow-up compared to those who ate the least (20g or less).
The link to red meat was less conclusive.
The researchers estimated that if we all ate less than 20g of processed meat (which is around a single small piece of bacon) a day, then 3.3% of all deaths could be avoided – which is where the media reports came from that processed meat is responsible for 1 in 30 deaths.
However, an important limitation (rightly highlighted by the authors) is the possibility that other health and lifestyle factors could be contributing towards premature death risks.
Nevertheless, the study does highlight the importance of eating a healthy balanced diet, containing a high amount of fruit and vegetables.
Where did the story come from?
The study was carried out by researchers from the Institute of Social and Preventive Medicine, University of Zurich, Switzerland, and a large number of other institutions across Europe.
Financial support was provided by a range of European organisations including, government, charity and academic institutions.
The media stories are generally representative of the findings of this research, with most including the common sense advice that eating an occasional bacon sarnie won’t kill you – just don’t do it every day.
The claims that processed meat is responsible for 1 in 30 deaths are based on the researchers’ estimate that 3.3% of the deaths in this study could have been prevented if all those who took part in the study ate less than 20g of processed meat a day.
What kind of research was this?
Many past observational studies have suggested that high levels of red meat and processed meat consumption could be linked to a range of diseases, including cardiovascular diseases and various cancers, such as bowel cancer.
However, it can be difficult in such studies to exclude the possibility that the effect is not directly due to red and processed meats as such, but is due to the influence of other health and lifestyle factors. For example, people who eat a small amount of red and processed meat may also be eating higher amounts of fruit and vegetables, exercising more, be less likely to be overweight, smoke, or drink excess amounts of alcohol.
In the same vein, people who eat lots of processed meat may have other unhealthy habits such as drinking lots of alcohol and being heavy smokers.
This was a large cohort study using data collected as part of the European Prospective Investigation into Cancer and Nutrition (EPIC) study.
EPIC is an ongoing cohort study including more than 500,000 participants from 10 European countries.
The researchers took data from the EPIC study to look at the association between red meat, processed meat, and poultry meat consumption, and the risk of overall mortality and cause-specific mortality.
What did the research involve?
Men (aged 40 to 70) and women (aged 35 to 70) were recruited to EPIC between 1992 and 2000 (depending on the European study centre). After excluding those with self-reported cancer or heart disease, or those who did not report on smoking status at the time of enrolment, there were 448,568 people in the study.
Dietary assessment was performed slightly differently depending on the country:
- seven countries gave self-administered dietary questionnaires (including data on 300-350 food items)
- three countries administered a similar questionnaire by direct interview
- two of the countries (UK and Sweden) also combined the questionnaires with a seven-day food diary
For the purposes of analysis, they grouped food products as follows:
- red meat (beef, pork, mutton/lamb, horse, goat)
- processed meat (including ham, bacon, sausages, or a small amount of minced meat as part of a ready-to-eat product – processed meat is mainly taken to be red meat, but it could be white as well)
- white meat (poultry, including chicken, hen, turkey, duck, goose, unclassified poultry, and rabbit)
Various other sociodemographic, health and lifestyle questions were also assessed at recruitment, including age, education, height and weight, medical history, alcohol consumption, and smoking history (current, past or never, including questions on frequency and type of tobacco smoked).
Follow-up of outcomes was to 2005-09, depending on the country, with an average follow-up of 12.7 years. Information on deaths and cause of death was obtained through record linkage with cancer registries, Boards of Health, and death indices in seven countries, and through active follow-up of participants (for example, mail, telephone, and medical records) in three countries.
Information on vital status could be obtained for 98% of the cohort, which is impressive given the size of the study.
Hazard ratios were calculated to examine the association between different types and quantities of meat and processed meat consumption and risk of death.
They adjusted analyses for the following cofounders:
- study centre
- weight and height
- smoking history
- alcohol intake
- overall energy intake
- physical activity levels
- educational level
What were the basic results?
Compared to men and women who ate lower amounts of red and processed meat, those who ate the highest amounts tended to also eat fewer fruit and vegetables, be more likely to smoke, and less likely to have a university degree. Men who ate the highest amounts of red meat also drank more alcohol than those who ate lower amounts. This result was not seen in women.
During the average 12.7 year follow-up, there were 26,344 deaths (6% of the cohort), of these, 37% were due to cancer, 21% due to cardiovascular disease, 4% due to respiratory disease, 3% to digestive tract diseases, and the remainder due to various other causes.
Overall, there was a link between increasing processed meat consumption and risk of all-cause mortality. In the model adjusted for all confounders:
- people who ate the highest amount of processed meat (160g per day) had 44% increased risk of death compared to those who ate 10-20g per day (hazard ratio (HR) 1.44, 95% confidence interval (CI) 1.24 to 1.66)
- people who ate 80-160g a day had 21% increased risk (HR 1.21, 95% CI 1.14 to 1.28) and those who ate 40-80g a day had 9% increased risk (HR 1.09, 95% CI 1.05 to 1.14) compared to those who ate 10-20g per day
- compared to those who ate 10-20g per day, there was no difference in risk with eating 0-10g or eating 10-40g
- overall, eating an additional 50g of processed meat a day gave 18% increased mortality risk (HR 1.18. 95% CI 1.11 to 1.25)
- eating an additional 50g of processed meat a day also gave a 30% increased risk of dying from any cardiovascular disease (HR 1.30, 95% CI 1.17 to 1.45), and an 11% increased risk of dying from any cancer (HR 1.11, 95% CI 1.03 to 1.21)
The link with red meat was not as strong as for processed meat:
- eating the highest intake of red meat (160g per day) was associated with 14% increased risk of all-cause mortality compared to eating 10-20g per day (HR 1.14, 95% CI 1.01 to 1.28)
- people who ate 20-160g of red meat a day were at no higher risk than those who ate 10-20g per day
- people who ate the lowest amount (0-10g a day) also had increased mortality risk compared to those who ate 10-20g per day.
- unlike with processed meat, the researchers found no overall significant increase in mortality risk for eating an additional 50g of red meat a day
There was no link between death risk and poultry consumption.
The researchers estimated that 3.3% of all deaths could be avoided if all people ate less than 20g per day of processed meat.
How did the researchers interpret the results?
The researchers conclude that their analysis supports a ‘moderate association’ between increased processed meat consumption and increased mortality, in particular due to cardiovascular diseases, but also cancer.
This is a useful study examining whether there is an increased risk of dying from any cause, and from specific causes, with increased consumption of red meat and processed meat. The link to red meat was less conclusive, but there appeared to be a consistent link between increasing processed meat consumption and mortality risk.
The study has many strengths, including that it followed a large number of adults from across 10 European countries for an average 12.7 years, with almost complete follow-up.
The study used reliable methods to assess mortality outcomes. Food frequency questionnaires will unavoidably include some inaccuracy (for example, inaccurate recall or estimation of intake).
However, the researchers did attempt to validate the information through a series of 24-hour recalls.
The researchers have adjusted their analyses for age, study centre, weight and height, smoking and alcohol intake, overall energy intake, physical activity levels and educational level.
However, as the authors rightly conclude, the main limitation of the study is that it cannot completely exclude the possibility of residual confounding – that is, that the effects of these demographic, health and lifestyle factors, or others unmeasured, have not been fully accounted for.
These limitations apart, the study provides reasonably good evidence to support the importance of eating a healthy balanced diet, containing a high amount of fruit and vegetables. At the same time, it is important to moderate your consumption of foods high in salt, fat and sugar, which includes many processed foods.
The occasional bacon sandwich or full English breakfast probably won’t do significant damage to your health. But these should be occasional treats and not a staple of your diet.
Analysis by Bazian
Edited by NHS Website
Links to the headlines
BBC News, 7 March 2013
The Guardian, 7 March 2013
Daily Mail, 7 March 2013
The Independent, 7 March 2013
The Daily Telegraph, 7 March 2013
Sky News, 7 March 2013
Links to the science
BMC Medicine. Published online March 7 2013