And as for those dodgy sounding numbers on hypothermia in the USA (hah!), well the procedures for reporting deaths and their causes are clearly laid down in this cheerful little guide:

On line (b) report the disease, injury, or complication, if any, that gave rise
to the immediate cause of death reported on line (a). If this in turn resulted
from a further condition, record that condition on line (c). If this in turn
resulted from a further condition, record that condition on line (d). For as
many conditions as are involved, write the full sequence, one condition per
line, with the most recent condition at the top, and the underlying cause of
death reported on the lowest line used in Part I. If more than four lines are
needed, add additional lines…


http://www.cdc.gov/nchs/data/misc/hb_cod.pdf

btw, don’t you just love that motto and title “Safer-Healthier-People ~ Physicians’ Handbook on Medical Certification of Death” ?!

So in theory all the causative breakdowns going back downstream to “government failure to provide adequate preventative healthcare” could be listed. But is it likely to happen in practice, especially if it tends to throw up uncomfortable political truths? I notice that over 65,000 Americans are listed as dying from ‘flu, for example.

And before you take a view about that last question, consider the answer to this simple query: how many American service personnel have died in combat since the “end” of the Iraq war? If statistics were neutral numbers, there could only be one answer to this – yet there are two in circulation. I know which I believe, and it ain’t Shrub.