I have to say, by the end of pregnancy I was actually quite happy to stay pregnant for a bit longer – I enjoyed feeling my baby wriggle and I liked the exclusion zone around me – cars would stop a good twenty feet away if I crossed the road, pedestrians would lurch out of the way if they came within five feet. I also strongly (and correctly) suspected that my baby was a lot easier to look after on the inside than on the outside. Despite this relaxed attitude to staying pregnant, my due date was still a big deal – I desperately wanted to know when my baby would arrive so that I could plan my maternity leave. I wanted a couple of weeks off before my baby arrived, I wanted to make an appointment to get my hair cut. I didn’t want to go into labour at the hairdresser, I didn’t want to sit around for a month feeling uncomfortable and waiting for an overdue baby.
Due dates are, as we all know, just an estimate of when the baby might arrive. A frequently-quoted statistic is that only 5% of babies arrive on their due date. But I wanted to see a graph of the likelihood of my baby being born on any given date. You’d think this would be easy to find, right? I mean, it doesn’t even require a complicated study, it just requires hospitals (or one hospital) to keep a tally of when babies arrive for a bit. And publish it. The 5% figure must come from somewhere – the evidence must be out there. Given that ‘when will my baby arrive?’ is a pretty big question for most pregnant women, why are data on the subject so hard to come by? Infuriating. I spent an alarming amount of time trying to derive the probability distribution from the bits of information that are provided – the likelihood of a baby being born prematurely, the likelihood of needing an induction, the number of babies born within x days of their due date, etc. Mostly, I was trying to figure out what made it a ‘due date’ – presumably it’s some sort of average – but is it the mean, median, mode? What shape distribution? It can’t be normally distributed because of inductions at 42 weeks. But is it normally-distributed with a cut off? Skewed? Poisson distributed?
Fortunately, this rant has a happy ending, because someone else must have had this rant before me and did something about it. After lots of searching (made more difficult by the sheer number of women asking this very question on forums) I finally found this site: http://spacefem.com/pregnant/due.php. Outstanding site. Calculator for the probability of going into labour on a given day, graphs of the distribution of arrival dates, and, best of all, links to real studies with real data. Brilliant.
Evidence-based rants 