The estimated due date is the most quoted number in obstetrics and the least understood. Patients hear a deadline. Clinically it’s a reference point — the thing almost every other decision gets measured against.

Counting from the wrong day, on purpose

Gestational age is counted from the first day of the last menstrual period, not from conception. That’s roughly two weeks before the pregnancy actually began.

It looks like a mistake and isn’t. Conception is rarely observed; the last period usually is. Dating from a date someone can actually report gives a consistent standard, and every reference range — lab values, growth charts, viability thresholds — is built on that same convention. It does mean a woman described as “6 weeks pregnant” has an embryo about 4 weeks old, which is worth explaining when it causes confusion.

Naegele’s rule

The classic arithmetic:

First day of the LMP − 3 months + 7 days + 1 year

An LMP of 1 March gives a due date of 8 December. It assumes a 28-day cycle with ovulation around day 14, and it puts pregnancy at 280 days.

That assumption is the rule’s weak point. For someone with a consistent 35-day cycle, ovulation is later, and the estimate runs about a week early unless it’s adjusted for cycle length.

Ask about the cycle, not just the date

“When was your last period?” is half the question. Cycle length and regularity change the answer, and so does recent hormonal contraception, which can make the last bleed a poor marker of anything. If the cycle isn’t ~28 days and regular, treat an LMP-based date as provisional.

Why ultrasound often wins

Early ultrasound dating is more accurate than LMP dating, and the earlier it happens the more accurate it is. In the first trimester, crown–rump length predicts gestational age within a few days, because early embryos grow at a remarkably consistent rate.

That consistency disappears later. By the third trimester, genuine variation in fetal size means ultrasound dating carries an error of a couple of weeks — which is why a due date established early is not re-dated by a later scan. A big baby at 32 weeks is a big baby, not a mis-dated one.

So in practice: an early scan that disagrees meaningfully with the LMP generally replaces it; once dating is settled, it stays settled.

Why the number does real work

The due date isn’t about the birthday. It’s the denominator for most of the decisions around it:

Viability and the threshold conversations in a threatened preterm delivery
Timing of antenatal steroids, which have a window
Screening and testing windows, which are gestational-age specific
Interpreting fetal growth — small for dates only means something if the dates are right
Induction and post-term management, which run on defined gestational-age limits

The terminology used around it is more precise than most people expect:

Term	Gestational age
Preterm	Before 37 weeks
Early term	37 0/7 – 38 6/7
Full term	39 0/7 – 40 6/7
Late term	41 0/7 – 41 6/7
Post-term	42 0/7 and beyond

Only about 1 in 20 babies arrive on the due date

The EDD is the midpoint of a distribution, not an appointment. Normal, healthy birth spans several weeks either side of it. Framing it as a due month — or at least a due fortnight — spares a lot of anxiety in the final weeks, and it’s a genuinely useful thing to say to a patient counting days.

The bottom line

The due date is an estimate built on a convention, refined by an early scan, and then held steady because later scans measure size rather than age. Get the dating right early and everything downstream is easier to interpret. Treat it as a deadline and it will mislead everyone involved.

Try it

The due date calculator applies Naegele’s rule, adjusts for cycle length, and shows current gestational age and trimester.

Keep going

What happens in the first minutes after all this: the Apgar score.

Sources

References

ACOG — Methods for Estimating the Due Date — the standard for dating and when redating is appropriate
StatPearls — Naegele Rule — the arithmetic and its assumptions