Friday, March 18, 2011

Today's hot health advice: Don't be rash, or saucy!

Well, in the latest issue of the Annals of Hyperprecision, we find the sage advice that to 'reduce' the risk of bowel cancer, you should not eat more than 70g of red meat per day.  That, we're told, means no more than 3 rashers of bacon or 2 measly sausages (but not both!).


Now the study reported more precisely that
Eating 100 to 120g of red and processed meat a day - things like salami, ham and  sausages - increases the risk of developing the condition by 20 to 30%, according to studies.
Although  very depressing for us bacophiles, at first glance this advice has all the panache of scientific soundness, but less so when you think seriously about it. The values seem very precise and while we haven't seen the original paper, we presume this was a regression analysis of cases of colorectal cancer per 100,000  per year vs mean grams of the raw Alley Oop diet per day.  The cutoff of 70g and the vague quoted values are of course subjective judgments.  And these would properly have to be based on some sort of statistical significance test--which entails issues about the size and nature of the sample, measurements, and other assumptions.

And of course 'two sausages' could mean the modest breakfast links shown above, or (preferably) the Cumberlands shown below--the report was  British, after all, and it seems just what you'd expect as public advice from a government committee!


We would not want to undermine public confidence in their government, and certainly not to trivialize the risk of the raw stuff for your intestinal health.   There are plenty of reasons to avoid sausage (without even thinking of the kind of meat that goes into them, nor the environmental cost of meat vs plant food). Colorectal cancers are complex, and dietary intake estimates notoriously imprecise (despite what committee reports may suggest, which is why researchers want to photograph every bite their study subjects eat).  Further, the report discusses whether cutting back in this way on meat consumption will lead to diseases due to iron deficiency.

Damn, this world is complicated!  Reports like this are an occasion to reflect beyond today's lunch, and what would go good with mash if we have to shun the links, and to consider the nature of scientific evidence in the observational setting where data are imprecise and hard to come by, assumptions many, and risk decisions complex.

If all we can realistically care too carefully about is keeping to a sensible level of the inevitable risks we face in life, then perhaps the oldest medical advice is still the best (and relevant to many other areas of the life sciences as well:  moderation in all things.

Well, take heart all!  The latest issue of the Annals also has some compensatory advice: moderate alcohol is good for you!  Just don't mix your liquid lunch with a BLT.

No comments: