Pregnancy Due Date & Gestational Age Calculator

Pregnancy Due Date and Gestational Age Calculator: Four Dating Methods, and the Window Around the Date

Gestational age in completed weeks and days, and an estimated due date, by four separately sourced dating methods — the 280-day rule adjusted for your own cycle length, a known conception or embryo transfer date, a measured crown-rump length through either of two published equations, and a gestational age a scan already reported. The due date is printed with the days on which 37 and 42 weeks fall, because a due date on its own is the middle of a five-week window.

Last updated 1 Oct 2026

These are estimates from dates and measurements, not findings about this pregnancy. A due date is a midpoint with weeks of normal variation either side, and a predicted fertile window is not a reliable way to avoid pregnancy. Nothing here replaces assessment by a midwife or doctor.

Gestational age today, the estimated due date, and the window either side of it

a date and a dating method → gestational age in days, completed weeks and days, trimester, and the due date with the 37-week and 42-week dates beside it
Every method works the same way underneath: it fixes the day on which gestational age was zero, and the due date is 280 days after that day. The four differ only in how that day is found, and they disagree — the page prints the gestational age the method assigned to the date you typed so you can see what it assumed.
The year of the date the method needs: the first day of the last period, the conception or transfer date, or the date of the scan. There is no clock in this calculator, so every date is one you type.
The month of the same date. Months are a list rather than a number because 03/04 means two different days on two sides of the world.
1 to 31. A day that does not exist in the month you chose — 31 April, 29 February in a common year — is refused rather than silently rolled forward into the next month.
The number of days from the first day of one period to the first day of the next. The 280-day rule assumes 28. Every day above 28 moves the due date one day later and makes you one day less pregnant today; every day below 28 moves it earlier. Most calculators ignore this field entirely.
0 for the day of egg collection, fertilisation or natural ovulation — a day-0 embryo. 3 for a day-3 transfer, 5 for a blastocyst, 6 for a day-6 blastocyst. The due date is 266 days after the day-0 event, so it is 266 minus this number of days after the transfer: 261 for a day-5 embryo and 263 for a day-3 embryo, which is the arithmetic ACOG Committee Opinion 700 states.
The measured crown-rump length. Both published equations on this page are printed whichever you choose, because they disagree — by up to 8 days at 5 mm and by a day or less above 25 mm. Above 84 mm crown-rump length stops being the measurement to date on; the page says so rather than refusing.
Robinson & Fleming is the equation behind most national dating tables and most ultrasound machines; INTERGROWTH-21st is newer, prospective, and was fitted at eight sites in eight countries including India. Neither is the right answer; the difference between them is printed so you can see how much the choice is worth.
Completed weeks, so a report of 12 weeks and 3 days means 12 here and 3 below. 12 completed weeks is 84 days, not 85 and not 91.
0 to 6. Weeks and days is a base-7 number: the total is 7 × weeks + days, and the days part never reaches 7.
Normally today. It is an input because the calculator has no clock of its own, which also means you can ask what the gestational age was or will be on any date — the date of a scan, an appointment or a planned delivery.
The month of the date you want the gestational age for.
1 to 31, and it has to exist in the month above. The default pair of dates — a period starting 12 January 2026 and a gestational age asked for 18 June 2026 — is the worked example below.
157days of gestationExample

a period that began on 12 January 2026, a usual cycle of 28 days, method 1, and a gestational age asked for 18 June 2026

Advertisement

One quantity, four ways of finding it, and the two equations behind method 3

Every method fixes J0, the day number on which gestational age was zero.  ·  method 1: J0 = JLMP + (C − 28)  ·  method 2: J0 = Jevent − 14 − aemb  ·  method 3: J0 = Jscan − GACRL  ·  method 4: J0 = Jscan − (7w + d)  ·  then EDD = J0 + 280, GA = Jtoday − J0, completed weeks = ⌊GA/7⌋, days = GA − 7⌊GA/7⌋  ·  Robinson & Fleming 1975: GACRL = 8.052 √(CRL × 1.037) + 23.73  ·  INTERGROWTH-21st 2014: GACRL = 40.9041 + 3.21585 √CRL + 0.348956 CRL, both rounded to the nearest day
J
a Julian day number — the count of days since a fixed epoch, which is how this page does date arithmetic. It is computed from the year, month and day you type by the standard integer formula and converted back the same way; the conversion and its inverse were checked against an independent calendar for all 73,414 dates from 1900 to 2100
J0
the dating zero point: the day on which gestational age is 0 days, 0 weeks and 0 days. For method 1 that is the first day of the last period, shifted by the cycle-length correction. Everything else on the page is this one number plus an offset, which is why the inverse operations round-trip exactly
C
your usual cycle length in days. The 280-day rule assumes 28, and (C − 28) is the whole of the correction. The day-count form used here differs from the calendar form of the same idea, Parikh’s formula, by 0 to 3 days
aemb
the age of the embryo in days at transfer: 0 for egg collection, fertilisation or natural ovulation, 3 for a day-3 transfer, 5 for a blastocyst. The convention puts the day-0 event at gestational age 14 days, so the due date is 266 − aemb days after the transfer — 261 for day 5 and 263 for day 3, the figures ACOG Committee Opinion 700 states
CRL
crown-rump length in mm. Useful for dating to 84 mm, about 14 weeks 0 days. The two equations agree to within a day above 25 mm and come apart below it, reaching 8 days at 5 mm
7w + d
a gestational age a scan reported, as completed weeks times seven plus days. 12 weeks 3 days is 87 days. The days part is 0 to 6 and never 7, and completed weeks means weeks that have finished
280
the convention, not a measurement: 40 × 7 days from the dating zero point, which ACOG states as the definition of the estimated due date. The Early Pregnancy Study’s median was 268 days from ovulation, which is 282 from a notional period under the same 14-day convention — two days more

Worked example

a period that began on 12 January 2026, a usual cycle of 28 days, method 1, and a gestational age asked for 18 June 2026
The dating zero point. 12 January 2026 is Julian day number 2,461,053. The cycle is 28 days, so the correction (C − 28) is zero and J0 is that same day. Gestational age on 18 June 2026, Julian day 2,461,210, is 2,461,210 − 2,461,053 = 157 days. In completed weeks and days that is ⌊157/7⌋ = 22 weeks and 157 − 154 = 3 days. Not 22 weeks 4 days, and not 23 weeks: 157 = 22 × 7 + 3 exactly, and 22 weeks have finished.
The due date. J0 + 280 = 2,461,333, which converts back to 19 October 2026 — printed above as the year 2026, the month 10, the day 19 and the single number 20261019. It is a Monday, which is the 1 in the day-of-the-week row. From 18 June that is 123 days away, and 157/280 of the conventional span has elapsed, which is 56.1%.
The window, which is the point. 37 weeks 0 days is J0 + 259 = 28 September 2026, and 42 weeks 0 days is J0 + 294 = 2 November 2026. That is a 35-day window, and it is the honest answer to “when is the baby due”. Of United States singleton births in 2022, 11.63% were at 37 completed weeks, 17.45% at 38, 38.84% at 39, 18.26% at 40, 4.88% at 41 and 0.27% at 42 or later. If births were spread evenly inside week 40, about 2.6% would land on 19 October itself.
Now change one thing. Set the cycle length to 35 days and nothing about the period changes, but the dating zero point moves 7 days later: the due date becomes 26 October 2026 and the gestational age on 18 June drops to 150 days, 21 weeks and 3 days. A week of difference, from a field most due-date calculators either do not have or do not use. Set it to 21 and the due date moves the other way, to 12 October.
And check the old rule against the new one. Naegele's calendar form on 12 January 2026 — add a year, subtract three months, add seven days — gives 19 October 2026, the same day, so the comparison row above reads 0. Move the period to 1 March 2026 and it reads 2: the calendar form gives 8 December 2026 where 280 days gives 6 December. Move it to 1 May and it reads 3. Across all 1,827 start dates from 2024 to 2028 the calendar form is 0 days later 797 times, 1 day later 519 times, 2 days later 370 times and 3 days later 141 times. It is never earlier, and it agrees with the modern rule on fewer than half of all dates.
And the same pregnancy dated by ultrasound. Switch to method 3, keep the scan date at 12 January 2026 and enter a crown-rump length of 60 mm. Robinson & Fleming gives 8.052 × √(60 × 1.037) + 23.73 = 87.24, which rounds to 87 days — 12 weeks 3 days. INTERGROWTH-21st gives 40.9041 + 3.21585√60 + 0.348956 × 60 = 86.75, which also rounds to 87. The two agree here. Drop the measurement to 10 mm and they give 50 and 55 days: five days apart, on an embryo small enough that both curves are being extrapolated.
What the page refuses. Ask for the gestational age on 18 June 2025, eighteen months before the period: the gestational age is negative and the page returns its incomplete message rather than a negative number. Ask for 18 June 2027: 522 days, outside the 0-to-336-day range, and it refuses again. Enter 29 February 2026, a date that does not exist, and it refuses rather than treating it as 1 March. Each of those is a case where a plausible-looking wrong number would be worse than no number.

What the two crown-rump length equations give for the same measurement

Crown-rump lengthRobinson & Fleming 1975INTERGROWTH-21st 2014Difference
5 mm6 weeks 0 days (42 d)7 weeks 1 day (50 d)8 days
10 mm7 weeks 1 day (50 d)7 weeks 6 days (55 d)5 days
15 mm7 weeks 6 days (55 d)8 weeks 3 days (59 d)4 days
20 mm8 weeks 4 days (60 d)8 weeks 6 days (62 d)2 days
25 mm9 weeks 2 days (65 d)9 weeks 3 days (66 d)1 day
30 mm9 weeks 6 days (69 d)9 weeks 6 days (69 d)none
50 mm11 weeks 5 days (82 d)11 weeks 4 days (81 d)1 day
70 mm13 weeks 1 day (92 d)13 weeks 1 day (92 d)none
84 mm14 weeks 1 day (99 d)14 weeks 2 days (100 d)1 day
95 mm14 weeks 6 days (104 d)15 weeks 0 days (105 d)1 day
Computed from the two published equations and rounded to the nearest day, which is the convention the published dating tables use. The INTERGROWTH-21st fit is published for 15 to 95 mm, so the 5 mm and 10 mm rows are extrapolations of it; 84 mm is the upper limit at which crown-rump length should be used for dating at all. The usable range of the measurement is therefore also the range in which the choice of equation stops mattering.

Where births actually happen, and what the due date is the middle of

Completed weeksShare of US singleton births, 2022Published designation
before 378.7%preterm
3711.63%early term
3817.45%early term
3938.84%full term
4018.26%full term — the week the due date falls in
414.88%late term
42 or later0.27%post-term
Singleton births by the obstetric estimate of gestation, from National Vital Statistics Reports vol 73 no 1 (2024). The 39-week peak is substantially the footprint of scheduled induction and caesarean section rather than of labour starting on its own, so this is a distribution of obstetric practice as much as of pregnancy length. For the natural spread, the better figure is the 37-day span measured in the Early Pregnancy Study, where the day of ovulation was known.

The four methods, and the one thing they all do

MethodWhat it fixesGestational age at the date you typeDue date
1 — last menstrual periodthe first day of the period as gestational age zero, shifted by (cycle − 28)28 − cycle, so 0 at a 28-day cyclethat date + 280 + (cycle − 28)
2 — conception or transferthe day-0 event at gestational age 14 days14 + embryo age in daysthat date + 266 − embryo age
3 — crown-rump lengththe scan date at the gestational age the equation givesthe equation’s output, rounded to a dayscan date + 280 − that figure
4 — reported gestational agethe scan date at the gestational age the report states7 × weeks + daysscan date + 280 − that figure
Every method reduces to one quantity: the Julian day number on which gestational age was zero. The due date is 280 days after it and today’s gestational age is the number of days from it. That is why the inverse operations round-trip exactly, and why the methods can be compared on one page — they disagree about where the zero is, not about what to do with it.

Why the gestational age is the headline and the due date is a row underneath it

A due date is the midpoint of a distribution, and the distribution is weeks wide. Every method on this page does the same two things: it fixes the single day on which gestational age was zero, and it adds 280 days to that day. The methods differ only in how they find that day, and they disagree with each other by days. What none of them can do is narrow the answer, because the spread is not measurement error — it is the length of pregnancy itself. In the Early Pregnancy Study, which followed 130 unassisted conceptions with the day of ovulation known from daily hormone measurement, the median time from ovulation to birth was 268 days and the length of gestation varied across a span of 37 days even after six preterm births were removed. Thirty-seven days is more than five weeks. A calculator that prints one date and stops has told you the middle of a five-week window and let you believe it told you a day.

So this page makes the gestational age the headline and the date a row underneath it. That is a deliberate inversion of what a due-date calculator usually does, and the reason is the paragraph above. Gestational age in completed days is the quantity that is actually well defined, it is the quantity six other calculators on this site need as an input, and it is the one the bar above is drawn against — a position in the pregnancy, with the published stage named. The estimated due date, the day 37 weeks 0 days falls on and the day 42 weeks 0 days falls on are all printed together, because those last two are the window and the first is only its middle. Dates come out as a year, a month and a day, and as a single YYYYMMDD number: this engine is numeric and has no date type, so 20261019 is how it writes 19 October 2026. There is also no clock, which is why the date you want the gestational age for is something you type — and that turns out to be a feature, because it lets you ask what the gestational age will be on the day of an appointment or a planned delivery.

Naegele’s rule, as Naegele stated it, is not the same as adding 280 days. The nineteenth-century form is a calendar operation: add a year, subtract three months, add seven days. The modern form is arithmetic on days. They are not equivalent, because calendar months are not equal in length. Sweeping every start date from 2024 to 2028 — 1,827 of them — the calendar form lands 0 days later than the 280-day form for 797 of them, 1 day later for 519, 2 days later for 370 and 3 days later for 141. It agrees with the 280-day rule on fewer than half of all possible start dates and is never earlier. This page implements 280 days, which is the convention ACOG states, and prints the gap for the dates you entered so that the discrepancy is visible rather than argued about. The same thing happens one level up: Parikh’s formula, the cycle-adjusted rule taught alongside Naegele’s, is stated as nine months minus 21 days plus the cycle length, which is the calendar form again and inherits the same 0-to-3-day drift. The day-count form used here, 280 plus the cycle length minus 28, is the same idea without the calendar in the way.

The cycle-length adjustment, which most calculators drop. The 280-day rule assumes a 28-day cycle with conception 14 days in. If your cycles run to 35 days, the assumption is wrong by a week in one direction, and the honest correction is to move the dating zero point by the difference: 280 plus (cycle minus 28). A 35-day cycle puts the due date a week later and makes you a week less pregnant today than an unadjusted calculator would say. The adjustment rests on the assumption that all of the variation in cycle length sits in the follicular phase and the luteal phase is fixed, and that assumption holds up better than it has any right to: across 612,613 ovulatory cycles logged by 124,648 app users, the follicular phase averaged 16.9 days with a standard deviation of 5.3 while the luteal phase averaged 12.4 days with a standard deviation of 2.4. The luteal phase varies less than half as much. But note the mean: 12.4 days, not the 14 the convention assumes. Taking 14 as fixed places conception about a day and a half earlier than the average real cycle, and the page keeps 14 anyway, because every downstream gestational-age convention on this site and in every report you will be shown is built on it. That is a convention being preserved for comparability, not a measurement.

Which ultrasound equation, and why it matters more than it looks. Crown-rump length to gestational age is a fitted regression, and the fits disagree. This page implements two and prints both. Robinson & Fleming (1975) is the equation behind most national dating tables and most ultrasound machines, in the form gestational age in days = 8.052 × √(crown-rump length × 1.037) + 23.73; the 1.037 factor is routinely dropped in secondary sources and dropping it costs a day, so the implementation here was checked against a published national dating table and reproduces it exactly at 50, 60, 70, 80 and 84 mm. INTERGROWTH-21st (2014) is the newer fit, prospective, from 4,265 women at eight sites in eight countries including India, in the form gestational age in days = 40.9041 + 3.21585√CRL + 0.348956 × CRL, with a published standard deviation of 2.39102 + 0.0193474 × CRL days. From 25 mm upward the two agree to within a day. Below that they come apart fast: 4 days at 15 mm, 5 at 10 mm, 8 at 5 mm. Since 84 mm is the upper limit at which crown-rump length should be used for dating at all, the usable range of this measurement is also the range in which the equation choice stops mattering — which is worth knowing before anybody redates a pregnancy on a 9 mm embryo.

What the term window actually looks like in a population. Of singleton births registered in the United States in 2022, 11.63% were at 37 completed weeks, 17.45% at 38, 38.84% at 39, 18.26% at 40, 4.88% at 41 and 0.27% at 42 or later; the remaining 8.7% were before 37 weeks. Two things follow. The first is that fewer than a quarter of births happen at or after 40 completed weeks, so arriving at your due date still pregnant puts you in a minority — but a large one. The second is a caution about reading that distribution as biology: the enormous 39-week peak is substantially the footprint of scheduled induction and caesarean section, not of labour starting on its own, and the same report exists because the distribution has been shifting earlier year on year. For the natural spread, the 37-day span from the Early Pregnancy Study is the better figure. The widely repeated claim that about 4% of births fall on the estimated date is consistent with both — one-seventh of the 18.26% born in week 40 is 2.6% — but this page could not find it reported directly in a primary source and does not print it as one.

What this page does not do. It does not decide whether a pregnancy should be redated: that is a clinical decision with published criteria attached to the size of the discrepancy and the gestational age at which the scan was done, and this page computes dates rather than applying those criteria. It does not handle twins, which are dated the same way but delivered earlier, and it says nothing about what any gestational age means for this pregnancy. It is not a clock — both dates are yours. For prematurity correction after birth, the paediatric corrected age calculator owns that and this page does not repeat it. The gestational age computed here is the input to the PAPP-A MoM calculator, free beta-hCG MoM, unconjugated estriol, NIPT positive predictive value, cell-free DNA fetal fraction and the neonatal early-onset sepsis risk calculator, none of which computes it. It is also what the pregnancy weight gain page needs, because a published gain range is only meaningful against a week of gestation. If you are working the other way round — from cycle to conception rather than from conception to birth — that is the ovulation and fertile window page, and the two share one cycle-length convention deliberately: ovulation plus 266 days and this page’s cycle-adjusted 280 give the same date for every cycle length from 20 to 45, which was checked exhaustively rather than assumed.

Advertisement

Frequently asked questions

Why is the big number a gestational age instead of my due date?

Because the due date is the least certain thing on this page and the gestational age is the most certain. Gestational age in days is exactly defined — it is the number of days since the dating zero point, and it is right to the day. The due date is that same quantity projected 280 days forward onto a distribution that spans about five weeks. Making the date the headline and burying the window would be the single most misleading thing this page could do, so the layout is the other way up: the date, the day 37 weeks falls on and the day 42 weeks falls on are printed together, a few lines below a number that is actually solid. The due date is also right there — four rows of it, as a year, a month, a day and a YYYYMMDD number.

Why do I have to type today’s date?

Because this calculator is a numeric expression evaluator with no clock and no date type. Everything you see is arithmetic on day numbers, and a day number has to come from somewhere. The upside is real: because the reference date is an input rather than the system clock, you can ask what the gestational age was at a scan three weeks ago, or will be on the date of a planned delivery, by typing that date instead of today’s.

Does “39 weeks and 2 days” mean 39 weeks have finished, or that I am in the 39th week?

Completed weeks. 39 weeks and 2 days means 39 whole weeks have passed and 2 days of the fortieth, which is 39 × 7 + 2 = 275 days. It does not mean the 39th week is in progress. The distinction matters because the two readings differ by a week, and a week is the difference between early term and full term. The same trap sits one level down: the first day of a period is gestational age zero, not gestational age one day, while it is cycle day 1 and not cycle day 0. Days elapsed and days counted differ by one, and this page always reports days elapsed.

The clinic gave me a different due date. Which is right?

Probably the clinic’s, and the reason is usually the method rather than the arithmetic. A due date set from a first-trimester scan supersedes one calculated from a remembered period, and once a clinic has set a due date it generally does not move. If you want to see where the difference comes from, put the scan’s own figures into method 3 or method 4 here and compare the gestational age each method assigns to the date you typed — the row is printed for exactly this. What this page will not do is tell you whether a discrepancy is large enough to justify redating; that decision has published criteria attached to it and belongs with the person who did the scan.

What happens if I enter a date in the future, or one from two years ago?

It refuses, and says so. The page computes a gestational age of 0 to 336 days — nought to 48 weeks — and anything outside that returns the incomplete message rather than a negative gestational age or a plausible-looking figure from the wrong branch. A dating date later than the date you are asking about gives a negative gestational age and is refused; one more than 48 weeks earlier is refused too. A day that does not exist in the month you chose, such as 31 April or 29 February 2026, is refused rather than rolled silently into the next month.

Why does changing my cycle length move the due date when other calculators ignore it?

Because the 280-day rule has a 28-day cycle built into it, with conception assumed 14 days after the first day of the period. If your cycles are 35 days, that assumption is a week out, and the correction is to move the dating zero point by the difference. Most calculators either have no field for it or accept the number and do nothing with it. This one shows you how many days it moved. It is still an assumption: it holds the luteal phase fixed at 14 days and puts all cycle-length variation in the follicular phase, which the large app-logged datasets support directionally — follicular standard deviation 5.3 days against luteal 2.4 — while also showing the mean luteal phase is 12.4 days rather than 14.

Which crown-rump length equation should I pick?

Above about 25 mm it barely matters: Robinson & Fleming and INTERGROWTH-21st agree to within a day all the way to the 84 mm limit, so pick whichever your report used and the answer will be the same. Below 25 mm it matters a lot and neither is reliable — they are 4 days apart at 15 mm and 8 days apart at 5 mm, and 15 mm is the bottom of the INTERGROWTH-21st fit, so anything smaller is extrapolation on both curves. The practical answer is to use the equation your ultrasound machine used, which is usually Robinson & Fleming, and to treat a dating decision taken on a very small embryo with suspicion whichever curve produced it.

How many babies are actually born on their due date?

Fewer than most people expect, and this page will not give you a single figure because it could not find one reported directly in a primary source. What can be sourced: of United States singleton births in 2022, 18.26% were recorded at 40 completed weeks, so if births were spread evenly across that week about 2.6% would fall on 40 weeks 0 days exactly. And in the one cohort where the day of ovulation was known from daily hormone measurement, gestation length spanned 37 days after preterm births were excluded. The frequently repeated figure of about 4% is in the same region as both, but it is a figure in circulation rather than one this page can point at, so it is quoted here as that and nothing more.

Does this page handle twins?

No. Twin pregnancies are dated the same way a singleton is — the same crown-rump length equations, the same 280 days — but they are delivered considerably earlier, so a due date computed as if for a singleton is a midpoint of a distribution that has moved. Rather than apply a shift this page cannot source cleanly, it computes the singleton figure and says plainly that it is the singleton figure. The population distribution quoted throughout is singleton births for the same reason.

Related calculators

Ovulation & Fertile Window CalculatorPregnancy & ObstetricsPregnancy Weight Gain Calculator (kg, by BMI)Pregnancy & ObstetricsPaediatric Weight Estimation Calculator (kg, Age-Based)Growth & PaediatricsEnergy Requirement Calculator (Mifflin, Harris-Benedict, Schofield)Nutrition SupportAdult Fluid Requirement Calculator (mL/day)Nutrition Support

References

  1. American College of Obstetricians and Gynecologists, American Institute of Ultrasound in Medicine and Society for Maternal-Fetal Medicine. Methods for Estimating the Due Date. Committee Opinion No. 700. Obstet Gynecol 2017;129(5):e150–e154. The source of three figures used here and cited by number rather than reproduced: the estimated due date is 280 days after the first day of the last menstrual period; an in vitro fertilisation pregnancy is dated from the age of the embryo and the date of transfer, giving 261 days from a day-5 transfer and 263 from a day-3 transfer; and crown-rump length is the first-trimester measurement to date on up to 84 mm, about 14 weeks 0 days. ACOG guidance is copyrighted, so the redating criteria table is deliberately not reproduced and this page does not apply it.
  2. Robinson HP, Fleming JE. A critical evaluation of sonar “crown-rump length” measurements. Br J Obstet Gynaecol 1975;82(9):702–10. The first equation implemented here, in the form used by national dating tables and ultrasound software: gestational age in days = 8.052 × √(crown-rump length in mm × 1.037) + 23.73, rounded to the nearest day. The 1.037 factor is frequently omitted in secondary renderings of this equation and omitting it shifts the result by about a day, so the implementation was verified against the Scottish Pregnancy and Newborn Screening Programme crown-rump length dating table, which states this exact form: at 50, 60, 70, 80 and 84 mm the published table gives 11+5, 12+3, 13+1, 13+6 and 14+1 and this page reproduces all five. The form without the 1.037 factor gives 11+4 at 50 mm and does not reproduce the table. A fitted regression is a method rather than an expression, and the 1975 publication is in any case long out of copyright.
  3. Papageorghiou AT, Kennedy SH, Salomon LJ, Ohuma EO, Cheikh Ismail L, Barros FC, et al. International standards for early fetal size and pregnancy dating based on ultrasound measurement of crown–rump length in the first trimester of pregnancy. Ultrasound Obstet Gynecol 2014;44(6):641–8. doi:10.1002/uog.13448. The second equation implemented here: gestational age in days = 40.9041 + 3.21585 × √CRL + 0.348956 × CRL, with the standard deviation of gestational age = 2.39102 + 0.0193474 × CRL, valid for crown-rump length 15 to 95 mm and about 58 to 105 days of gestation, fitted on 4,265 women recruited at eight sites in eight countries — Brazil, China, India, Italy, Kenya, Oman, the United Kingdom and the United States. The Indian site is part of why this equation is offered on a site read largely in India. Note on licensing: the article is distributed under CC BY-NC-ND, which is a NonCommercial licence and this site carries advertising, so no part of the paper, its charts or its tables is reproduced here — only the published equation is implemented, as a method, with attribution. INTERGROWTH-21st is an Oxford-led consortium and is not a WHO publication; the separate WHO fetal growth charts are excluded from this site under the WHO NonCommercial terms.
  4. Jukic AM, Baird DD, Weinberg CR, McConnaughey DR, Wilcox AJ. Length of human pregnancy and contributors to its natural variation. Hum Reprod 2013;28(10):2848–55. doi:10.1093/humrep/det297. The source of the spread figures. The Early Pregnancy Study followed 130 singleton pregnancies from unassisted conception, with the day of ovulation identified from daily urinary hormone measurement rather than inferred from a remembered period; 125 were analysed. Median time from ovulation to birth 268 days, which is 38 weeks and 2 days from ovulation and corresponds to 282 days from a notional last menstrual period under the 14-day convention — two days later than the 280-day rule. The variation in pregnancy length spanned 37 days after excluding women with complications or preterm births, with a coefficient of variation of 3.7% when dated from ovulation.
  5. Martin JA, Osterman MJK. Shifts in the Distribution of Births by Gestational Age, United States, 2014–2022. National Vital Statistics Reports, vol 73, no 1. Hyattsville, MD: National Center for Health Statistics, 31 January 2024. The population distribution quoted on this page, for singleton births by the obstetric estimate of gestation in completed weeks: 11.63% at 37 weeks, 17.45% at 38, 38.84% at 39, 18.26% at 40, 4.88% at 41 and 0.27% at 42 or later in 2022, with the balance of 8.7% before 37 weeks. A US Government work and freely usable. The report exists because this distribution has been moving earlier, which is also the reason it should not be read as the natural distribution of gestation length — the 39-week peak reflects scheduled delivery.
  6. Eunice Kennedy Shriver National Institute of Child Health and Human Development. Know Your Terms: gestational age designations. NICHD, National Institutes of Health. The term definitions behind the bands above: early term 37 weeks 0 days to 38 weeks 6 days, full term 39 weeks 0 days to 40 weeks 6 days, late term 41 weeks 0 days to 41 weeks 6 days, post-term 42 weeks 0 days and beyond. These replaced the undifferentiated 37-to-42-week definition of “term” in a joint ACOG and Society for Maternal-Fetal Medicine committee opinion of November 2013 (ACOG Committee Opinion No. 579). Taken from the NIH page, which is a US Government work, rather than from the copyrighted committee opinion; a set of boundary definitions is in any case a definition and not an expression.
  7. Bull JR, Rowland SP, Berglund Scherwitzl E, Scherwitzl R, Gemzell Danielsson K, Harper J. Real-world menstrual cycle characteristics of more than 600,000 menstrual cycles. npj Digital Medicine 2019;2:83. doi:10.1038/s41746-019-0152-7. CC BY 4.0, which permits commercial use with attribution. 612,613 ovulatory cycles from 124,648 users aged 18 to 45: mean cycle length 29.3 days (SD 5.2), mean follicular phase 16.9 days (SD 5.3), mean luteal phase 12.4 days (SD 2.4), and a mean within-woman variation in cycle length of 2.6 days. Mean follicular phase length rose from 10.4 days in cycles of 15 to 20 days to 26.8 days in cycles of 36 to 50 days. This is the evidence for the assumption the cycle-length adjustment rests on — that cycle-length variation sits mainly in the follicular phase — and also the evidence that the 14-day luteal phase the convention uses is about a day and a half longer than the average.
  8. Edwards KI, Itzhak P. Estimated Date of Delivery. StatPearls. Treasure Island, FL: StatPearls Publishing. Cited for one naming fact only: the cycle-adjusted rule is known as Parikh’s formula and is stated as the first day of the last menstrual period plus nine months, minus 21 days, plus the length of the previous cycle. No primary publication for the formula could be located. Because that statement is a calendar operation it carries the same 0-to-3-day drift against a day count as the calendar form of Naegele’s rule; the day-count equivalent used here is 280 + (cycle − 28).
  9. Naegele FK. Lehrbuch der Geburtshilfe für Hebammen. Heidelberg, 1812. The origin of the calendar rule — add a year, subtract three months, add seven days. An 1812 publication and long out of copyright. The claim that this is not equivalent to adding 280 days was not taken on trust: every start date from 1 January 2024 to 31 December 2028 was computed both ways, 1,827 dates, and the calendar form was 0 days later in 797 cases, 1 day later in 519, 2 days later in 370 and 3 days later in 141, and never earlier.
  10. Wilcox AJ, Weinberg CR, Baird DD. Timing of sexual intercourse in relation to ovulation. N Engl J Med 1995;333(23):1517–21. Cited here for the convention this page’s method 2 rests on: conception happens on or within a few days before the day of ovulation, which the gestational-age convention fixes at 14 days. The fertile window itself, and the day-specific probabilities, are on the ovulation and fertile window page.
  11. Verification note. Every date identity on this page was swept rather than sampled. The Gregorian-to-Julian-Day-Number conversion and its inverse were checked against an independent calendar implementation for all 73,414 dates from 1 January 1900 to 31 December 2100, with no disagreement, and the inverse recovered the original year, month and day in every case. The round trip last menstrual period → due date → last menstrual period was checked for every start date from 1990 to 2060 against every cycle length from 20 to 45 days, 674,258 combinations, with no failure. The round trip gestational age → date → gestational age, and the identity 7 × completed weeks + days = total days, were checked for every gestational age from 0 to 310 days against every start date over a 21-year span, 2,385,681 combinations, with no failure. The cross-check against the ovulation page — that ovulation plus 266 days equals this page’s cycle-adjusted 280-day due date whenever the luteal phase is 14 — held for every cycle length and every start date tested.

CalcEngines health calculators are for education and for checking arithmetic that has already been decided elsewhere. They are not medical advice, they do not decide what to give, and they do not replace the judgement of a doctor, nurse, midwife or dietitian who knows the person in front of them. Every figure depends on the values you enter and on the assumptions stated on the page — check it against the prescription, the product label and your local policy before acting on it.