How is a pregnancy due date worked out?

Forty weeks is counted from before conception

The most confusing thing about pregnancy dating is that the clock starts before the pregnancy does, and almost nothing explains this clearly.

Gestational age is measured from the first day of the last menstrual period, not from conception. Since ovulation occurs roughly two weeks into a cycle, and conception within a day or so of ovulation, a pregnancy described as four weeks along involves an embryo that is about two weeks old. Weeks one and two of a forty-week pregnancy contain no pregnancy at all.

This looks absurd and has a practical justification. The date of the last period is something most people can report, often to the day. The date of conception is not observable without tracking that most people are not doing, and even people trying to conceive usually know the fertile window rather than the exact day. Dating from an observable event is more reliable than dating from an inferred one, so obstetrics adopted the convention and kept it.

The consequence is that actual embryonic development runs about two weeks behind the stated gestational age. Where a source refers to fetal age or conceptional age, it is using the other convention and its week numbers are two lower for the same point. Mixing the two is a common source of confusion when comparing information across sources.

The forty-week figure follows from the same convention: 280 days from the last period, which is roughly 266 days of actual gestation. A pregnancy is considered full term from thirty-seven weeks, and early term, full term, late term and post term are now distinguished — the shift away from treating everything from thirty-seven to forty-two weeks as equivalent came from evidence that outcomes differ measurably across that range.

What Naegele's rule assumes

The calculation itself is old and simple. Naegele's rule, attributed to the German obstetrician Franz Naegele in the early nineteenth century, takes the first day of the last period, adds a year, subtracts three months, and adds seven days. That is 280 days, expressed in a form that could be done on a calendar without arithmetic.

The rule rests on two assumptions, and both are frequently wrong for an individual. It assumes a cycle of exactly twenty-eight days, and it assumes ovulation on day fourteen.

Cycle length varies substantially. Large studies of tracked cycles find a mean nearer twenty-nine days, with wide variation between people and meaningful variation within the same person from month to month. A cycle consistently longer than twenty-eight days shifts ovulation later, which makes the true due date later than the rule produces — roughly a day for each day of cycle length beyond twenty-eight. A thirty-five-day cycle puts the estimate about a week early.

Ovulation timing varies even within a standard-length cycle. Research tracking ovulation directly finds that the luteal phase — from ovulation to the next period — is the more consistent part at around fourteen days, while the follicular phase before ovulation is what varies. That makes counting back from the next expected period a better predictor than counting forward from the last one, which is the logic ovulation calculators use.

Some calculators let you enter a cycle length and adjust accordingly. That is a genuine improvement on the flat rule for anyone whose cycle is consistently long or short, though it still assumes regularity. For an irregular cycle, no calculation from a period date is reliable.

How accurate the date turns out to be

The estimated date of delivery is a central estimate of a fairly wide distribution, and it is routinely treated as a deadline instead.

Only around four percent of births occur on the estimated date. Roughly half occur within a week either side, and the large majority within two weeks. Spontaneous labour clusters slightly after the estimated date for first pregnancies, which is why a first baby arriving a few days late is unremarkable rather than a complication.

A first-trimester ultrasound is substantially more accurate than the period-based calculation, and this is the part worth knowing. Early embryos grow at a highly consistent rate, so a crown-rump length measurement between about eight and thirteen weeks dates a pregnancy to within a few days. Current obstetric guidance is to redate the pregnancy when an early scan disagrees with the period-based estimate by more than about five to seven days, and the scan date then becomes the official one.

Accuracy falls as pregnancy progresses, because growth rates diverge between individuals. A scan in the third trimester can be out by a fortnight or more in either direction, which is why a late scan does not override an early one. If an early scan was done, its date is the best available and should not be revised by later measurements.

None of this makes the estimated date unimportant. It anchors screening windows, decisions about when intervention is appropriate, and the definition of preterm and post term — all of which depend on knowing gestational age. It is a planning instrument with a known margin, not a prediction of a day.

Counting the practical dates around it

Several dates hang off the due date and each has its own counting rules, which is where a general date calculator earns its place.

Trimesters are conventionally weeks one to thirteen, fourteen to twenty-seven, and twenty-eight onward, though the exact boundaries vary slightly between sources — a difference that matters only when something is scheduled relative to them.

Leave entitlements are counted in a mixture of units and it is worth checking which. Some are expressed in calendar weeks from the birth, some in working days, and some from the due date rather than the actual date of birth, which differ. Notice periods to an employer are typically counted in weeks before the expected week of childbirth, so they depend on the estimated date rather than the real one.

Age after birth is the last of these and has its own convention. Paediatric guidance frequently uses corrected age for a baby born preterm — chronological age minus the weeks of prematurity — when assessing development, which for a baby born two months early means development is compared against a point two months behind the calendar. Growth charts and milestone timelines generally assume corrected age through the first couple of years.

Questions