Maternal & Newborn Atlas
About
Menu

Questions to ask about induction

What the sources say about why labour may be induced, how induction is done, and what the ARRIVE and SWEPIS trials studied, with questions to ask a clinician.

What induction is and why it may be done

  • MedlinePlus says induction uses medicines or other methods to start labour.1
  • NICHD says induction is used to make contractions start so that delivery can occur.2
  • MedlinePlus says induction is usually done only for one of two reasons.3, 4
  • One is that the due date has passed.3
  • The other is a problem with the pregnancy that risks the health of the mother or the fetus.4
  • NICHD also says induction is usually used only for those two reasons.5
  • The NICHD page on induction shows a last reviewed date of 9/1/2017.6
  • The ARRIVE results were published in 2018 and the SWEPIS results in 2019, after that review.7, 8
  • NICHD says that in most cases induction is limited to a problem with the pregnancy or to a pregnancy that has gone past the infant's due date.9, 10
  • NICHD says that women who want induction for non-medical reasons can discuss it with their health care providers.11
  • The US Pitocin label lists induction for a medical indication, with examples that include maternal diabetes and pre-eclampsia at or near term.12

Getting the cervix ready

  • NICHD says providers use a scoring system, called the Bishop score, to judge how ready the cervix is for labour.13
  • If the cervix is not ready, NICHD says a provider may suggest one of three steps to ripen it.14
  • The three steps are stripping the membranes, giving prostaglandins and inserting a catheter.15, 16, 17
  • NICHD describes stripping the membranes as separating the thin tissue of the amniotic sac from the wall of the uterus.15
  • NICHD says this makes the body release prostaglandins, which soften the cervix and cause contractions.15
  • NICHD says prostaglandins may be inserted into the vagina or given by mouth.16
  • NICHD adds that the body makes prostaglandins naturally to ripen the cervix, but that sometimes extra amounts are needed.16
  • NICHD describes the catheter as a small tube with an inflatable balloon on the end that can be placed in the cervix to widen it.17

Starting contractions

  • NICHD says a provider may recommend amniotomy or oxytocin once the cervix is ripe.18, 19, 20
  • NICHD says these techniques start contractions or make them stronger.18
  • NICHD describes amniotomy as making a small hole in the amniotic sac, so that the water breaks and contractions start.19
  • NICHD says oxytocin, also called Pitocin, is a hormone the body makes that causes contractions.20
  • NICHD says oxytocin is given to start labour or to speed up labour that has already begun.20

What the US labels say about three products

  • The US label for Cervidil, a vaginal insert, covers starting or continuing cervical ripening at or near term.21
  • It applies to pregnant women who have a medical or obstetrical (pregnancy-related) reason for induction.21
  • FDA cleared the Cook Cervical Ripening Balloon in 2013.22
  • The cleared use is mechanical dilation of the cervical canal before labour induction at term, when the cervix is unfavorable (not ready) for induction.22
  • The Pitocin label has a boxed notice about elective induction.12
  • The notice defines elective induction as starting labour in a pregnant individual who has no medical indication for induction.23
  • The notice says "the available data are inadequate to evaluate the benefits-to-risks considerations", and that Pitocin is not indicated for elective induction.12, 24
  • The atlas records for these products note that a label or clearance is not a recommendation to use a product for any woman.12, 21, 22
  • They add that labelled uses differ by country.12, 21

What the ARRIVE trial studied

  • MedlinePlus says labour usually begins between 37 and 42 weeks of pregnancy.25
  • In both trials, expectant management means waiting instead of inducing at the set week.26, 27
  • ARRIVE was a multicentre trial in the United States of low-risk first-time mothers.28
  • The women were randomly assigned at 38 weeks 0 days to 38 weeks 6 days.28
  • They were assigned either to induction between 39 weeks 0 days and 39 weeks 4 days, or to expectant management.28
  • The ARRIVE registry record labels the induction group "Elective Induction of Labor".29
  • The registry record says the induction group was to be induced with oxytocin.30
  • The registry record says the other group was managed expectantly, unless a medical indication arose, until at least 40 weeks 5 days.26
  • The ARRIVE registry lists no masking, so women and staff knew each group.28
  • The main outcome was a composite, a single count of several outcomes: death around birth or severe newborn complications.28
  • It occurred in 4.3% of babies in the group assigned to induction and 5.4% in the group assigned to expectant management.28
  • The authors report that induction did not result in a significantly lower frequency of that combined outcome, meaning a difference no larger than chance alone would explain.28
  • Caesarean delivery was the main secondary outcome.31
  • It occurred in 18.6% of women in the group assigned to induction and 22.2% in the group assigned to expectant management, a significantly lower frequency with induction.31
  • The ARRIVE results concern low-risk first-time mothers assigned to induction at 39 weeks.28, 31
  • They do not describe induction for a medical reason or at other times.28, 31

What the SWEPIS trial studied

  • SWEPIS was a Swedish trial in women with an uncomplicated, low-risk pregnancy of one baby.27
  • The women were randomly assigned either to be induced at 41 weeks or to wait, with induction planned at 42 weeks.27
  • The main outcome was a composite that included stillbirth, newborn death and a list of serious newborn problems.27
  • It occurred in 2.4% of pregnancies in the group assigned to induction, and in 2.2% of those assigned to expectant management.27
  • The difference was not significant.27
  • No deaths around birth occurred in the group assigned to induction.27
  • Six occurred in the group assigned to expectant management: five stillbirths and one death in the early newborn period.27
  • The trial was stopped early because deaths around birth were more frequent among the women managed expectantly.27
  • The authors call the lower mortality a secondary outcome and ask for cautious interpretation.27
  • The groups did not differ in caesarean births, in instrumental vaginal births, or in major maternal morbidity.32
  • The trial was open label, meaning women and staff knew each group, and it was stopped early.27
  • The expectant management group could still be induced at 42 weeks.27
  • So SWEPIS compares two timings of induction, not induction against no induction.27

Reading the two trials together

  • The two trials used different weeks, countries and groups of women, so each result describes its own group.27, 28

Topics the sources raise about induction

  • The reason for induction, which MedlinePlus and NICHD give as a problem with the pregnancy or a due date that has passed.3, 4, 5
  • How ready the cervix is, which NICHD says providers judge with the Bishop score.13
  • Which steps would be used to ripen the cervix or to start contractions, among those NICHD lists.14, 18
  • The week of pregnancy, because ARRIVE and SWEPIS each tested induction at a set week.27, 28
  • The kind of pregnancy, because both trials studied low-risk women.27, 28

Questions to ask

  1. What is the reason induction is being suggested: a problem with the pregnancy, or a due date that has passed?
  2. Is there a medical reason for induction here, and what would waiting for labour to start on its own involve?
  3. How ready is the cervix, and would it need to be ripened before labour is induced?
  4. Which methods would be used to ripen the cervix or start contractions, and what does each one involve?
  5. Which medicine or device might be used, and what is it labelled for?
  6. Do the ARRIVE or SWEPIS results apply to this pregnancy, given who took part in each trial?
  7. What are the possible benefits and risks of induction at this point in pregnancy, compared with waiting?
  8. If waiting is chosen, at what point would induction be offered?

Sources

  1. Childbirth (MedlinePlus health topic) Webpage, National Library of Medicine (MedlinePlus), Summary, What are the stages of labor?, paragraph 4, sentences 1 and 2. Retrieved .
  2. What is induction of labor? (NICHD health topic) Webpage, Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), Opening paragraph, sentence 2. Retrieved .
  3. Childbirth (MedlinePlus health topic) Webpage, National Library of Medicine (MedlinePlus), Summary, What are the stages of labor?, paragraph 4, sentence 3, part 1. Retrieved .
  4. Childbirth (MedlinePlus health topic) Webpage, National Library of Medicine (MedlinePlus), Summary, What are the stages of labor?, paragraph 4, sentence 3, part 2. Retrieved .
  5. What is induction of labor? (NICHD health topic) Webpage, Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), When would a provider induce labor?. Retrieved .
  6. What is induction of labor? (NICHD health topic) Webpage, Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), Page footer, Last Reviewed Date. Retrieved .
  7. Labor Induction versus Expectant Management in Low-Risk Nulliparous Women Publication, The New England journal of medicine, Journal issue, publication date. Retrieved .
  8. Induction of labour at 41 weeks versus expectant management and induction of labour at 42 weeks (SWEdish Post-term Induction Study, SWEPIS): multicentre, open label, randomised, superiority trial Publication, BMJ (Clinical research ed.), Journal issue, publication date. Retrieved .

    Source problem. A cited source is corrected. Check the sources before relying on this claim.

  9. What is induction of labor? (NICHD health topic) Webpage, Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), Can induction be requested?, sentence 1, part 1. Retrieved .
  10. What is induction of labor? (NICHD health topic) Webpage, Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), Can induction be requested?, sentence 1, part 2. Retrieved .
  11. What is induction of labor? (NICHD health topic) Webpage, Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), Can induction be requested?, sentence 3. Retrieved .

  12. FDA lists an original approval of NDA018261 for Pitocin (oxytocin injection) on 1980-11-19, classified Type 5, new formulation or new manufacturer; the current sponsor in Drugs@FDA is PH Health and the labelling is by Par Health USA, LLC. The US label (version 15, effective 2026-05-05) lists use before birth to start or improve uterine contractions, where this is desirable and considered suitable for reasons of fetal or maternal concern, in order to achieve vaginal delivery, and use after birth to produce contractions in the third stage of labour and to control postpartum bleeding. The antepartum indications are (1) induction of labour in patients with a medical indication such as Rh problems, maternal diabetes or pre-eclampsia at or near term, (2) stimulation or reinforcement of labour in selected cases of uterine inertia, and (3) adjunctive therapy in incomplete or inevitable abortion. A boxed notice says Pitocin is not indicated for elective induction of labour.

    Applies to Maternal
    • The antepartum uses on the US label (initiation or improvement of uterine contractions, where this is desirable and considered suitable for reasons of fetal or maternal concern, in order to achieve vaginal delivery: (1) induction of labour in patients with a medical indication, such as Rh problems, maternal diabetes or pre-eclampsia at or near term; (2) stimulation or reinforcement of labour in selected cases of uterine inertia; (3) adjunctive therapy in incomplete or inevitable abortion) and the postpartum use (producing uterine contractions in the third stage of labour and controlling postpartum bleeding or haemorrhage). The label says the product is not indicated for elective induction of labour.
    • Not stated in the source
    • United States (FDA-approved labelling)
    • Pregnancy, labour and the time just after birth
    • Gestational age Not reported
    • Postnatal age Not reported
    Follow-up Not applicable (regulatory context)

    Limitations. The date is the original submission's action date in Drugs@FDA; it is not the date of the current label, which has changed many times since (the application lists supplements for labelling and manufacturing). The indication wording is from the openFDA copy of the current label and was not compared with the DailyMed page. Approval means FDA found the product acceptable for the labelled uses in the United States; it is not a recommendation to use it for any woman, and labelled uses differ by country. Three other label set ids exist for the same application and were not read.

  13. What is induction of labor? (NICHD health topic) Webpage, Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), Monitoring the Body's Preparation for Labor, paragraph 2, sentence 1. Retrieved .
  14. What is induction of labor? (NICHD health topic) Webpage, Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), Preparing the Cervix for Labor, lead-in. Retrieved .
  15. What is induction of labor? (NICHD health topic) Webpage, Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), Preparing the Cervix for Labor, list item 1. Retrieved .
  16. What is induction of labor? (NICHD health topic) Webpage, Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), Preparing the Cervix for Labor, list item 2. Retrieved .
  17. What is induction of labor? (NICHD health topic) Webpage, Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), Preparing the Cervix for Labor, list item 3. Retrieved .
  18. What is induction of labor? (NICHD health topic) Webpage, Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), How is labor induced?, lead-in. Retrieved .
  19. What is induction of labor? (NICHD health topic) Webpage, Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), How is labor induced?, list item 1. Retrieved .
  20. What is induction of labor? (NICHD health topic) Webpage, Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), How is labor induced?, list item 2. Retrieved .

  21. FDA lists an original approval of NDA020411 for Cervidil (dinoprostone vaginal insert) on 1995-03-30, classified Type 3, new dosage form; the sponsor in Drugs@FDA is Ferring Pharmaceuticals. The US label (version 10, effective 2023-06-02) says Cervidil is indicated for the initiation and/or continuation of cervical ripening in pregnant women at or near term in whom there is a medical or obstetrical indication for the induction of labor.

    Applies to Maternal
    • Pregnant women at or near term in whom there is a medical or obstetrical indication for the induction of labour, for initiation or continuation of cervical ripening.
    • Not stated in the source
    • United States (FDA-approved labelling)
    • Pregnancy at or near term
    • Gestational age Not reported
    • Postnatal age Not reported
    Follow-up Not applicable (regulatory context)

    Limitations. The date is the original submission's action date in Drugs@FDA, not the date of the current label. The indication wording is from the openFDA copy of the label and was not compared with the DailyMed page. Approval is not a recommendation for any woman, and labelled uses differ by country.

  22. FDA cleared the Cook Cervical Ripening Balloon from applicant Cook, Inc. under 510(k) K131206, with the decision Substantially Equivalent dated 2013-09-27. The 510(k) summary states the intended use: "indicated for mechanical dilation of the cervical canal prior to labor induction at term when the cervix is unfavorable for induction". FDA classifies the product code, PFJ, as a catheter, balloon, for dilation of the cervical canal prior to labor, in device class 2.

    Applies to Maternal
    • Women at term for whom labour induction is planned and whose cervix is unfavorable for induction. The cleared use is mechanical dilation of the cervical canal prior to labor induction at term when the cervix is unfavorable for induction (510(k) summary, Intended Use). It is not cleared for other times in pregnancy or for other uses.
    • Not stated in the source
    • United States (FDA clearance)
    • Term pregnancy, before labour induction
    • Gestational age Not reported
    • Postnatal age Not reported
    Follow-up Not applicable (regulatory context)

    Limitations. Clearance means FDA found the device substantially equivalent to a legally marketed predicate; it is not an approval and not a recommendation to use the device for any woman. The summary is a scanned PDF read by OCR text; contraindications, warnings and performance data were not read. The openFDA record is the dataset copy as of its update date.

  23. Pitocin drug label (openFDA record, set id 6e5a66fc-e507-497c-b5ce-44a8c95898ad) Label, U.S. Food and Drug Administration (openFDA drug label endpoint), results[0].boxed_warning[0], sentence 1. Retrieved .
  24. Pitocin drug label (openFDA record, set id 6e5a66fc-e507-497c-b5ce-44a8c95898ad) Label, U.S. Food and Drug Administration (openFDA drug label endpoint), results[0].boxed_warning[0], sentence 2. Retrieved .
  25. Childbirth (MedlinePlus health topic) Webpage, National Library of Medicine (MedlinePlus), Summary, What is childbirth?, paragraph 2, sentence 2. Retrieved .
  26. A Randomized Trial of Induction Versus Expectant Management Registry, U.S. National Library of Medicine (ClinicalTrials.gov), protocolSection.armsInterventionsModule.armGroups[0].description. Retrieved .

  27. SWEPIS randomly assigned women with an uncomplicated, low-risk pregnancy of one baby, one to one, to induction of labour at 41 weeks or to expectant management with induction at 42 weeks. The composite perinatal outcome (stillbirth, neonatal death and a list of serious newborn problems) was found in 33 of 1381 pregnancies (2.4%) after induction and 31 of 1379 (2.2%) with expectant management. The relative risk was 1.06 (95% confidence interval 0.65 to 1.73; P=0.90), so the difference was not significant. After induction there were no perinatal deaths. With expectant management there were six, made up of five stillbirths and one death in the early neonatal period (P=0.03). The study was stopped early because deaths around birth were more frequent in the expectant management group (the monitoring board advised stopping on 2 October 2018; recruitment ended on 13 October 2018). The authors call the lower mortality a secondary outcome and ask for cautious interpretation.

    Applies to Mixed recipients
    • Fetuses and the babies born of them, fetal and newborn together, because the trial counted stillbirth and neonatal outcomes in one composite and counted perinatal deaths (stillbirths and early neonatal deaths) together. The mother's gestational age at randomisation was 40+6 to 41+1 weeks (the paper spells the window out in weeks and days joined by a plus sign; no age bound is recorded in structured form because that spelling cannot be bound to a quote). Use this context only for the composite outcome and perinatal deaths.
    • Fourteen hospitals in Sweden: five university clinics and nine county hospitals.
    • Sweden
    • Late term and the perinatal period
    • Gestational age Not reported
    • Postnatal age Not reported
    Follow-up Randomisation to the perinatal period; the abstract does not give the follow-up time.

    Limitations. Open-label, stopped early (the independent monitoring board advised stopping on 2 October 2018), so the mortality difference rests on a small number of deaths and an early stop; the authors say the results should be interpreted cautiously. Perinatal mortality was a secondary outcome. The expectant management group could still be induced at 42 weeks, so this compares timing, not induction against no induction. Only low-risk women were included. A published erratum corrects the neonatal intensive care admission figures in table 3; those figures are not used here, and the erratum does not change the numbers above. PubMed and the PMC full text were read. The ISRCTN registry record was not fetched.

  28. ARRIVE randomly assigned low-risk first-time mothers at 38 weeks 0 days to 38 weeks 6 days of gestation either to labour induction at 39 weeks 0 days to 39 weeks 4 days or to expectant management. The primary outcome, a composite of perinatal death or severe neonatal complications, occurred in 4.3% of neonates in the induction group and 5.4% in the expectant-management group (relative risk 0.80, 95% confidence interval 0.64 to 1.00). The authors report that induction did not result in a significantly lower frequency of the composite adverse perinatal outcome.

    Applies to Mixed recipients
    • Fetuses and newborns of low-risk nulliparous women who were randomly assigned at 38 weeks 0 days to 38 weeks 6 days of gestation either to labour induction at 39 weeks 0 days to 39 weeks 4 days or to expectant management. The composite counts fetal and newborn outcomes together: the registry says perinatal death includes antepartum stillbirth, intrapartum stillbirth and neonatal death. The gestational bounds are the mother's gestational age at randomisation. Use this context only for the composite perinatal outcome.
    • Multicentre trial in the United States; the abstract does not say what kind of site took part.
    • United States
    • Antepartum pregnancy, birth and the newborn period to hospital discharge
    • Gestational age 38 weeks 0 days to 38 weeks 6 days, measured at enrolment
    • Postnatal age Not reported
    Follow-up Per the registry, the composite runs from delivery through 72 hours after birth, except perinatal death, which runs from the antepartum pregnancy period through hospital discharge.

    Limitations. The abstract and the registry record were read; the full text was not, because the PMC copy is not in the Open Access subset, so the analysis methods are not described here. The registry lists the composite's components: perinatal death, respiratory support within 72 hours after birth, an Apgar score of 3 or less at 5 minutes, hypoxic-ischemic encephalopathy, seizure, infection, meconium aspiration, birth trauma, intracranial or subgaleal haemorrhage, and hypotension needing vasopressor support. The registry lists masking as none. The percentages are as printed in the abstract, and the abstract does not give the number of neonates in each group, so none is computed. The result applies to low-risk first-time mothers offered induction at 39 weeks; it does not apply to induction for a medical reason or at other times. The registry shows the expectant-management arm continued until at least 40 weeks 5 days unless a medical indication arose.

  29. A Randomized Trial of Induction Versus Expectant Management Registry, U.S. National Library of Medicine (ClinicalTrials.gov), protocolSection.armsInterventionsModule.armGroups[1].label. Retrieved .
  30. A Randomized Trial of Induction Versus Expectant Management Registry, U.S. National Library of Medicine (ClinicalTrials.gov), protocolSection.armsInterventionsModule.interventions[0].description. Retrieved .

  31. In ARRIVE, caesarean delivery was the principal secondary outcome. It occurred in 18.6% of women in the induction group and 22.2% with expectant management (relative risk 0.84, 95% confidence interval 0.76 to 0.93), a significantly lower frequency with induction. The women were low-risk first-time mothers randomly assigned at 38 weeks 0 days to 38 weeks 6 days of gestation to induction at 39 weeks 0 days to 39 weeks 4 days or to expectant management.

    Applies to Maternal
    • Low-risk nulliparous women (no previous pregnancy beyond 20 weeks per the registry) with a singleton pregnancy who were randomly assigned at 38 weeks 0 days to 38 weeks 6 days of gestation either to labour induction at 39 weeks 0 days to 39 weeks 4 days or to expectant management. The gestational bounds are the gestational age at randomisation.
    • Multicentre trial in the United States; the abstract does not say what kind of site took part.
    • United States
    • Pregnancy from 38 weeks 0 days, to delivery
    • Gestational age 38 weeks 0 days to 38 weeks 6 days, measured at enrolment
    • Postnatal age Not reported
    Follow-up To delivery (caesarean delivery was the principal secondary outcome).

    Limitations. This is a secondary outcome. The abstract and registry were read, not the full text, so adjustment, missing data and other methods are not described here. The abstract gives no number of women in each group for this outcome, so none is computed. The result concerns low-risk first-time mothers and does not apply to other women. The primary composite outcome measured in babies is a separate claim in a separate context, and it did not differ significantly between the groups.

  32. In SWEPIS, the groups did not differ in how often women had a caesarean birth, an instrumental vaginal birth or any major maternal morbidity. One group was induced at 41 weeks and the other was managed expectantly and induced at 42 weeks. The study compared the two timings in 2760 women with an uncomplicated, low-risk pregnancy of one baby, and was stopped early.

    Applies to Maternal
    • Women with an uncomplicated, low-risk pregnancy of one baby, head down. They were assigned at random, one to one, through the Swedish Pregnancy Register, either to induction of labour at 41 weeks or to expectant management with induction at 42 weeks. Their gestational age at randomisation was 40+6 to 41+1 weeks (no structured age bound is recorded, as for the babies' context).
    • Fourteen hospitals in Sweden: five university clinics and nine county hospitals.
    • Sweden
    • Late term, from randomisation to delivery
    • Gestational age Not reported
    • Postnatal age Not reported
    Follow-up To delivery and maternal discharge; the abstract does not give the follow-up time.

    Limitations. The abstract gives no figures for these maternal outcomes, so none are stated. The study was open label and stopped early. The women were low-risk and the expectant group could be induced at 42 weeks. The perinatal results are a separate claim in a separate context. The SWEPIS paper has a published erratum (BMJ 2021;375:n3072) that corrects only the neonatal intensive care admission row of table 3; it does not touch these maternal outcomes.

About this guide

Written for
People who are pregnant and thinking about induction, and the people with them
What it covers
Written from two US government health pages, a trial registry record, summaries of two trials and three US product records. The NICHD page was last reviewed in 2017. It does not compare methods of induction.
Pages it explains
Sources checked
2026-10-15
Next check due
2027-04-15
Reading level
Estimated grade 8.6 (target 9)

Every sentence cites a source. Words in quotation marks are the source's own; the rest is the atlas's summary.

https://maternal.openmedicalatlas.org/guides/induction-questions/ · Release cb1244a2c7003676, built 2026-10-15