Sausages: the absolute truth

The World Health Organization's International Agency for Research on Cancer (WHO-IARC) have just released a statement about red and processed meats ( http://www.iarc.fr/en/media-centre/pr/2015/pdfs/pr240_E.pdf ).  They reviewed the evidence about the increased risk of things like bowel cancer, and concluded that processed meats definitely cause cancer in people, and red meat "probably" causes cancer.

We've known for a long while that the best way to report health risks to the public is to include both relative and absolute risks.  So, in one study, they divided people up into how much processed meat they eat, and looked at the prevalence of bowel cancer, and saw there was a 18% increased risk of bowel cancer among those who eat a 'dose' of processed meat (about a sausage every day; a second sausage increases the risk by another 18%).  Nearly every newspaper reported that relative risk. 

But 18% is from what to what?  The absolute increase in that study was from 56/1000 to 66/1000.  I think only one newspaper reported that: take a bow, The Independent.

****************************************
****************oooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
****************************************
**************************oooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
Lowest group: lifetime risk of bowel cancer = 56/1000 Sausage-a-day group: lifetime risk of bowel cancer = 66/1000

http://www.independent.co.uk/life-style/health-and-families/health-news/cancer-definitely-caused-by-processed-meat-what-you-need-to-know-a6709266.html

Several newspapers also made much of the decision to place processed meat is the same IARC class as smoking (hang your head, Guardian, who declared "Processed meats rank alongside smoking as cancer causes" http://www.theguardian.com/society/2015/oct/26/bacon-ham-sausages-processed-meats-cancer-risk-smoking-says-who ). The IARC class 1 includes everything for which there is sufficient evidence that it 'causes cancer in humans', while 2A 'probably causes cancer' (red meat).  It doesn't tell you anything about the size of any risks, or what sort of cancer you are at risk of.  Let's look at an example.  The increased risk of lung cancer from smoking in one study of Canadian men went from about 1/1000 among non-smokers to 17/1000 – say, 15 times as much, or up to a 1500% increase. 

*ooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
*****************ooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
oooooooooooooooooooooooooooooooooooooooo
Non-smokers: lifetime risk of lung cancer = 1/1000 Smokers: lifetime risk of lung cancer = 17/1000

Some other estimates give the relative increase as up to 30 times. Moreover, nearly 60% of people with bowel cancer are still alive five years after diagnosis; for lung cancer, it's just 5%.  Smoking, because of these effects, is such a hugely different size of risk that it's a big confounding factor in any study that looks at cancer (if, for example, people who eat more sausages and bacon are also more likely to smoke), and is rather difficult to cater for.  So, to coin a phrase, smoking tobacco is still in a class of its own.

The trouble with reporting absolute risks is that it puts everything in perspective, and makes for undramatic headlines  -- not so much "Processed meats rank alongside smoking as cancer causes" as "Eating a sausage a day causes small increase in risk of unlikely event which probably won't kill you, and boy is that different to smoking cigarettes". 

But if you don't see the absolute risks, then the journalist isn't doing their job properly.