Objective
We sought to describe details of labor induction, including precursors and methods, and associated vaginal delivery rates.
Study Design
This was a retrospective cohort study of 208,695 electronic medical records from 19 hospitals across the United States, 2002 through 2008.
Results
Induction occurred in 42.9% of nulliparas and 31.8% of multiparas and elective or no recorded indication for induction at term occurred in 35.5% and 44.1%, respectively. Elective induction at term in multiparas was highly successful (vaginal delivery 97%) compared to nulliparas (76.2%). For all precursors, cesarean delivery was more common in nulliparas in the latent compared to active phase of labor. Regardless of method, vaginal delivery rates were higher with a ripe vs unripe cervix, particularly for multiparas (86.6-100%).
Conclusion
Induction of labor was a common obstetric intervention. Selecting appropriate candidates and waiting longer for labor to progress into the active phase would make an impact on decreasing the national cesarean delivery rate.
The induction rate in the United States has more than doubled from 9.5% of all deliveries in 1990 to 22.5% in 2006. The increasing induction rate may be due in part to increased rates of complications such as gestational diabetes and preeclampsia as a result of increasing maternal age and body mass index (BMI) in the current obstetric population. Another key contributor, however, has been a rise in elective induction, when there is no medical or obstetrical indication for delivery. Induction of labor has been implicated in an increased risk of cesarean delivery in some studies, while other observational studies and small randomized controlled trials have demonstrated a decreased risk of cesarean and potentially improved neonatal outcomes with elective delivery. Elective inductions may be preferentially performed in healthy pregnancies and for women who have a more favorable Bishop score, factors that typically translate to better maternal and neonatal outcomes. Yet the evidence of the benefits of elective induction <41 0/7 weeks of gestation is insufficient, since many of these studies have been limited by methodological deficiencies. These discrepant results also suggest that the success of induction and maternal and perinatal outcomes may differ depending on the indication for induction, as well as on cervical readiness. The purpose of this study is to describe in detail the maternal and obstetrical characteristics of induction of labor in a recent obstetrical cohort in the United States, including precursors and methods for induction of labor and associated vaginal delivery rates.
Materials and Methods
The Consortium on Safe Labor was a study conducted by the Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, and has been described in detail elsewhere. In brief, this was a retrospective cohort study involving 228,668 deliveries from 2002 through 2008 from 12 clinical centers and 19 hospitals representing 9 American College of Obstetricians and Gynecologists (ACOG) US districts. Institutional review board approval was obtained by all participating institutions. The population was then standardized by assigning a weight to each subject using ACOG district, maternal race/ethnicity, parity, and plurality based on 2004 National Natality data. To limit overcontribution by women who had multiple pregnancies during the study period, we included the first pregnancy if a woman had >1 pregnancy in the present study.
Induction of labor was a predefined variable when either the patient’s electronic medical record indicated that there was an induction and/or a method or start time for induction was recorded in the patient’s chart. There was a separate variable for labor augmentation. We defined attempts at vaginal delivery to include all women with at least 2 vaginal examinations in the labor progression database. We established 3 categories of precursors for induction: “indicated,” “elective,” and “no recorded indication.” The indication for induction was used to identify the precursors for delivery and classified using the following hierarchy. First, women with premature rupture of membranes (ROM) were always classified as such. Thus, if a woman undergoing induction of labor had both premature ROM and another indication, she was only counted once in the premature ROM category. Second, we supplemented the indication for induction with all potential maternal, fetal, or obstetric complications of pregnancy, which were then included in the “indicated” category. For example if a woman had an induction and no indication was recorded but the pregnancy was complicated by preeclampsia, then she was included in the “indicated” precursor “preeclampsia.” A woman could have >1 pregnancy condition in the “indicated” category. We also included in the “indicated” category those women who were admitted to labor and delivery for an unspecified fetal or maternal reason. Third, if a site indicated that the induction was elective; no other indications for induction were provided; and there were no other obstetric, fetal, or maternal conditions complicating the pregnancy, then the precursor for induction was designated “elective.” If a delivery indication was noted as postdates or postterm with no other indications listed but <41 weeks of gestation, these were also coded as elective. Finally, the “no recorded indication” category encompassed all inductions as identified by the site with no other obstetric, fetal, or maternal conditions of the pregnancy, including if no reason for induction was provided. Method of labor induction included all methods for cervical ripening and induction with multiple values was allowed.
Demographic data were summarized. We then investigated the precursors among women undergoing induction of labor at 24-41 weeks, which amounts to a weighted sample size of 1,323,407. One site did not provide indications for induction and was not included in the precursor analysis leaving a weighted sample size of 1,281,193.
For the next analysis, the number and percentage of women with singleton gestations and vertex presentation (n = 1,231,662) who underwent induction of labor were calculated for each method of induction and stratified by both the need for cervical ripening and preterm (24-36 weeks of gestation) vs term (37-41 weeks of gestation). Kaplan-Meier curves were created for nulliparous women with a singleton gestation in vertex presentation and show the cervical dilation reached at the time of intrapartum cesarean delivery. These curves are stratified by the major precursor categories and by preterm (24 to <37 weeks) and term (37 to <42 weeks). Four hospitals did not report methods of induction, and 2 hospitals did not report cervical dilation at admission, leaving cases from 13 of the 19 hospitals available for analysis (n = 986,009). An additional 157,768 of the inductions did not have a method of induction reported and were excluded from the analysis. Of the remaining cases (n = 828,241), only 146,793 (17.7%) had an original Bishop score available with all 5 components reported, while 703,105 (84.9%) had information available on cervical dilation, effacement, and station. Therefore, as previously described elsewhere, we used a simplified Bishop score, comprised of dilation, effacement, and station only, to determine the cervical readiness (ripe vs unripe) for induction. We defined an unripe cervix as a simplified Bishop score ≤4 because of similar sensitivity and specificity to the original Bishop score ≤6, the definition of an unfavorable cervix. We compared maternal, obstetrical, and hospital characteristics between those with missing information and those where either the method of induction or cervical information was known. Women with a missing method of induction were slightly more likely to have a ripe cervix (21.8% vs 18.1%), but slightly more likely to deliver by cesarean (22.9% vs 20.1%). Cesarean delivery was also higher in women with a missing simplified Bishop score (25.8% vs 19.8%).
Given that this is a descriptive analysis with very large sample size, no significance testing was performed. All analyses were performed using software (SAS, version 9.2; SAS Institute Inc, Cary, NC).
Results
The prevalence of induction of labor by maternal and obstetric characteristics is presented in Table 1 . Overall, induction was common, occurring in 42.9% of nulliparous women and 31.8% of multiparous women. Among women attempting vaginal delivery, induction was more prevalent with increasing maternal age, those with higher BMI, and postdate pregnancies. The rate of induction varied modestly by race/ethnicity, ranging from 41.6% among non-Hispanic whites to 46.7% among Hispanic nulliparas, with multiracial/other/unknown appreciably lower rate (35.3%), and ranging in multiparas from 28.0% for multiracial/other/unknown to 33.8% among non-Hispanic whites. Induction in nulliparas varied substantially by hospital type, occurring most frequently in university-affiliated teaching hospitals (47.6%), followed by teaching community hospitals (36.7%), then nonteaching community hospitals (27.4). Rates by hospital type for multiparas were more similar (30.0-37.4%). The prevalence of induction in multiparous women with a previous uterine scar was low (8.3%). Induction was less common at gestational ages <34 weeks, and had the highest prevalence at 41 weeks of gestation (63.4% in nulliparas and 50.1% in multiparas).
| Nulliparous | Multiparous | |||||
|---|---|---|---|---|---|---|
| Variable | Proportion of population, % n = 1,612,035 | Prevalence in women undergoing induction (42.9%), % | Prevalence in women attempting vaginal delivery a (47.6%), % | Proportion of population, % n = 2,033,140 | Prevalence in women undergoing induction (31.8%), % | Prevalence in women attempting vaginal delivery a (41.0%), % |
| Maternal age, y | ||||||
| <18 | 6.5 | 37.2 | 38.9 | 0.4 | 23.4 | 26.3 |
| 18-34 | 83.0 | 43.3 | 47.4 | 77.8 | 31.9 | 40.0 |
| ≥35 | 10.4 | 43.9 | 55.7 | 21.7 | 31.8 | 45.6 |
| Missing | 0 .1 | 22.6 | 34.5 | 0.1 | 14.2 | 20.3 |
| Race/ethnicity | ||||||
| Non-Hispanic white | 58.7 | 41.6 | 46.6 | 54.3 | 33.8 | 43.7 |
| Non-Hispanic black | 13.6 | 44.1 | 48.1 | 14.6 | 29.1 | 37.9 |
| Hispanic | 20.6 | 46.7 | 50.8 | 25.0 | 29.7 | 38.0 |
| Asian/Pacific Islander | 3.8 | 44.4 | 48.8 | 2.9 | 30.4 | 38.6 |
| Other/unknown | 3.3 | 35.3 | 39.3 | 3.2 | 28.0 | 35.9 |
| Health insurance | ||||||
| Private | 55.8 | 41.0 | 45.9 | 51.2 | 32.9 | 42.7 |
| Public | 31.1 | 39.7 | 43.2 | 34.8 | 27.4 | 36.0 |
| Self-pay | 1.2 | 47.1 | 51.0 | 1.6 | 28.9 | 35.4 |
| Other/unknown | 11.9 | 60.0 | 66.2 | 12.4 | 39.8 | 48.2 |
| BMI at delivery, kg/m 2 | ||||||
| <25.0 | 12.9 | 38.0 | 42.1 | 10.3 | 29.3 | 35.3 |
| 25.0-29.9 | 32.7 | 42.2 | 46.2 | 29.7 | 32.4 | 39.9 |
| 30.0-34.9 | 21.6 | 46.5 | 51.8 | 23.1 | 33.7 | 44.2 |
| ≥35.0 | 15.0 | 53.5 | 61.0 | 18.5 | 35.1 | 51.3 |
| Unknown | 17.9 | 34.7 | 38.1 | 18.4 | 26.5 | 33.7 |
| No. of fetuses | ||||||
| Singleton | 99.0 | 43.2 | 47.6 | 97.7 | 32.2 | 41.2 |
| Multiple | 1.0 | 18.1 | 36.1 | 2.3 | 15.6 | 31.6 |
| Previous uterine scar b | 0.6 | 9.5 | 49.0 | 25.0 | 8.3 | 28.8 |
| Gestational age at delivery, wk | ||||||
| <34 | 5.1 | 23.4 | 33.2 | 4.9 | 15.6 | 27.3 |
| 34-36 | 8.2 | 41.6 | 49.9 | 9.5 | 25.6 | 36.8 |
| 37 | 9.0 | 43.0 | 49.2 | 10.9 | 30.2 | 40.2 |
| 38 | 18.5 | 41.9 | 47.3 | 22.7 | 29.6 | 40.5 |
| 39 | 26.7 | 39.8 | 43.7 | 30.6 | 34.3 | 44.6 |
| 40 | 23.4 | 44.0 | 46.0 | 16.3 | 33.7 | 36.9 |
| ≥41 | 9.2 | 63.4 | 65.0 | 5.1 | 50.1 | 53.5 |
| Pregnancy complications | ||||||
| Diabetes c | 4.9 | 52.4 | 64.0 | 6.5 | 35.9 | 56.3 |
| Hypertensive disorder d | 10.6 | 68.1 | 78.6 | 7.2 | 47.1 | 69.3 |
| Fetal anomaly e | 7.9 | 37.4 | 47.4 | 7.6 | 26.6 | 40.1 |
| Hospital type | ||||||
| University-affiliated teaching hospital | 60.0 | 47.6 | 53.6 | 60.5 | 32.5 | 43.1 |
| Teaching community hospital | 36.8 | 36.7 | 39.6 | 35.7 | 30.0 | 37.3 |
| Nonteaching community hospital | 3.2 | 27.4 | 29.2 | 3.8 | 37.4 | 44.9 |
a Denominator excludes prelabor cesarean delivery;
b Includes previous myomectomy in nulliparous women;
c Pregestational or gestational;
d Chronic hypertension, preeclampsia, superimposed preeclampsia, eclampsia, gestational hypertension, unspecified;
Precursors of induction
Indicated precursors of induction were the largest category of precursors in singleton gestations, regardless of gestational age or parity ( Table 2 ). The most common indicated precursor for preterm induction in singleton gestations was hypertensive disease (42.6% in nulliparas and 31.6% in multiparas), followed by fetal (28.7% in nulliparas and 27.3% in multiparas) and maternal (24.7% and 27.1%, respectively) conditions. At term, the prevalence of indicated precursors was reversed, with fetal (20.6% in nulliparas and 12.8% in multiparas) and maternal (14.8% in nulliparas and 16.6% in multiparas) conditions more common than hypertensive disease (14.2% in nulliparas and 8.7% in multiparas). Inductions were elective in 1.2% of preterm nulliparas and 2.3% of preterm multiparas, compared to 15.6% of nulliparas and 25.4% multiparas at term. However, 9.9% of all preterm inductions were classified as elective or no recorded indication. Approximately 95% of pregnancies with the unknown precursors for induction delivered at ≥37 weeks of gestation; 4.3% were at late preterm; and 0.9% were <34 weeks of gestation. Elective induction was the most common individual precursor in multiparas at term.
| Preterm 24-36 wk | Term 37-41 wk | |||
|---|---|---|---|---|
| Precursor | Nulliparous n = 69,796 | Multiparous n = 58,432 | Nulliparous n = 595,585 | Multiparous n = 557,380 |
| With precursor, % | ||||
| Any precursor a | ||||
| PROM | 22.3 | 22.0 | 10.5 | 6.3 |
| Chorioamnionitis | 2.6 | 2.3 | 1.2 | 0.3 |
| Decidual hemorrhage/abruption | 3.7 | 5.4 | 1.5 | 1.9 |
| Hypertensive disease | 42.6 | 31.6 | 14.2 | 8.7 |
| Gestational hypertension | 13.3 | 6.9 | 4.2 | 2.0 |
| Preeclampsia | 14.1 | 7.4 | 4.5 | 2.0 |
| Superimposed preeclampsia | 6.3 | 6.3 | 1.1 | 0.9 |
| Eclampsia | 1.2 | 0.4 | 0.03 | 0.1 |
| Chronic hypertension | 2.3 | 2.4 | 1.1 | 1.3 |
| Unspecified | 5.3 | 8.3 | 3.4 | 2.5 |
| Maternal condition b | 24.7 | 27.1 | 14.8 | 16.6 |
| Diabetes | 11.1 | 11.9 | 5.6 | 7.1 |
| Fetal anomaly | 15.6 | 11.3 | 6.0 | 5.7 |
| Antepartum stillbirth | 4.6 | 5.6 | 0.2 | 0.7 |
| Suspected fetal macrosomia | 0.6 | 0.9 | 2.1 | 2.4 |
| Fetal condition c | 28.7 | 27.3 | 20.6 | 12.8 |
| Maternal fever on admission | 1.5 | 0.4 | 0.6 | 0.2 |
| Admission for fetal reason, not otherwise specified d | 0.2 | 0.6 | 0.6 | 0.6 |
| Admission for maternal reason, not otherwise specified d | 1.4 | 1.1 | 0.6 | 0.6 |
| History of maternal/obstetrical condition e | 0.03 | 0.4 | 0.02 | 0.6 |
| History of fetal condition e | 0.6 | 8.0 | 0.3 | 6.3 |
| Postdates | 0 | 0 | 11.4 | 6.4 |
| Prior uterine scar | 0.2 | 7.6 | 0.1 | 6.3 |
| Total “indicated” | Vaginal delivery, % | 91.2 | 62.8 | 88.9 | 80.3 | 64.5 | 63.7 | 55.9 | 85.5 |
| Elective | Vaginal delivery, % | 1.2 | 92.3 | 2.3 | 95.8 | 15.6 | 76.2 | 25.4 | 97.0 |
| No recorded indication | Vaginal delivery, % | 7.6 | 83.4 | 8.8 | 93.4 | 19.9 | 75.9 | 18.7 | 94.6 |
a Categories for “indicated” precursors can add up to more than total indicated percentage because women could have >1 pregnancy condition;
b Maternal medical problems–percent of women with diabetes is listed;
c Included conditions such as intrauterine growth restriction and abnormal antenatal testing;
d Were included only if there was no other pregnancy condition–these are the only 2 indicated categories that are exclusive of other indications;
e Included pregnancy complications in prior pregnancy (eg, traumatic first delivery or history of fetal demise).
Success of vaginal delivery by precursor for induction
Results for delivery outcomes are also presented in Table 2 . Among nulliparous women, indicated induction was associated with the lowest vaginal delivery rates (62.8% preterm and 63.7% term). At term, elective induction in multiparas was associated with a high vaginal delivery rate of 97% vs 76.2% for nulliparas. Vaginal delivery rates for elective induction were similar to those with no recorded indication. Among nulliparous women, both preterm and term, the proportions still in labor at each cervical dilation were highest for elective induction, those with no recorded indication, and ROM ( Figure ). With respect to precursors of preterm delivery, the largest percentage of cesarean deliveries in the first stage of labor were performed <6-cm dilation, especially for fetal indications and preeclampsia, followed by diabetes ( Figure , A). The rate of cesarean delivery >6 cm decreased modestly in active labor. At term, the largest percentages of cesarean deliveries in the first stage of labor were also performed <6-cm dilation, and occurred more often for fetal indications, diabetes, postdates, and preeclampsia compared to those with elective induction, no recorded indication, or ROM.
