I came across a convincing critique
of the recent headlines
claiming that eating meat greatly increases your chance of dying. The discussion is fairly detailed; the critic pretty clearly doesn't believe that the conclusions reported can be consistent with the data, given what appear to be internal inconsistencies: A larger effect after controlling for other factors then before, despite the fact that the other factors are mostly getting worse as the amount of meat eaten increases.
But the real killer fact is much simpler. The studies reported on covered two groups of people over a period of more than twenty years. Total mortality from all causes was less than one percent. So what is being reported as a 13% increase in mortality due to eating one more serving of meat is, assuming everything in the study is correct, a 13% increase in a mortality rate of less than one percent, or an additional mortality of about one in a thousand over twenty plus years.
I also found another piece
offering a more general criticism of results of this sort—claims about the effect of nutrition based on studies of what some large group of people ate and what happened to them. Such results are often newsworthy but, at least according to the author of that piece, always or almost always fail to be supported by followup studies.
Looking at the reports more carefully, I think the critique is wrong in at least one important respect. The death rate is not less than 1% for the whole study period, it is less than 1% per year. The total death rate over the period appears to be about 20%, which greatly weakens the force of the argument I cited.