This is a follow up from an 7 week old post, where I asked if anyone knew the SD on estimation of human gestation length (ie baby due dates).
Currently the human gestational period is estimated using a nineteenth-century algorithm (Naegels rule) that does not take into account biologic variability - apparently this was based on lunar months. Something I thought was a bit dodgy to be using these days.
Well, anyway I found a sort-of-answer. Bordering on MPSIMS, but hey its a follow-up to a real question.
A paper (AM J OBSTET GYNECOL 1993;168:480-4) was published in 1993 by Mittendorf et al, which statistically analysed a sample of deliveries (n=9355) and came up with an algorithm to more accurately predict delivery date.
The paper details the fudge factors to apply to a base value. eg Multiparity is -3.1 days (95% confidence interval -2.4 to -3.8 days). So taking into account a whole range of factors you can more accurately predict the gestational length with confidence limits.
The equation is basically:
GL = 256.447 + 1.5648age - 0.0259age^2 + 0.1295*ponderal index
Add -3.1169 days for multiparity (not first baby)
Add -2.5328 for race being black
There are other factors for alcohol ( 1.0463 if drinking - not sure of how much from paper) and coffee consumption ( -0.3928 x cups per day), Incompetent cervix ( -18.8054), DES exposure ( -2.6281), Miscarriage (-0.6990), stillbirth (-4.4849) and many more.
This can give due dates up to 3 weeks different from Naegels rule.
I asked several midwives about this. None had heard of the Mittendorf observations. They guessed the reason that the old rule is used is everyone is used to it. I suppose it will take another 200 years and a few follow up studies before the more accure algorithm is used.
Actually, that’s pretty cool. I wonder how more accurate it is. I think people are comfortable with Naegels rule because the bottom line is it doesn’t really matter. Except in very rare cases the baby will come out when it comes out and all these calculations are simply to give the parents some ballpark, but irrelevant figure.
Ah, but it does matter. If you are seeing a traditionally trained OB, he is firmly entrenched in Naegel’s rule. It doesn’t matter that the Harvard study (the first real study of empirical data) found an average gestation for first time (healthy, well-cared for) mothers was 41 weeks plus a day. Or that the average gestation for subsequent births was 40 weeks plus 3 days.
So if, these days, you have a healthy, full term pregnancy, you are quite likely to be induced (and, likely, sectioned when it doesn’t work) when you fail to deliver at 40 weeks.
It does matter also because if the calculation is under by more than 2 weeks you can have a medically induced premature birth. More likely if the doctors underestimate date of conception.
And since we are now in MPSIMS - this didn’t happen with us I am happy to say - all went 100% at the birth.