Originally Posted by Serenyd
It's hard to compare our statistic to U.S. statistics because of many factors.
These were the best statistics I could find:
U.S. 2003 - 6.74 deaths per 1,000 fetal deaths and live births. This includes
Infant deaths of less than 7 days and fetal deaths with stated or presumed period of gestation of 28 weeks or more. This is the perinatal death rate.
This is a death rate of about 0.7%
I will note that a certain blogger is comparing the UC data presented here with the CNM hospital neonatal death rate and coming out saying that the UC death rate is 20 times higher...
I take issue with comparing the UC numbers with CNM numbers on the CDC wonder data set. My biggest reason for taking issue is that in the case of the CNM data, I'd bet my house that there are a good number of neonatal deaths not being attributed to CNMs in that data set that started out as intended CNM births, but at some point transferred to OB care, and are thus reflected in the OB numbers.
How many of the 4 deaths reported here as UC deaths had actually already transferred to OB care prior to the death?
Then there is the issue of that blogger only cherry picking what she considers to be "low risk" CNM births--after 37 weeks, white non-Hispanic, blah, blah, blah.
Is this REALLY reflective of the UC population birthing here? How many of the UCer's here are doing VBACs which many CNMs are not "allowed" to attend? How many are less than 37 weeks? How many have undiagnosed GD, are "too old" for CNMs...or whatever else risk factor you want to throw in there?
As others have pointed out...you are working with a "self selected" data set...so the true results could be skewed a LOT in either the direction of a higher or a lower death rate.
And finally...statistical significance. If there had been only ONE death in the UC group, that blogger would be crowing that the UC death rate is 5 times the CNM death rate...but obviously there is an issue of statistical significance going on.