Counting down to the birth

By Michael Blastland
Producer, BBC Radio 4's More or Less

In Britain women are given a due date of 280 days

Could a surfeit of overdue babies be all in the counting? Producer, Michael Blastland weighs up the odds.

How many medical judgements aren't really judgements at all but depend on a simple measurement, a number?

I ask because of this week's surprising story from BBC Radio 4's More or Less that the due date for new babies when calculated from the date of the last period seems not to be based on the research evidence.

Discrepancy

The mean date for delivery (the average) is 281 days after the first day of a woman's last period. But that average includes premature births.

In France, it seems, the due date is given as 287 days - a time when some British women will already find themselves being induced.

The most typical date for birth, the day on which more women give birth than any other, is 283 days (the mode).

Yet in Britain women are given a due date of 280 days and offered induction - sometimes with a good deal of encouragement - after 287 days.

We asked several doctors and experts on obstetrics for an explanation for this discrepancy but didn't get a convincing one. The answer seemed to be that 280 days had frozen into a kind of medical certainty over the centuries from comments by a Dutch doctor in the 1700s.

If anyone knows why British women are often told that their baby is due three days earlier than is most likely, do let us know. Click here to have your say.

In France, it seems, the due date is given as 287 days - a time when some British women will already find themselves being induced. If anyone can tell us why the French choose this number, we would be just as interested.

Peculiarity

Numbers have a habit of achieving a kind of reverence. On More or Less, we love numbers and we're not keen to diminish their standing.

Good numbers can be powerful ways of understanding what's going on in the world. But not all numbers are good. Finding which you can trust and which you can't is often difficult.

There are, after all, quite a lot of babies and finding the typical length of pregnancy can be done.

When we asked, as we often do on More or Less, "is that a good number?" we didn't expect to find quite such vagueness about pregnancy.

There are, after all, quite a lot of babies and finding the typical length of pregnancy can be done - indeed has been done - with a survey of 400,000 women in Sweden in 1990. So the real peculiarity here seems to be that there are reasonably good numbers around, we just ignore them.

But back to my original question. How often does medicine use some kind of quantitative indicator to reach diagnosis; blood counts, for example, to decide if a patient has leukaemia?

Not knowing much about medicine, I can't tell you. But the thought won't leave us that there may be a good deal of reverence for certain numbers out there which avoid the need for making more difficult judgements, and might benefit from some questioning.

If anyone has any thoughts, do let us know.

---

BBC Radio 4's More or Less was broadcast on Thursday, 8 January, 2004, at 1500 GMT.

---

You can e-mail us by using the form below: