“ When you give birth to a baby, you give birth to a mother as well. ”

Home Classes FAQ Bio Testimonials Photos Doula Services Doula Training Articles/Blog

<< Back


Reduction in c-sections when women given more time!

Obstetrics & Gynecology:
July 2016 - Volume 128 - Issue 1 - p 145–152
doi: 10.1097/AOG.0000000000001488
Contents: Original Research
Reduction in the Cesarean Delivery Rate After Obstetric Care Consensus Guideline Implementation
Wilson-Leedy, Jonas G. MD; DiSilvestro, Alexis J. MD; Repke, John T. MD; Pauli, Jaimey M. MD

Department of Obstetrics and Gynecology, Penn State—Milton S. Hershey Medical Center, Hershey, Pennsylvania.

Corresponding author: Jonas G. Wilson-Leedy, MD, Department of Obstetrics and Gynecology, Penn State—Milton S. Hershey Medical Center, 500 University Drive, Hershey, PA 17033; e-mail: jwilsonleedy@hmc.psu.edu.

Financial Disclosure The authors did not report any potential conflicts of interest.
The authors thank Allen R. Kunselman for his assistance with statistical analysis.

OBJECTIVE: To evaluate the rate of primary cesarean delivery after adopting labor management guidelines.

METHODS: This is a before–after retrospective cohort study at a single academic center. This center adopted guidelines from the Consensus for the Prevention of the Primary Cesarean Delivery. Nulliparous women attempting vaginal delivery with viable, singleton, vertex fetuses were included. For the primary outcome of cesarean delivery rate among induced or augmented patients, 200 consecutive women managed before guideline adoption were compared with 200 similar patients afterward. Secondary outcomes of overall cesarean delivery rate, maternal morbidity, neonatal outcomes, and labor management practices were analyzed with inclusion of intervening spontaneously laboring women.

RESULTS: Between September 13, 2013, and September 28, 2014, 275 women preguideline and 292 postguideline were identified to include 200 deliveries after induction or augmentation each. Among women delivering after induction or augmentation, the cesarean delivery rate decreased from 35.5% to 24.5% (odds ratio [OR] 0.59, 95% confidence interval [CI] 0.38–0.91). The overall cesarean delivery rate decreased from 26.9% to 18.8% (adjusted OR 0.59, CI 0.38–0.92). Composite maternal morbidity was reduced (adjusted OR 0.66, CI 0.46–0.94). The frequency of cesarean delivery documenting arrest of dilation at less than 6 cm decreased from 7.1% to 1.1% postguideline (n=182 and 176 preguideline and postguideline, respectively, P=.006) with no change in other indications.

CONCLUSION: Postguideline, the cesarean delivery rate among nulliparous women attempting vaginal delivery was substantially reduced in association with decreased frequency in the diagnosis of arrest of dilation at less than 6 cm.

During the 1970s the cesarean delivery rate in the United States nearly tripled; the National Institutes of Health Cesarean Task Force identified dystocia as the primary associated diagnosis.1 Despite subsequent attempts of the American College of Obstetricians and Gynecologists (the College) to lower the cesarean delivery through recommendations standardizing diagnosis of arrest of labor, the cesarean delivery rate continued to rise to 32.7% by 2013.2,3 In low-risk pregnancies, cesarean delivery is associated with increased maternal morbidity and mortality and greater financial cost relative to vaginal delivery without proven benefit for the neonate.4,5The cesarean delivery rate among hospitals, health care providers, and regions is highly variable with no association between higher rates and improved neonatal outcomes.5–7 Interpreting available data, the College and the Society for Maternal-Fetal Medicine concluded that a large proportion of primary cesarean deliveries may be medically unindicated.4,8 Diagnoses related to arrest of labor and nonreassuring fetal heart tones were identified as major modifiable contributors. In response, “Obstetric Care Consensus No. 1: Safe Prevention of the Primary Cesarean Delivery” was jointly released by the College and the Society for Maternal-Fetal Medicine in March 2014, encouraging multiple changes in labor management.4 Key recommendations included defining 6 cm dilatation as the threshold for active labor and advocating for longer durations of expectant management before cesarean delivery for labor arrest (Appendix 1, available online at http://links.lww.com/AOG/A815).The primary outcome of the present study was the cesarean delivery rate for nulliparous patients undergoing induction or augmentation after consensus implementation with the central hypothesis that a reduction would be achieved.


We conducted a single-center before and after retrospective cohort study at Penn State—Milton S. Hershey Medical Center relative to adoption of consensus recommendations. The center averages 1,900 deliveries per year. On April 15, 2014, guidelines for labor management consistent with the consensus were adopted as the practice standard at our institution (Policy number 0014-MFM [Appendix 2, available online at http://links.lww.com/AOG/A815]). The associated policy was presented at departmental meetings, circulated to all faculty and residents by e-mail, included among nursing policies, and made available online for reference. Adherence to guidelines was at the discretion of individual health care providers. Of note, no formal guidelines regarding labor management existed at our institution before adoption of the consensus recommendations.The Human Subjects Protection Office of the institutional review board approved the study (STUDY00001687). All data were compiled through retrospective review of electronic records by study authors. To confirm accuracy of data abstracting, tabulated patient characteristics, intrapartum interventions, and outcomes were compared against an electronic database maintained by nursing staff for quality review and a second maintained by resident physicians for patient care. Each of the independent databases contained information regarding maternal comorbidities, receipt of oxytocin or cervical ripening, epidural use, timing of membrane rupture, route of delivery, use of internal monitoring, third- or fourth-degree lacerations, and neonatal outcomes. The resident-maintained database included additional data describing maternal outcomes. When discrepancies were identified, documentation within the patient chart was reviewed and abstracted data were corrected as appropriate.The preguideline period was defined as ending in the month before publication of guidelines on February 28, 2014. The postguideline review period began on May 1, 2014, 2 weeks after policy approval to allow for integration into practice. Electronic birth logs were reviewed to identify eligible patients and we confirmed that the total number of deliveries reported within the logs was identical to the number for which billing was submitted. Nulliparous patients attempting vaginal delivery with a singleton vertex gestation and no history of myomectomy were included. Nonviable pregnancies and those complicated by fetal anomalies were excluded.Our primary outcome consisted of the rate of cesarean delivery among induced or augmented patients. Patients were identified as having been induced or augmented if they received oxytocin or cervical ripening (consisting of mechanical methods or prostaglandins) at any time intrapartum. Patients delivering within these study periods who met inclusion criteria but labored spontaneously without augmentation were included for subsequent analysis of the overall cesarean delivery rate as well as other secondary outcomes.A planned secondary analysis of labor management practices and maternal and neonatal outcomes was completed. Consistent with the consensus, the threshold for active labor was defined as 6 cm dilatation. Given the potential effect of redefinition of active labor on the decision to admit a laboring patient, and our inclusion of augmented patients in evaluating the primary outcome, spontaneously laboring patients were included when evaluating all secondary outcomes. For each cesarean delivery performed, the indication was documented as well as the duration allowed before cesarean delivery for arrest-related diagnoses. Indications for cesarean delivery were categorized as nonreassuring fetal heart tracing, arrest disorder at less than 6 cm dilation, arrest of dilation at 6 cm or greater, arrest of descent, worsening maternal condition, and patient request. Maternal request was recorded if the operative report listed the indication as elective or nonindicated for patients who initially attempted vaginal delivery but subsequently requested cesarean delivery. Cesarean deliveries were classified as having been performed for worsening maternal condition if the indication listed on the operative report was related to maternal status rather than labor progress or fetal monitoring (eg, worsening preeclampsia remote from delivery). If multiple indications were documented for a single cesarean delivery, each was recorded for analysis. For further review of cesarean delivery indications among patients at less than 6 cm dilatation (latent labor), cesarean deliveries performed for arrest disorders below this threshold were categorized as having been performed for either arrest of dilation or failed induction. For this purpose, any cesarean delivery performed at less than 6 cm dilatation that listed arrest of dilation as an indication was categorized as such.Secondary maternal outcomes included chorioamnionitis, endometritis, maternal birth injury, postpartum hemorrhage, and receipt of transfusion. Chorioamnionitis and endometritis were documented if these diagnoses were recorded in the patient chart. Maternal birth injury consisted of third- or fourth-degree perineal lacerations or hysterotomy extension. Postpartum hemorrhage was recorded if estimated blood loss was 500 mL or greater for a vaginal delivery or 1,000 mL or greater for cesarean delivery, if uterotonics other than oxytocin were administered, or if a delayed postpartum hemorrhage was recorded in the patient record. Our composite maternal outcome was defined to include any of the aforementioned secondary outcomes. Secondary neonatal outcomes included shoulder dystocia, birth injury, 5-minute Apgar score less than 5, umbilical artery pH less than 7.0, umbilical artery base deficit greater than 12, and receipt of antibiotics for greater than 48 hours. Neonatal birth injury was defined as cephalohematoma, fracture, intracranial hemorrhage, or brachial plexus injury associated with delivery. The composite neonatal outcome included all secondary neonatal outcomes.A sample size of 200 patients undergoing labor induction or augmentation was chosen for our primary outcome such that this investigation was powered to detect a 10% absolute decrease in the primary outcome of cesarean delivery rate from 25% at baseline to 15% postguideline with α error=5% and 80% power using a one-sided analysis. An estimate of the preguideline cesarean delivery rate for nulliparous women was used for the baseline, and a 10% decrease was based on the clinically significant nature of this difference as well as the predicted effect of specific labor management guidelines. A one-sided analysis was used with the expectation that delayed performance of cesarean delivery would not increase the cesarean delivery rate. The initial preguideline and postguideline study periods were restricted relative to the length of time required to include 200 consecutive deliveries after induction or augmentation. The sample for secondary outcomes consisted of consecutive induced or augmented patients as well as intervening nulliparous patients who met inclusion criteria but labored spontaneously for total cohorts of 275 patients preguideline and 292 patients postguideline.Statistical analysis was performed using the software R. Categorical data were evaluated with χ2 test or Fisher exact test when the expected cell count was less than five. Continuous variables were compared using two-sample t tests or Wilcoxon rank-sum test if the data were not normally distributed. Multivariable logistic regression was performed and adjusted odds ratios calculated to determine the independent influence of postguideline status on outcomes of interest. Principle exposures of interest were included initially and forced to remain in all models regardless of significance. For cesarean delivery rate and composite maternal morbidity, these minimum covariates consisted of induction or augmentation status, neonatal weight, gestational age at delivery, and maternal body mass index. For the neonatal composite only, neonatal weight and gestational age at delivery were required for model inclusion. Additional covariates were considered for model inclusion if a relationship (P<.05) was identified on bivariate analysis. Final models were selected based on backward elimination with covariates removed when P>.10. All analyses were two-sided and a probability value <.05 was considered significant.


After defining the study periods to include samples of 200 consecutive nulliparous patients undergoing induction or augmentation, the preguideline data collection period spanned from September 13, 2013, to February 28, 2014, and the postguideline period from May 1, 2014, to September 28, 2014. These date ranges included 799 and 805 total deliveries, respectively, with patients excluded as outlined in Appendix 3, available online at http://links.lww.com/AOG/A815. Cohorts for secondary outcomes included both induced or augmented and spontaneously laboring nulliparous patients with 275 patients preguideline and 292 patients postguideline. There were no differences in comorbidities, demographics, cervical status at admission, internal monitoring, or maternal or fetal characteristics for the entire sample of nulliparous patients or for the subset of patients who were induced or augmented (Table 1; Appendix 4 [Appendix 4 available online at http://links.lww.com/AOG/A815]). For the entire sample of nulliparous patients, the gestational age at delivery ranged from 29 6/7 weeks to 42 weeks (data not shown).

For the subset of patients who were induced or augmented, the cesarean delivery rate was significantly reduced from 71 of 200 (35.5%) to 49 of 200 (24.5%) with odds ratio 0.59 and 95% confidence interval 0.38–0.91 (Appendix 5, available online at http://links.lww.com/AOG/A815). Secondary outcomes were evaluated for all nulliparous women including those who labored spontaneously (Table 2). The reduction in cesarean delivery rate remained significant when spontaneously laboring patients were included. Performance of cesarean delivery for labor arrest was significantly reduced (Table 2). No significant differences in the frequency of other indications were identified. No significant differences in neonatal outcomes were apparent. Rates of postpartum hemorrhage and composite maternal morbidity were significantly reduced.

Multivariable logistic regression was performed to account for potential confounders and determine the independent influence of postguideline status on outcomes of interest for the entire cohort including spontaneously laboring patients (Table 3). Reductions in the cesarean delivery rate and maternal composite morbidity remained significant after adjustment. Composite neonatal morbidity was not significantly different after accounting for the influence of potentially confounding variables.

The duration of expectant management, proportions of patients with prolonged expectant management, and cesarean delivery rate were compared within subsets of patients who were induced or augmented in latent labor, augmented for arrest of dilation in active labor, or reached the second stage (Table 4). For patients induced or augmented at less than 6 cm dilatation, the cesarean delivery rate resulting from arrest-related indications was significantly reduced and the frequency with which arrest of dilation was diagnosed decreased significantly. No significant differences were noted in the frequencies of cesarean delivery for failed induction, arrest of dilation at 6 cm or greater, or arrest of descent. Durations of expectant management for arrest of dilation at 6 cm or greater and within second-stage labor were unchanged.

The proportion of patients induced or augmented for at least 18 hours after membrane rupture at less than 6 cm dilatation increased postguideline. There were no differences in proportions induced or augmented for at least 12 hours. Similar calculations were performed to compare the duration of expectant management at 4 cm dilated and less than 80% effaced or less than 4 cm dilated regardless of effacement. In the preguideline period, rates of continued expectant management below this cutoff were 14 of 113 (12.4%) and 6 of 113 (5.3%) at 12 and 18 hours respectively, and were not different postguideline at 13 of 109 (11.9%) and 6 of 109 (5.5%) (data not shown, P=1.00 and 1.00, respectively).Management of patients before cesarean delivery performed solely for arrest-related indications was compared (Table 5). No statistically significant differences in management of patients before cesarean delivery in latent or active labor were identified. The average duration of second-stage labor before diagnosis of arrest of descent was significantly increased postguideline as was the proportion of patients with a second stage of at least 4 hours.


For nulliparous patients undergoing induction or augmentation, the cesarean delivery rate decreased significantly subsequent to adoption of consensus recommendations. Significance was maintained with inclusion of patients laboring without induction or augmentation and after correction for multiple predictors. The overall cesarean delivery rate fell from 26.9% to 18.8% after guideline implementation, well below the goal of 23.9% set by Healthy People 2020 for nulliparous patients with singleton term pregnancy in vertex presentation.9In conjunction with the reduced cesarean delivery rate, we identified decreases in rates of postpartum hemorrhage and composite maternal morbidity. The reduced rate of composite maternal morbidity is well aligned with known associations between cesarean delivery and adverse maternal outcomes and appears to have been primarily driven by a reduction in postpartum hemorrhage.10,11 Although we did not identify significant increases in maternal or neonatal morbidities, our investigation was insufficiently powered to asses these outcomes. The absence of a significant increase in neonatal morbidity would be consistent with available data suggesting no increase in adverse neonatal outcomes with prolonged expectant management and population-level studies suggesting no improvement in neonatal outcomes as the cesarean delivery rate increases above 10%.4,8,12,13Our data suggest that a significant decrease in cesarean deliveries for labor arrest at less than 6 cm dilatation contributed to the reduction in cesarean delivery rate. Consistent with recommendations discouraging diagnosis of arrest of dilation in latent labor, a sixfold reduction in this indication was identified postguideline. Adoption of altered strategies for management of latent labor in the postguideline period was further suggested by a nearly threefold increase in the proportion of patients induced or augmented for a minimum of 18 hours after membrane rupture.Of note, data supporting delaying cesarean delivery for failed induction until a minimum of 12–18 hours after membrane rupture in latent labor were derived from investigations using definitions of active labor beginning at 4 cm and 80–90% effaced or 5 cm regardless of effacement.14–16 Using a comparable definition, proportions of patients expectantly managed for 12 or 18 hours after membrane rupture were no different postguideline and were similar or in excess to investigations on which the consensus recommendations were based.15,16 This suggests that the observed increase in the proportion of patients induced or augmented for at least 18 hours at less than 6 cm dilatation resulted from avoidance of cesarean delivery for arrest of dilation rather than adoption of longer durations of expectant management in general.There were multiple strengths to this analysis. Where previous data have evaluated the cesarean delivery rate as well as maternal and neonatal outcomes in relation to prolonged durations of latent, active, and second-stage labor or compared protocol adherent and nonadherent patients, direct comparison with an alternate practice standard was lacking.2,14–21 The present work suggests that adoption of consensus recommendations for latent labor can be expected to substantially reduce the cesarean delivery rate with decreased composite maternal morbidity.The retrospective nature of this study and absence of data characterizing guideline adherence represent significant limitations, which preclude conclusive attribution of the observed reduction in the cesarean delivery rate to specific changes in practice. Although our findings suggest that adoption of 6 cm as the threshold for active labor holds promise for reducing the cesarean delivery rate, risk may have been shifted from latent to active or second-stage labor given that available data suggest protracted latent labor may be associated with an increased risk of cesarean delivery at a later phase or stage.14,22 It is possible that the increase in average duration of the second stage before diagnosis of arrest of descent may have resulted in a statistically significant reduction in cesarean delivery in second-stage labor with a larger sample size. Similarly, our investigation was insufficiently powered to comment definitively on the influence of postguideline practices on secondary maternal and neonatal outcomes. As a result of our inclusion of all gestational ages, we are unable to comment on the effect of consensus recommendations on the cesarean delivery rate for term compared with preterm pregnancies. The homogeneity of our patient cohort and restriction to nulliparous women may limit generalization to more diverse populations.In conclusion, this investigation confirms that adoption of consensus guidelines was associated with a reduction in the primary cesarean delivery rate among nulliparous patients at our institution. The observed decrease is both statistically and clinically significant. Although underpowered to fully assess secondary outcomes, we did not identify any association with increased rates of adverse maternal or neonatal outcomes.

<< Back