Abstract
We examined the interaction between race and labor induction in cesarean delivery in a cohort of 600,000 deliveries in the Cerner Health Facts database. Black women had higher likelihood cesarean (28.9 vs. 26.5%) and lower likelihood of induction of labor at delivery compared to white women (27.2 vs. 32.5%). Induction modified the association between race and cesarean—Black women (odds ratio=1.36, 95% confidence interval 1.30, 1.43) who were induced had significantly increased odds of cesarean delivery.
Introduction
Cesarean delivery (CD), the surgical delivery of a fetus through incisions in the abdominal or uterine walls, is the most common inpatient surgery in the United States. One in three deliveries occur via CD. The proportion of births delivered via CD has increased dramatically since the 1990s with over 1.2 million CD in 2018. CD is a necessary and beneficial procedure when indicated, but the U.S. rates of CD greatly exceed the estimated threshold of 10–20% that optimally balances risks with maternal and neonatal outcomes. Excess CDs increase concerns about unnecessary maternal (infection, repeat CD, severe morbidity, and maternal mortality) and neonatal (laceration, respiratory morbidity, and shoulder dystocia) complications.1,2
Accompanying this increase in the rate of CD in the U.S. has been the emergence of racial disparities in the frequency of CD in Black and non-Hispanic white (NHW) women. Of note, the rate of CD has increased 61% among Black mothers compared to 33% among NHW women since 1994. In 2018 the overall CD rate was 31.9 percent of deliveries with 30.8 percent of deliveries to white mothers via cesarean compared with 36.1 percent among Black mothers.3 This pattern was mirrored in women without a history of cesarean (21.4% vs. 25.2%) and in low risk pregnancies (singleton, primiparous, full term pregnancies with infant in head down position; 24.9% vs. 30.3%). Black mothers had rates of overall, primary, and low-risk CD 17.2 to 21.7% higher than white mothers. When stratified by age, Black women had rates of primary (11.3 to 29.8%) and low-risk (18.2 to 45.1%) CD higher than white women.
During this same time period, specific labor management practices, including the use of labor induction, increased markedly— between 1990 and 2015 rates of medical induction have more than doubled (9.5% to 23.8%). As with CD, Black women have had much steeper rates of increase of induction of labor (IOL) than non-Hispanic white women (93% vs. 43% since 1995). Further, Black women undergoing induction may have greater risk of cesarean; Black nulliparous women who have been induced have a rate of cesarean 32% higher than their peers who were not induced.
To better understand the intersections between race, induction of labor, and cesarean delivery, we have examined the extent to which induction of labor explained risk of cesarean delivery among Black and NHW women in a cohort of 600,000 women in the Cerner Health Facts database. We estimated the odds of cesarean delivery for Black women by receipt of induction of labor and type of induction.
Methods
Data for this analysis were obtained from Cerner Health Facts® database, a data warehouse aggregating electronic health records from participating Cerner client inpatient and outpatient facilities (Cerner Corporation, North Kansas City, Missouri). Health Facts uses an automated electronic medical record system to capture hospital procedures, laboratory testing and results, diagnostic information, demographics, medical history, and inpatient and outpatient encounters over time. This instance of Health Facts contains 63 million unique patients from over 600 U.S. healthcare facilities that have a data use agreement in place. Data include comprehensive clinical records from inpatient and outpatient encounters and includes data from pharmacy, laboratory, registration, and billing data. Cerner Corporation has established Health Insurance Portability and Accountability Act-compliant operating policies to establish de-identification for Health Facts. Health Facts research has been reviewed and approved by the Institutional Review Board at the University of Missouri, Kansas City (IRB# 14-567).
Inpatient stays with the delivery of a pregnancy documented in Health Facts data between 2000 and 2014 were identified. This analysis reports on deliveries of 606,435 Black and NHW women between 2000 and 2014 from inpatient facilities. For consistency when relying on validated definitions of clinical outcomes utilizing International Classification of Disease, 9th Edition-Clinical Modification (ICD9-CM) codes, this analysis excluded deliveries in 2015 or later. Deliveries were excluded if they were multiple gestations (N=56,776), extreme maternal ages (<15 years or > 45; N=7,205) or other pregnancy related-conditions (hydatiform mole, ectopic, abnormal products of conception or abortion; N=2018). The final analytic sample included 562,050 deliveries. The mean maternal age for deliveries was 27.7 years with a standard deviation ±6 years. Patients contributed from 1 to 11 pregnancies (mean=1.1, median=1).
Identification of Deliveries
Inpatient deliveries were identified using an enhanced definition developed by Kuklina and colleagues that incorporates ICD-9CM codes for delivery outcome (V27) as well as relevant diagnosis-related groups and procedures. ICD-9CM codes included in this algorithm are summarized in Table 1. This definition has been validated and determined to accurately identify 99% of deliveries.4
Table 1.
International Classification of Disease-9 CM Codes used to define key variable
| Concept | Code(s) | |
|---|---|---|
| Any delivery4 | Outcome of delivery | V27 |
| Normal delivery | 650 | |
| Forceps | 720, 721,721.1, 721.9, 723.1, 723.9, 724, 726 | |
| Breech extraction | 725.1, 725.2, 725.3, 725.4 | |
| Vacuum extraction | 727.1, 727.9 | |
| Other delivery or extraction | 728, 729, 732.2, 735.9 | |
| Episiotomy | 736 | |
| Cesarean section | 740, 741, 742, 744.9 | |
| Exclusions: hydatiform mole (630), abnormal products of conception (631), ectopic pregnancy (633), abortion (632, 634, 635, 636, 638, 639, 69.01, 69.51, 74.91, 75.0); Validation: Identifies approximately 99.4% deliveries | ||
|
| ||
| Any induction5,6 | Medical | 73.4 |
| Rupture of membranes | 73.0, 73.01, 73.09 | |
| Surgical | 73.1 | |
| Validation: Positive predictive value: 97%; negative predictive value 75% | ||
|
| ||
| Bateman comorbidity index*7 | Alcohol abuse | 291, 303, 305 |
| Cardiac valvular disease | 394, 395, 396, 397, 424 | |
| Chronic congestive heart failure | 428.22, 428.23, 428.33, 428.42, 428.43 | |
| Chronic ischemic heart disease | 412, 413, 414 | |
| Chronic renal disease | 581, 582, 583, 585, 587, 588, 646.2 | |
| Congenital heart disease | 745, 746, 747.1–747.4, 648.5 | |
| Drug abuse | 304, 305.2–305.9, 648.3 | |
| Gestational hypertension | 642.3 (excluding other hypertension) | |
| HIV | 042 | |
| Mild/unspecified preeclampsia | 642.4, 642.7 | |
| Placenta previa | 641.1, 641.1 | |
| Pre-existing diabetes | 250, 648.0 | |
| Pre-existing hypertension | 401, 402, 403, 404, 405, 642.0–642.2, 642.7 | |
| Pulmonary hypertension | 416.0, 416.8, 416.9 | |
| Severe preeclampsia/eclampsia | 642.5, 642.6 | |
| Sickle cell disease | 282.4, 282.6 | |
| Systemic lupus erythematosus | 710.0 | |
| *Bateman index also includes definitions for maternal asthma, multiple gestation, advanced maternal age, and repeat cesarean provided above | ||
|
| ||
| Indications for cesarean delivery8,9 | Malpresentation; Antepartum bleeding; Herpes infection; Hypertension; Multiple gestation; Macrosomia; Unengaged fetal head; Soft tissue condition; Uterine scar; Preterm gestation; Congenital fetal anomaly; Renal, liver, or thyroid abnormality; hypotension; Diabetes; Substance use, mental health; Heart disease; Isoimmunization; Fetal demise, growth restriction; Oligohydramnios, Polyhydramnios; Rupture of membranes; Chorioamnionitis; Infection; Maternal age; Uterine rupture; Pulmonary embolism; Cerebral hemorrhage; Asthma | 652 (excluding 652.1 and 652.5); 641, 656.0; 647.6; 642, 642.5, 642.6; 651; 656.6; 652.5; 654.0, 654.1, 654.4, 654.5, 654.6, 654.7; 654.9; 644.2; 655.0, 655.1; 646.2, 646.7, 648.1; 669.1, 669.2; 648.0, 648.8; 648.3, 648.4; 648.5, 648.6; 656.1, 656.2; 656.4, 656.5; 658.0, 657; 658.2; 658.4; 659.2, 659.3; 659.5, 659.6; 665.0, 665.1; 673.0, 673.1, 673.3. 673.8; 431, 432, 433, 434; 493 |
Definition of Variables
Induction of labor was defined according to the algorithms established by Levine and colleagues5 and Yasmeen and colleagues.6 Induction variables included documentation of codes for rupture of membranes, medical or surgical induction, and number of induction modalities documented. Burden of maternal illness was summarized using the Bateman comorbidity index,7 a measure that incorporates 20 pregnancy-related comorbidities and maternal age. Modeled after the Charlson and Elixhauser comorbidity indices, the Bateman index demonstrates improved discriminative ability compared to these measures with the endpoints maternal death/endorgan damage or maternal intensive care unit stay. Indications for primary and repeat CD indications were based on definitions provided by Gregory and colleagues8 and augmented by data from Henry et al.9 In all, we examined 33 potential stated indications that range from malpresentation, macrosomia, and antenatal bleeding to advanced maternal age, maternal asthma, prior CD and mental health diagnoses. Other covariates included year of delivery, maternal age, payer (no insurance, self-pay, private, Medicaid or Medicare/other public), teaching facility (yes/no), urban hospital, number of inpatient beds (0–5, 6–99, 100–199, 200–299, 300–499, 500, or more), and Census region (Midwest, Northeast, South, West).
Statistical Analyses
Categorical data were summarized using column and row percents. Continuous data were summarized using mean and standard deviation or median and interquartile range as appropriate. Estimates of relative risk for race and induction were estimated from a logistic regression. Due to concerns that about bias in estimates of the odds ratio, we conducted sensitivity analyses using Poisson regression with robust error variance to confirm that absence of bias.10,11 Estimates were adjusted for year of delivery, maternal age, hospital size, Census region, teaching hospital, indication for CD and Bateman comorbidity index score. Additive interaction was assessed in the relative risk scale using PROC NLMIXED to estimate the relative excess risk due to interaction (RERI).12,13 Analyses were conducted in SAS version 9.4 (SAS Corporation, Cary, NC).
Results
Of the 562,050 deliveries included in this analysis, 20.2% (N=113,674) were among non-Hispanic Black women. Most deliveries (54%) were in women between the ages of 25 and 34 (Table 2). One quarter of pregnancies were supported by Medicaid and 37% by private health insurance plans. Sixty-seven percent of deliveries occurred at teaching hospitals, 64% at medium sized hospitals (100–499 beds), and 85% were at urban hospitals. Black mothers were on average younger, more likely to report Medicare or other public health insurance, living in the South, sought care at a large hospital, and had higher comorbidity scores than NHW mothers.
Table 2.
Demographic, Hospital and Clinical Characteristics by Race and Cesarean Delivery
| Race | ||||||
|---|---|---|---|---|---|---|
| Overall | Black | White | ||||
| N | (Col %) | N | (Row %) | N | (Row %) | |
| All | 562,050 | (100.0) | 113,674 | (20.2) | 448,376 | (79.8) |
| Age group | ||||||
| 15–24 | 181,772 | (32.3) | 54,066 | (29.7) | 127,706 | (70.3) |
| 25–34 | 301,385 | (53.6) | 47,712 | (15.8) | 253,673 | (84.2) |
| 35 & older | 78,893 | (14.0) | 11,896 | (15.1) | 66,997 | (84.9) |
| Health insurance | ||||||
| Medicaid | 131,932 | (23.5) | 34,734 | (26.3) | 97,198 | (73.7) |
| Medicare/other public | 15,743 | (2.8) | 7,281 | (46.2) | 8,462 | (53.8) |
| No private or public | 199,382 | (35.5) | 42,434 | (21.3) | 156,948 | (78.7) |
| Private/Commercial | 205,360 | (36.5) | 27,821 | (13.5) | 177,539 | (86.5) |
| Self-pay | 9,633 | (1.7) | 1,404 | (14.6) | 8,229 | (85.4) |
| Teaching hospital | ||||||
| Missing | 493 | (0.1) | 94 | (19.1) | 399 | (80.9) |
| No | 185,551 | (33.0) | 21,581 | (11.6) | 163,970 | (88.4) |
| Yes | 376,006 | (66.9) | 91,999 | (24.5) | 284,007 | (75.5) |
| Bed size | ||||||
| 0 to 5 beds | 21,993 | (3.9) | 682 | (3.1) | 21,311 | (96.9) |
| 6–99 | 69,967 | (12.4) | 4,754 | (6.8) | 65,213 | (93.2) |
| 100–199 | 90,365 | (16.1) | 12,036 | (13.3) | 78,329 | (86.7) |
| 200–299 | 146,239 | (26.0) | 36,581 | (25.0) | 109,658 | (75.0) |
| 300–499 | 125,501 | (22.3) | 26,500 | (21.1) | 99,001 | (78.9) |
| 500+ | 107,985 | (19.2) | 33,121 | (30.7) | 74,864 | (69.3) |
| Census region | ||||||
| Midwest | 132,681 | (23.6) | 25,812 | (19.5) | 106,869 | (80.5) |
| Northeast | 211,560 | (37.6) | 33,787 | (16.0) | 177,773 | (84.0) |
| South | 167,359 | (29.8) | 51,741 | (30.9) | 115,618 | (69.1) |
| West | 50,450 | (9.0) | 2,334 | (4.6) | 48,116 | (95.4) |
| Failed induction | 7,444 | (1.3) | 1,897 | (25.5) | 5,547 | (74.5) |
| Vaginal birth after cesarean | 31,119 | (5.5) | 7,039 | (22.6) | 24,080 | (77.4) |
| Prior cesarean delivery | 81,590 | (14.5) | 17,554 | (21.5) | 64,036 | (78.5) |
| Indication for cesarean Delivery | 268,676 | (47.8) | 55,559 | (20.7) | 213,117 | (79.3) |
| Indications for CD (no.) | ||||||
| 0 indications | 293,374 | (52.2) | 58,115 | (19.8) | 235,259 | (80.2) |
| 1 to 2 | 231,538 | (41.2) | 46,420 | (20.0) | 185,118 | (80.0) |
| 3 or more | 11,525 | (2.1) | 3,228 | (28.0) | 8,297 | (72.0) |
| Missing | 25,613 | (4.6) | 5,911 | (23.1) | 19,702 | (76.9) |
| Bateman comorbid conditions (no.) | ||||||
| 0 conditions | 397,040 | (70.6) | 74,346 | (18.7) | 322,694 | (81.3) |
| 1–2 | 135,838 | (24.2) | 29,235 | (21.5) | 106,603 | (78.5) |
| 3–4 | 19,427 | (3.5) | 6,620 | (34.1) | 12,807 | (65.9) |
| 5 or more | 9,745 | (1.7) | 3,473 | (35.6) | 6,272 | (64.4) |
Over the 15 years of observation of Health Facts data, the proportion of cesarean deliveries was higher in Black women compared to white (28.9% vs. 26.5; See Table 3). In contrast, rates of labor induction were lower (27.2% vs. 32.5; OR = 0.77, 95% CI 0.75–0.78) among Black mothers. This finding was consistent for rupture of membranes and medical induction, while the rates of surgical induction were higher in Black women compared to white (1.6% vs. 1.0). Black women were younger, had higher Bateman comorbidity scores and were more likely to deliver at teaching and large hospitals, factors associated with lower rates of induction.
Table 3.
Frequency of Cesarean Delivery and Labor Induction by Race
| Black | White | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| By Race | Cesarean Delivery | Cesarean Delivery | ||||||||
| All Deliveries | Black | White | No | Yes | No | Yes | ||||
| N | (%) | N | (%) | N | (%) | (Row %) | (Row %) | (Row %) | (Row %) | |
| Cesarean delivery | ||||||||||
| No | 410,377 | (73.0) | 80,824 | (71.1) | 329,553 | (73.5) | -- | -- | -- | -- |
| Yes | 151,673 | (27.0) | 32,850 | (28.9) | 118,823 | (26.5) | -- | -- | -- | -- |
| Labor induction | ||||||||||
| No | 385,527 | (68.6) | 82,718 | (72.8) | 302,809 | (67.5) | (68.4) | (31.6) | (68.4) | (31.6) |
| Yes | 176,523 | (31.4) | 30,956 | (27.2) | 145,567 | (32.5) | (78.3) | (21.7) | (84.2) | (15.8) |
| Induction - rupture of membranes | ||||||||||
| No | 433,056 | (77.0) | 92,614 | (81.5) | 340,442 | (75.9) | (68.6) | (31.4) | (69.3) | (30.7) |
| Yes | 128,994 | (23.0) | 21,060 | (18.5) | 107,934 | (24.1) | (82.0) | (18.0) | (86.9) | (13.1) |
| Induction - medical | ||||||||||
| No | 476,302 | (84.7) | 99,049 | (87.1) | 377,253 | (84.1) | (70.6) | (29.4) | (72.0) | (28.0) |
| Yes | 85,748 | (15.3) | 14,625 | (12.9) | 71,123 | (15.9) | (74.4) | (25.6) | (81.5) | (18.5) |
| Induction - surgical | ||||||||||
| No | 555,667 | (98.9) | 111,846 | (98.4) | 443,821 | (99.0) | (71.2) | (28.8) | (73.5) | (26.5) |
| Yes | 6,383 | (1.1) | 1,828 | (1.6) | 4,555 | (1.0) | (62.5) | (37.5) | (69.6) | (30.4) |
Racial differences in CD for Black women were observed only in the presence of labor induction (see Figure 1). Among women without documentation of labor induction, the odds of CD for Black women was slightly lower than that of white women (OR =0.89, 95% CI 0.87, 0.91) after adjustment for calendar year, maternal age, maternal health conditions (Bateman comorbidity index, CD indication) and facility characteristics (hospital size, location and teaching status). Among women with any method of induction, the odds of CD for Black women was significantly higher than for white women (OR = 1.36, 95% CI 1.30, 1.43). The interaction between race and induction (any induction or type of induction) was statistically significant. The relative excess risk due to interaction - or the proportion of CD among Black women that were induced attributable to their interaction - was 20% (95% CI 17%, 22%). The odds of CD among Black women with rupture of membranes and another form of induction was 1.72 to 2.14 times higher than white women.
Figure 1.
Adjusted Odds Ratios* of Cesarean Delivery for Black Women Compared to White Stratified by Labor Induction and Method of Induction
*Estimates adjusted for year of delivery, maternal age, hospital size, Census region, teaching hospital, indication for CD and Bateman comorbidity index score
Discussion
Black women had higher odds of cesarean delivery compared to white women. However, the likelihood of CD was modified by receipt of induction of labor. Black women were less likely to receive induction of labor at delivery compared to white women. In particular, Black women had lower rates of medical induction and rupture of membranes. Among women with documented labor induction, the odds of CD was considerably higher for Black women.
Of note, rates of labor induction in this cohort of deliveries identified via electronic health records were lower than estimates from the Eunice Kennedy Shriver National Institute of Child Health and Human Development supported Consortium on Safe Labor (CSF) conducted between from 2002 to 2008. CSF found rates of IOL higher as high as 42.7% overall with the rate of induction higher among Black women (44.6% vs. 42.3%) than non-Hispanic white women.14 However, estimates from HF are consistent with estimates from vital statistics data with respect to the overall rate of induction and the higher rate among white women.3,15
Professional guidelines published in 2014 provided guidance for the use of induction with the objective of reduced CD.1,2 IOL may reduce the rate of cesarean delivery compared to expectant management depending on gestational age and the use of cervical ripening agents. However, these guidelines recommend a minimum of 24 hours for latent labor before declaration of failed induction. Since the publication of these guidelines, evidence has been inconsistent - some facilities have found a decrease in CD,16,17 but this has not been consistent across institutions.18 A recent study examined whether a standardized labor protocol reduced the rates of cesarean delivery among Black and found a significant decrease in cesarean delivery rate in black women in the induction protocol group compared with the observation group (25.7% vs 34.2%, P=.02).19
Data from 4,000 deliveries found no differences in the length of first stage of labor for Black women overall,20 but differences in body mass index21 and in the length of the second stage may contribute to CD.22 A recent study found that among women undergoing induction of labor, Black women had higher rates of dissatisfaction with the birth process compared to white women after adjustment for mode of delivery (cesarean or vaginal) and duration of delivery.23 Specifically, Black women reported higher rates that the labor was excessively long and lower rates of preparedness for labor. Attanasio and colleagues found that Black women reported lower rates of shared decision-making which was associated with both induction and CD.24
The strengths of this analysis include the large cohort of deliveries with geographic variability and multiple years of observation—Health Facts (HF) deliveries represent diverse patient populations with ample representation of urban and under-served patients (Table 1). Further, HF includes detailed information available on procedures, diagnoses, and patient, facility and clinical event characteristics at the encounter level. Finally, this analysis incorporated validated definitions for key variables (CD, IOL) and cohort identification.
Limitations include lack of data or validated algorithm to measure gestational age at delivery, maternal gravidity and parity, and reason for cesarean delivery. Also, HF may under-represent obstetric diagnoses and procedures of the female reproductive tract. While Health Facts discharges was similar to the Nationwide Inpatient Sample—a nationally representative sample of hospital discharges—in terms of geographic distribution, length of stay, patient sex; there were significantly fewer but differences in diagnoses specific to pregnancy and childbirth and procedures on the female reproductive tract.25 The underrepresentation of pregnancies may be attributable to variability in reporting common to electronic health record data as well as to variability in participating hospitals.26 For example, in our sample, more than 80% of deliveries occurred in urban hospitals, which may have higher CD rates.27
Conclusion
Disparities in cesarean delivery among Black women were significant among women with labor induction in this cohort. Additional studies of the predictors of CD are needed in light of recent recommendations on labor management following induction and adherence to guidelines that fully weighs risks and benefits to both mother and child.
Footnotes
Xi Wang, MS, David Walsh, and Jenifer E. Allsworth, PhD, (above), are in the Department of Biomedical and Health Informatics, University of Missouri-Kansas City School of Medicine, Kansas City, Missouri.
Disclosure
This work was supported by a CTSA grant from NCATS awarded to the University of Kansas for Frontiers: University of Kansas Clinical and Translational Science Institute (# UL1TR002366) The contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH or NCATS. Cerner Health Facts data are hosted at UMKC at the Center for Health Insights (CHI) in the Department of Biomedical and Health Informatics (DBHI).
References
- 1.Spong CY, Berghella V, Wenstrom KD, Mercer BM, Saade GR. Preventing the first cesarean delivery: summary of a joint Eunice Kennedy Shriver National Institute of Child Health and Human Development Society for Maternal-Fetal Medicine, and American College of Obstetricians and Gynecologists workshop. In reply. Obstet Gynecol. 2013 Mar;121(3):687. doi: 10.1097/AOG.0b013e3182854b36. [DOI] [PubMed] [Google Scholar]
- 2.Gynecologists ACoOa, Medicine SfM-F. Obstetric care consensus (1): safe prevention of the primary cesarean delivery. Obstet Gynecol. 2014 Mar;123(3):693–711. doi: 10.1097/01.AOG.0000444441.04111.1d. [DOI] [PubMed] [Google Scholar]
- 3.Martin JA, Hamilton BE, Osterman MJK, Driscoll AK. Births: Final Data for 2018. Natl Vital Stat Rep. 2019 Nov;68(13):1–47. [PubMed] [Google Scholar]
- 4.Kuklina EV, Whiteman MK, Hillis SD, et al. An enhanced method for identifying obstetric deliveries: implications for estimating maternal morbidity. Matern Child Health J. 2008 Jul;12(4):469–77. doi: 10.1007/s10995-007-0256-6. [DOI] [PubMed] [Google Scholar]
- 5.Levine LD, Limaye M, Srinivas SK. The utility of ICD9-CM codes in identifying induction of labor. Am J Perinatol. 2015 Apr;32(5):475–80. doi: 10.1055/s-0034-1390351. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Yasmeen S, Romano PS, Schembri ME, Keyzer JM, Gilbert WM. Accuracy of obstetric diagnoses and procedures in hospital discharge data. Am J Obstet Gynecol. 2006 Apr;194(4):992–1001. doi: 10.1016/j.ajog.2005.08.058. [DOI] [PubMed] [Google Scholar]
- 7.Bateman BT, Mhyre JM, Hernandez-Diaz S, et al. Development of a comorbidity index for use in obstetric patients. Obstet Gynecol. 2013 Nov;122(5):957–65. doi: 10.1097/AOG.0b013e3182a603bb. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Gregory KD, Korst LM, Gornbein JA, Platt LD. Using administrative data to identify indications for elective primary cesarean delivery. Health Serv Res. 2002 Oct;37(5):1387–401. doi: 10.1111/1475-6773.10762. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Henry OA, Gregory KD, Hobel CJ, Platt LD. Using ICD-9 codes to identify indications for primary and repeat cesarean sections: agreement with clinical records. Am J Public Health. 1995 Aug;85(8 Pt 1):1143–6. doi: 10.2105/ajph.85.8_pt_1.1143. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.McNutt LA, Wu C, Xue X, Hafner JP. Estimating the relative risk in cohort studies and clinical trials of common outcomes. Am J Epidemiol. 2003 May;157(10):940–3. doi: 10.1093/aje/kwg074. [DOI] [PubMed] [Google Scholar]
- 11.Zhang J, Yu KF. What’s the relative risk? A method of correcting the odds ratio in cohort studies of common outcomes. JAMA. 1998 Nov;280(19):1690–1. doi: 10.1001/jama.280.19.1690. [DOI] [PubMed] [Google Scholar]
- 12.Knol MJ, VanderWeele TJ, Groenwold RH, Klungel OH, Rovers MM, Grobbee DE. Estimating measures of interaction on an additive scale for preventive exposures. Eur J Epidemiol. 2011 Jun;26(6):433–8. doi: 10.1007/s10654-011-9554-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Richardson DB, Kaufman JS. Estimation of the relative excess risk due to interaction and associated confidence bounds. Am J Epidemiol. 2009 Mar;169(6):756–60. doi: 10.1093/aje/kwn411. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Singh J, Reddy UM, Huang CC, Driggers RW, Landy HJ, Grantz KL. Racial/Ethnic Differences in Labor Induction in a Contemporary US Cohort: A Retrospective Cohort Study. Am J Perinatol. 2018 Mar;35(4):361–368. doi: 10.1055/s-0037-1607285. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Martin JA, Hamilton BE, Osterman MJK, Driscoll AK, Drake P. Births: Final Data for 2016. Natl Vital Stat Rep. 2018 Jan;67(1):1–55. [PubMed] [Google Scholar]
- 16.Wilson-Leedy JG, DiSilvestro AJ, Repke JT, Pauli JM. Reduction in the Cesarean Delivery Rate After Obstetric Care Consensus Guideline Implementation. Obstet Gynecol. 2016 Jul;128(1):145–52. doi: 10.1097/AOG.0000000000001488. [DOI] [PubMed] [Google Scholar]
- 17.Bell AD, Joy S, Gullo S, Higgins R, Stevenson E. Implementing a Systematic Approach to Reduce Cesarean Birth Rates in Nulliparous Women. Obstet Gynecol. 2017 Nov;130(5):1082–1089. doi: 10.1097/AOG.0000000000002263. [DOI] [PubMed] [Google Scholar]
- 18.Rosenbloom JI, Stout MJ, Tuuli MG, et al. New labor management guidelines and changes in cesarean delivery patterns. Am J Obstet Gynecol. 2017 Dec;217(6):689.e1–689.e8. doi: 10.1016/j.ajog.2017.10.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Hamm RF, Srinivas SK, Levine LD. A standardized labor induction protocol: impact on racial disparities in obstetrical outcomes. Am J Obstet Gynecol MFM. 2020 Aug;2(3):100148. doi: 10.1016/j.ajogmf.2020.100148. [DOI] [PubMed] [Google Scholar]
- 20.Tuuli MG, Odibo AO, Caughey AB, Roehl K, Macones GA, Cahill AG. Are there differences in the first stage of labor between Black and White women? Am J Perinatol. 2015 Feb;32(3):233–8. doi: 10.1055/s-0034-1382254. [DOI] [PubMed] [Google Scholar]
- 21.Norman SM, Tuuli MG, Odibo AO, Caughey AB, Roehl KA, Cahill AG. The effects of obesity on the first stage of labor. Obstet Gynecol. 2012 Jul;120(1):130–5. doi: 10.1097/AOG.0b013e318259589c. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Greenberg MB, Cheng YW, Hopkins LM, Stotland NE, Bryant AS, Caughey AB. Are there ethnic differences in the length of labor? Am J Obstet Gynecol. 2006 Sep;195(3):743–8. doi: 10.1016/j.ajog.2006.06.016. [DOI] [PubMed] [Google Scholar]
- 23.Hamm RF, Srinivas SK, Levine LD. Risk factors and racial disparities related to low maternal birth satisfaction with labor induction: a prospective, cohort study. BMC Pregnancy Childbirth. 2019 Dec;19(1):530. doi: 10.1186/s12884-019-2658-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Attanasio LB, Kozhimannil KB, Kjerulff KH. Factors influencing women’s perceptions of shared decision making during labor and delivery: Results from a large-scale cohort study of first childbirth. Patient Educ Couns. 2018 Jun;101(6):1130–1136. doi: 10.1016/j.pec.2018.01.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.DeShazo JP, Hoffman MA. A comparison of a multistate inpatient EHR database to the HCUP Nationwide Inpatient Sample. BMC Health Serv Res. 2015 Sep;15:384. doi: 10.1186/s12913-015-1025-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Parsons A, McCullough C, Wang J, Shih S. Validity of electronic health record-derived quality measurement for performance monitoring. J Am Med Inform Assoc. 2012 Jul-Aug;19(4):604–9. doi: 10.1136/amiajnl-2011-000557. 2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Kozhimannil KB, Law MR, Virnig BA. Cesarean delivery rates vary tenfold among US hospitals; reducing variation may address quality and cost issues. Health Aff (Millwood) 2013 Mar;32(3):527–35. doi: 10.1377/hlthaff.2012.1030. [DOI] [PMC free article] [PubMed] [Google Scholar]