I am a nutritional researcher. Work for industry but not beef/meat.
I know why these groups claim that X causes Y, but I do not agree with it.
The substantial evidence is based upon observational studies. Observational studies are great for looking at associations, but less so for defining if something causes something else. The propsective cohorts remove some of the time relative aspects of observational studies (a happened before b) but not much else. Sure people love to bring up Bradford-Hills ground breaking work on tobacco and cancer, but the relatively risks here are enormous compared to that seen from food and cancer. This increases the likelyhood of finding a true association. Plus the animal studies are pretty consistent as a means to perform controlled trials of the exposure. The WHO/IARC admit their animal work is inadeqate to define a causal link. So the reported increased risk fo red meat of 17% and 18% for processed is from
http://www.ncbi.nlm.nih.gov/pubmed/21674008
In this they explain the problem, the dataset, and compile by known statistical techniques to improve hte power of the analysis to come up with the values. Nine studies made up the processed meat pool and eight made up the red meat pool.
on average there was a follow up of ~10 years. Food intake was estimated via whats known as a food frequency questionnaire (FFQ) at teh begining of the trials. Two had another assessment some point, but with a different questionnaire asking a different thing. The FFQ is usually asking people how often they eat something, and how big the serving is (with a few twists) and then the servings are then converted into amounts *But not always*. Sometimes they are 'validated' but not always for all foods/nutrients, and even then, validating does not mean accurately correcting for the variables - never mind the method used for validation is also imprecise. In cases the meta-analysis authors couldnt extract grams of food from a trial (ie if it presented servings rather than weight) they converted it into grams (g) using 120 g as a standard portion size for red and processed meat combined and for red meat, and 50 g was assumed as standard portion size for processed meat. Others have used similar values but still adds noise.
Just think of that for an analysis set, that people are meant to take as substantial evidence. One unreliable measure of food intake at the begining, and associating this to 10years of disease risk.
Its vague assocations, with poor quality data.
the relationship may be bigger or smaller, it may be true or false. But with data like this, its wishful thinking if they can assign causality, But that has never stopped do-good public health folk.