The evaluation on the carcinogenicity of red meat by the International Agency for Research on Cancer (IARC) – and not by WHO as many media reported – has given rise to a range of misunderstandings and is an example of how scientific communication on topics with a major public impact may prove extremely difficult.  As coordinator of the epidemiology Working Group that classified the consumption of red meat, I followed the process from the beginning, about a year ago, and I may be able to shed some light on it.  All the information was included in the IARC press release but apparently journalists do not always read releases from top to bottom.

1. Many confused the ‘strength of evidence’ with the likelihood of developing cancer as the result of eating red meat. Processed meats – basically packaged meats containing nitrites – have been classified as group 1 (carcinogenic to humans). This was done on the basis of strong evidence a group of international experts deemed sufficient, following a careful evaluation of 800 scientific publications that started a year ago.  It should be noted that the evaluation was carried out by a group of independent experts who followed IARC consolidated and well tested methodologies: the evaluation is neither a WHO nor an IARC decision but the result of the work of the group of independent scientists convened by IARC.
2. As for individual risk, eating more than 50 g of processed red meat a day raises the risk of developing colon cancer in a lifetime to about 6 percent as compared to 5 percent in those who eat little or no processed red meat. A very modest increase if compared say to the risk caused by smoking which increases the risk of developing lung cancer 25-fold (1 percent in the life of non-smokers as compared to 25 percent in the heavy smokers).  Most people mistook the strength of evidence for the strength of an increased risk.  Another misunderstanding is the belief that there is an ideal threshold below which there are no risks: there is a direct relationship between amounts eaten and the risk of cancer, which is the more you eat the greater the risk.  For those who wish to have further information, all this is clearly explained in the website of Cancer Research UK.
3. On the basis of the above it is meaningless to say that meat consumption is as carcinogenic as Plutonium. They are in the same group, Group 1, because using standard criteria the working group convened by IARC concluded that evidence for each of these exposures was strong.  In the case of meat, already in 1997 the World Cancer Research Fund and later the American Cancer Society and Cancer Research UK had already reached those conclusions.  What the IARC Working Group did was to carry out the most rigorous assessment so far, as the enormous amount of work proves.  All the institutions and agencies I mentioned stop at the point in which they conclude that there is strong evidence and a relationship between the quantity consumed and risk.  However, they do not issue nutritional guidelines concerning the nutritional value of meat as a whole.
4. I believe it is impossible to appreciate the implications of the above unless one has a basic understanding of what cancer is and of how it develops.  Cancer is a multistage and multi-factorial disease given it develops through stages.  In any population there are some individuals who may be especially predisposed to developing cancer for a number of reasons – inherited genetic variations, mutations induced by other carcinogenic substances and so on.  In other words, a small number of people in any given population may lack the activation of just the last stage or hit to complete the carcinogenic process - even at low exposure doses.  A larger number of individuals may lack the activation of two stages and so on.  Luckier individuals have no activated stage, but they are probably a lucky few.  How many stages are required?  We don’t know exactly but it could be anywhere between 5 and 6 – this is in fact more of an educated guess.  In this respect, dose and risk are related (dose definitely counts); however, the comparisons like the one some people have made with botulin or other forms of acute toxicity do not hold, because these are ‘one hit’ toxicants (the activation of one stage is sufficient) and a notion of threshold applies to them.  Multi-factorial diseases require the combination of several exposures, which increase risk.  For instance, in the case of heart attacks, obese hypertensive males with high cholesterol and glycaemia who are strong smokers experience the highest risk.  If no stage has previously been activated, then even high doses of nitrosamines present in processed meats (sausages, cured meats and the like) may have no effect in a single shot.  However, it is possible that they lead to mutations that in combination with those due to other exposures will eventually lead to the development of colon cancer.  In this case, probability is a notion that implies the biological model typical of cancer.  High doses of arsenic inactivate enzymes that are essential for cell life: a high dose is necessary to generate an intoxication and there is a threshold; in fact, we are all exposed to very low doses of arsenic without suffering any signs of acute intoxication.  Even drinking 10 litres of water all in one go may prove fatal.  The development of cancer, or carcinogenesis as it is known, works in a different way and there are real and complex issues behind the understanding of carcinogenesis; it is not, as the press would like us to believe, the whim of a scientist or of an international agency.
5. The last point is an answer to a recent article by Guyatt, which was published in the Financial Times – and one wonders why this journal was chosen to write about public health. The writer suggests we failed to consider two Randomised Controlled Trials (RCTs) in our assessment.  RCTs are considered the most reliable scientific procedure to establish a causal relationship and for instance they are used with drug trials.  In fact, we did consider them in the Working Group but then disregarded them for a number of very valid reasons: one was called Women’s Health Initiative, and focussed on complex nutritional styles and not just on the reduction of meat consumption.  It is therefore impossible to tell the difference between the effect the reduction of meat had from the impact of other features of the diets the women underwent. The other publication was based on a Polyp Prevention Trial that led to a 52 percent reduction of colon cancer risk among women following the recommendations of the American Cancer Society, which include lower meat consumption.  The aim of this second RTC was to prevent the recurrence of polyps in high risk people.  In this case the meat reduction was so small – a few grams a day of processed meat – that the impact cannot be measured as such, independently of other dietary changes.