OK, so let's see.
The number of biweekly deaths in the agegroup <65 is clearly elevated
in El Paso in weeks 14-15 with 83 while the average in weeks 1-13 is 59.
Even when we consider that reporting habits change from week to week
and just consider the percentage, it's still elevated with 33.3% while
the average was 28.0%.
This is not so much elevated as the P+I deaths but still should
only happen ~once per thousand biweeks statistically,
when we assume that all deaths are independent
and there is no outbreak of some unusual disease or disaster or such.
For comparison, if P+I deaths were independent (which they aren't, it's infectious)
then I calculate a 1:17million chance of seeing 43 or more
in a 2-weeks period when the average is 12.
So there is indeed reason to assume that the increase in P+I deaths
and the increase in youngerly(<65) deaths is correlated.
I think that's what niman meant, but how he formulated it,
that was obviously wrong.
The percentage of weekly deaths occurring in the youngerly(<65) in El Paso
in weeks 1-16 , 2011 was:
25.6,21.3,23.6,39.6,28.4,26.3,22.2,32.2,26.4,29.9,24.3,32.4,27.3,30.5,35.9,34.6
for week 16 this looks even more significant than the 10.3% ratio of P+I deaths.
This is consistent with H1N1 deaths, since the H1N1-deaths-age is lower than
the normal flu death-age. (but no H1N1 reported this season in region 6)
But it is also consistent with deaths from accidents or violence or disaster like fire.
Anyway, it still supports that something unusual is going on and this is not
mere coincidence.
we should check this with other cities or regions,
eg. Houston did look unusual to me, but I didn't check the age-distribution yet.
El PasoHouston
Login
Register
FAQ
Search
