Abstract
Background:
Nationally, rates of cesarean birth are highest among Black patients, compared with other racial/ethnic groups. These observed inequities are a relatively new phenomenon (in the 1980s, cesarean birth rates among Black patients were lower than average) indicating an opportunity to narrow the gap. Cesarean birth rates vary greatly among hospitals, masking racial disparities that are unseen when rates are reported in aggregate.
Objective:
This study explored reasons for the current large Black-White disparity in first-birth cesarean rates by first examining the hospital-level variation in first-birth cesarean rates among different racial/ethnic groups. We then identified hospitals that had low first-birth cesarean rates among Black patients and compared them to hospitals with high rates. We sought to identify differences in facility or patient characteristics could provide insights for the racial disparity.
Study Design:
A population cross-sectional study was performed on 1,267,493 California live births from 2018 through 2020 using birth certificate data linked with maternal patient discharge records. Annual nulliparous term singleton vertex cesarean delivery (first-birth) rates were calculated for the most common racial/ethnic groups statewide and for each hospital. Self-identified race/ethnicity categories as selected on the birth certificate were used. Relative risk (RR) and 95% confidence intervals for first-birth cesarean comparing 2019 to 2015 were estimated using a log-binomial model for each racial/ethnic group. Patient and hospital characteristics were compared between hospitals that had first-birth cesarean rates <23.9% for Black patients compared with hospitals with rates ≥23.9% for Black patients.
Results:
Hospitals with at least 30 nulliparous term singleton vertex Asian, Black, Hispanic and White patients each were identified. Black patients had a very different distribution with significantly higher rate (28.4%), wider standard deviation (7.1) and interquartile range (6.5) than other racial groups (P<0.01). 29 hospitals with a low first-birth cesarean rate among Black patients were identified using the Healthy People 2020 target of 23.9% and compared to 106 hospitals with higher rates. The low group has a cesarean rate of 19.9% compared to 30.7% in the higher group. There were no significant differences between the groups in hospital characteristics (ownership, delivery volume, Neonatal level, proportion of midwife deliveries) or patient characteristics (age, education, insurance, onset of prenatal care, BMI, hypertension, diabetes). Among the 106 hospitals that did not meet the target for Black patients, 63 met it for white patients with a mean rate of 21.4%. In the same hospitals the mean rate for Black patients was 29.5%. Cesarean indications among Black patients in the group that did not meet the 23.9% target were significantly higher for all indications: labor dystocia, fetal concern (spontaneous labor), and no labor (e.g. macrosomia), all indications with a high degree of subjectivity.
Conclusion:
The statewide cesarean rate for Black patients is significantly higher and has substantially greater hospital variation than other racial or ethnic groups. The lack of difference in facility or patient characteristics between hospitals with low cesarean rates among Black patients and those with high rates suggests unconscious bias and structural racism potentially play important roles in creating these racial differences.
Keywords: cesarean delivery, disparity, hospital variation, Black, Asian, Hispanic, Nulliparous Term Singleton Vertex
Introduction
U.S. cesarean birth rates have risen by 50% since the mid-1990’s.1,2,3 Cesarean birth rates among Black patients were lower than white patients in the 1980’s but this difference reversed over time beginning in the mid 1990’s and accelerating in the 2010’s with Black patients currently having a 6 percentage point higher rate than white patients for both total cesarean birth rate and nulliparous term singleton vertex (NTSV, low-risk first birth) cesarean birth rate.3 Higher rates of cesarean birth have been associated with greater risk of immediate morbidities and future pregnancy complications4,5 and are likely a contributor to the well-publicized disparities in severe maternal morbidity and mortality among Black patients.6,7,8 Furthermore, a recent study found that Black patients having cesarean birth had even higher risk adjusted rates of maternal complications than for White patients having a cesaran birth.9 Prior studies attempted to explain Black-white cesarean disparities by focusing on patient characteristics such as education and age and comorbidities such as obesity and hypertensive disorders but after risk adjustment for these factors, significant inequalities remain unexplained.10,11,12,13
Hospital-level variation in NTSV cesarean birth rates is striking with rates ranging 4-6 fold and only minimally explained by patient and hospital characteristics.14,15 Both the rise and variation in cesareans rates appear to be driven by indications that can develop during labor--failure to progress in labor and concerning fetal heart rate patterns, both of which are prone to physician subjectivity.16,17 In the setting of large variation, an important quality improvement strategy has been to recognize high-performing hospitals and identify best practices to share with other facilities.18,19 Objectives of this study are first to analyze hospital variation by race for NTSV cesarean rates among 238 California hospitals and then to compare hospitals with low NTSV cesarean rates for Black patients to hospitals with high NTSV cesarean rates for Black patients to identify patient, hospital, or patient care characteristics that may account for the disparities.
Materials and Methods
We conducted a cross-sectional study of 1,267,493 deliveries that occurred between January 1, 2018 to December 31, 2020 in 238 California hospitals with maternity services. Births at military hospitals, birthing centers and at home were not included. The California Maternal Quality Care Collaborative (CMQCC) obtained patient discharge records for all females aged 8-60 years in California hospitals from the California Department of Health Care Access and Information (HCAI), and de-identified birth certificate records from the California Department of Public Health-Vital Records. Records were linked using a previously validated probabilistic linkage algorithm with linkage rates exceeding 98%.20,21 Institutional review board approval was obtained from Stanford University as the study host, and the California Committee for the Protection of Human Subjects for the use of state data sets.
The primary outcome measure for this study is cesarean rate among nulliparous patients at term, with a singleton fetus in a vertex position (NTSV cesarean). Following the Joint Commission (JC) PC-02 measure specifications22, we identified 432,595 NTSV patients (flowchart in Figure 1). Race/ethnicity of the birthing person was calculated based on self-reported race and Hispanic origin status from birth certificates. Four categories utilized for this study included: Hispanic, non-Hispanic Asian (Asian), non-Hispanic Black (Black), and non-Hispanic white (white). We did not include Pacific Islander, Native American and Other race categories as too few hospitals had sufficient NTSV births of these groups to analyze at the hospital level. Unknown or missing race/ethnicity (4.8%) were also excluded.
Fig 1.
Study flow diagram
We then examined whether the NTSV cesarean birth rate among Black patients in each hospital met the national Healthy People 2020 target of 23.9%, and compared hospital- and individual-level characteristics among hospitals that met the target for Black patients to those that did not meet the target. We chose 23.9% as it was both the national HP2020 target, and the target used in the recent statewide Supporting Vaginal Collaborative resulting in over 60% of hospitals meeting the target. However, we also performed a sensitivity analysis examining other cut-points including 22.9%, 25%, and 26%. We considered the following hospital characteristics: percent of Medicaid deliveries in NTSV births, percent of Certified Nurse-Midwives (CNMs) attendance among all vaginal births, percent of Black NTSV patients among all NTSV patients, hospital ownership, hospital delivery volume, and neonatal level of care. No data was available for the utilization of hospitalists and furthermore their involvement in labor management varies greatly from facility to facility. Medicaid deliveries were identified from the patient discharge record and, if missing, from the birth certificate. CNM attendance was identified using the license number of the delivery provider on the birth certificate. Hospital ownership was retrieved from annual hospital utilization and financial reports from HCAI. Level of neonatal care was derived as a self-reported variable based on the 2012 definition from the American Academy of Pediatrics.23 Patient-level characteristics included maternal demographic and clinical data. Education, prenatal care onset, and pre-pregnancy BMI were obtained from birth certificate. Age at delivery was calculated from discharge records. Comorbidities, including preeclampsia, chronic hypertension, diabetes, and gestational diabetes were identified using ICD-10-CM diagnosis codes from discharge data. Cesarean indications were divided into 6 indication groups (spontaneous labor with labor dystocia, spontaneous labor with fetal intolerance, induced labor with labor dystocia, induced labor with fetal intolerance, no labor, and others) using ICD-10-CM diagnosis and procedure codes from discharge records (all codes available in supplemental materials).
Statistical Analyses
We conducted descriptive analysis of mean, standard deviation, median, interquartile range (IQR), and range to assess the variation of hospital NTSV cesarean rate among Asian, Black, Hispanic, and white populations. We restricted the analysis to hospitals with at least 30 NTSV births over 3 years for each racial/ethnic group (30 was chosen as a minimum denominator to maintain an expected numerator of at least 6). We used Levene’s test and Brown-Forsythe test to examine whether variance of NTSV cesarean rates were different across racial/ethnic groups.24 We used white population as a comparison group as hospital variation was smallest among white patients (SD=5.1, IQR=4.9%).
We categorized hospitals into four groups based on their Black and white NTSV cesarean rate and compared their frequency of hospital- and individual-level characteristics: 1) hospitals meeting the Healthy People 2020 NTSV cesarean target (<23.9%) for both groups, 2) hospitals meeting the target for Black but not white patients, 3) hospitals meeting the target for white but not Black patients, and 4) hospitals not meeting the target for both Black and white groups.
To investigate the potential association between a hospital’s characteristics and the success of meeting the NTSV cesarean rate target, we performed robust poisson regression analysis to estimate the relative risk (RR) and 95% confidence interval (CI) for each hospital- and aggregated individual-level characteristic. This allowed us to compare characteristics of hospitals that did not meet the target for Black patients with those that did meet the target. We also examined indications for cesarean births among Black patients in hospitals that did not meet the NTSV cesarean birth rate target and compared to those in hospitals that met the target using multinomial logistic regression to estimate odds ratio (OR) and 95% CI. Multinomial logistic regression is an extension of binary logistic regression that allows for simultaneous evaluation of more than two categories of the outcome variable. We evaluated both crude and adjusted associations and estimated adjusted relative risk (aRR) and adjusted odds ratio (aOR) through multivariable analysis that included all characteristics. All hypothesis testing was 2-sided, with a prespecified significance threshold of p<0.05 and SAS Enterprise Guide 8.3 (SAS Institute Inc., Cary, NC, USA) was used for statistical analysis.
Results
In 2018-2020, the racial and ethnic composition of California NTSV patients included: Hispanic 43.4%, white 31.5%; Asian 18.9%, and Black 5.3%. Hospital-level NTSV Cesarean birth rates for white, Black, Asian and Hispanic patients were compared using histograms and descriptive statistics (Figure 2). There was substantial variation among hospitals for all race/ethnicity categories; hospital NTSV cesarean birth rates for Black patients had the highest mean and median and the greatest variation (Levene’s and Brown-Forsythe tests both P<0.01) with the widest and flattest histogram. For each race and ethnicity there was a large proportion of hospitals that met the national Healthy People 2020 target of <23.9%. 63.5% of hospitals met the target for Hispanic patients, 59.8% of hospitals met the target for White patients, and 45.8% of hospitals met the target for Asian patients. However, only 21.5% of hospitals met the national target for Black patients.
Figure 2.
Histograms and descriptive statistics for hospital NTSV CD rates for White, Black, Asian, and Hispanic gravidas in 2018-2020. Data was restricted to hospitals with at least 30 NTSV births within each race/ethnic group. P-value from Levene’s test (P=0.01) and Brown-Forsythe test (P < 0.01) suggested that hospital variance of NTSV cesarean rates were different across the groups.
Table 1 compares hospital-level characteristics for hospitals that met the national target for Black patients versus hospitals that did not. A secondary stratification asks whether hospitals met the target separately or together for Black and White patients. As expected, NTSV cesarean birth rates were markedly different among hospitals that met the target versus those that did not (19.9% vs. 30.7%, p<0.01). Of note, the NTSV cesarean birth rate was much closer for White patients between the two groups (19.7% vs 23.6%) and between Black and White patients in the hospitals meeting the national target for Black patients (19.9% vs 19.7%). 64 hospitals met the target for white patients but did not for Black patients. The mean NTSV cesarean rates were markedly different in these hospitals: 21.4% for white patients and 29.5% for Black patients.
Table 1.
Hospital characteristics, stratified by whether the 2018-2020 NTSV CD rate met the national Healthy People 2020 target of 23.9% (among 135 hospitals with at least 30 NTSV births among both Black and White gravidas)
| Hospitals Meeting Target for Black Gravidas | Hospitals NOT Meeting Target for Black Gravidas | |||||
|---|---|---|---|---|---|---|
|
|
||||||
| Hospitals meeting target for both Black and White gravidas | Hospitals meeting target for Black but not White gravidas | Total | Hospitals meeting target for White but not Black gravidas | Hospitals not meeting target for both Black and White gravidas | Total | |
| Number of hospitals | 25 | 4 | >29 | 63 | 43 | 106 |
|
| ||||||
| Average number of Black NTSV Births | 149.0 | 66.3 | 137.6 | 143.4 | 171.2 | 154.7 |
|
| ||||||
| Average number of White NTSV Births | 611.0 | 744.3 | 629.4 | 942.0 | 778.4 | 875.7 |
|
| ||||||
| Hospital level rates (2018-2020) | Mean (SD) | Mean (SD) | Mean (SD) | Mean (SD) | Mean (SD) | Mean (SD) |
|
| ||||||
| Average NTSV CD rate for Black gravidas | 20.0 (3.3) | 19.4 (4.7) | 19.9 (3.4) | 29.5 (4.3) | 32.4 (5.9) | 30.7 (5.2) |
| Average NTSV CD rate for White gravidas | 18.7 (3.3) | 25.7 (1.3) | 19.7 (3.9) | 21.4 (1.7) | 26.8 (3.3) | 23.6 (3.6) |
| Average proportion of NTSV births to Black gravidas to all NTSV births | 8.1 (6.3) | 4.5 (4.4) | 7.6 (6.1) | 5.7 (4.6) | 6.8 (5.6) | 6.2 (5.0) |
| Hospital Characteristics | N (%) | N (%) | N (%) | N (%) | N (%) | N (%) |
|
| ||||||
| >50% Medicaid in NTSV deliveries | 8 (32.0) | 1 (25.0) | 9 (31.0) | 25 (39.7) | 23 (53.5) | 48 (45.3) |
| Significant Midwifery presence (>20% of vaginal births attended by CNM in years 2018-2020) | 12 (50.0) | 0 (0.0) | 12 (41.4) | 20 (31.7) | 6 (14.0) | 26 (24.5) |
| Hospital Ownership | ||||||
| Private nonprofit | 12 (48.0) | 3 (75.0) | 15 (51.7) | 30 (47.6) | 29 (67.4) | 59 (55.7) |
| Private investor | 2 (8.0) | 0 (0.0) | 2 (6.9) | 5 (7.9) | 3 (7.0) | 8 (7.5) |
| Public: university, city, county | 4 (16.0) | 1 (25.0) | 5 (17.2) | 11 (17.5) | 6 (14.0) | 17 (16.0) |
| Integrated Health System | 7 (28.0) | 0 (0.0) | 7 (24.1) | 17 (27.0) | 5 (11.6) | 22 (20.8) |
| Average annual Delivery Volume | ||||||
| <1,500 births | 10 (40.0) | 1 (25.0) | 11 (37.9) | 9 (14.3) | 7 (16.3) | 16 (15.1) |
| 1,500-2,499 births | 8 (32.0) | 3 (75.0) | 11 (37.9) | 24 (38.1) | 11 (25.6) | 35 (33.0) |
| >=2500 births | 7 (28.0) | 0 (0.0) | 7 (24.1) | 30 (47.6) | 25 (58.1) | 55 (51.9) |
| Neonatal Level of care | ||||||
| 1 (Well newborn) | 6 (24.0) | 0 (0.0) | 6 (20.7) | 3 (4.8) | 1 (2.3) | 4 (3.8) |
| 2 (Special care) | 13 (52.0) | 2 (50.0) | 15 (51.7) | 11 (17.5) | 8 (18.6) | 19 (17.9) |
| 3 (NICU) | 6 (24.0) | 2 (50.0) | 8 (27.6) | 38 (60.3) | 28 (65.1) | 66 (62.3) |
| 4 (Regional NICU) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 11 (17.5) | 6 (14.0) | 17 (16.0) |
Table 2 compares patient-level characteristics among Black patients for hospitals that met the national target for Black patients versus those that did not and includes a secondary stratification noting whether hospitals met the target separately or together for Black and white patients. Age, pre-pregnancy BMI, education, insurance status, onset of prenatal care, and presence of hypertensive or diabetic comorbidities for Black patients were all similar between hospitals with high versus low NTSV cesarean birth rates among Black patients.
Table 2.
Patient characteristics in Black gravidas, by whether the 2018-2020 NTSV cesarean birth rate met the national Healthy People 2020 target of 23.9% (among 135 hospitals with at least 30 NTSV births among both Black and white gravidas)
| Hospitals Meeting Target for Black Gravidas | Hospitals NOT Meeting Target for Black Gravidas | |||||
|---|---|---|---|---|---|---|
|
|
||||||
| Hospitals meeting target for both Black and White gravidas | Hospitals meeting target for Black but not White gravidas | Total | Hospitals meeting target for White but not Black gravidas | Hospitals not meeting target for both Black and White gravidas | Total | |
| Number of Black gravidas | 3,725 | 265 | 3,990 | 9,037 | 7,362 | 16,399 |
|
| ||||||
| Patient level rates | Mean (SD) | Mean (SD) | Mean (SD) | Mean (SD) | Mean (SD) | Mean (SD) |
|
| ||||||
| Age at Delivery (years) | 25.9 (5.8) | 27.2 (6.6) | 26.0 (5.8) | 26.6 (6.0) | 26.0 (6.0) | 26.4 (6.0) |
| Pre-Pregnancy BMI (kg/m2) | 28.1 (7.3) | 27.0 (6.5) | 28.1 (7.3) | 27.9 (7.4) | 28.0 (7.3) | 27.9 (7.3) |
| Patient level characteristics | N (%) | N (%) | N (%) | N (%) | N (%) | N (%) |
|
| ||||||
| Education | ||||||
| High school/GED or less | 1,488 (39.9) | 89 (33.6) | 1,577 (39.5) | 3,483 (38.5) | 2,830 (38.4) | 6,313 (38.5) |
| Some college | 1,258 (33.8) | 80 (30.2) | 1,338 (33.5) | 2,736 (30.3) | 2,569 (34.9) | 5,305 (32.3) |
| College grad or more | 759 (20.4) | 82 (30.9) | 841 (21.1) | 2,399 (26.5) | 1,645 (22.3) | 4,044 (24.7) |
| Missing | 220 (5.9) | 14 (5.3) | 234 (5.9) | 419 (4.6) | 318 (4.3) | 737 (4.5) |
| Medi-Cal/other government-sponsored insurance as the primary payment for delivery | 1,988 (53.4) | 149 (56.2) | 2,137 (53.6) | 4,426 (49.0) | 4,244 (57.6) | 8,670 (52.9) |
| Prenatal care started at the first trimester | 3,052 (81.9) | 232 (87.5) | 3,284 (82.3) | 7,604 (84.1) | 6,024 (81.8) | 13628 (83.1) |
| Comorbidities | ||||||
| Preeclampsia or chronic hypertension | 935 (25.1) | 48 (18.1) | 983 (24.6) | 2,166 (24.0) | 1,692 (23.0) | 3,858 (23.5) |
| Preexisting or gestational diabetes | 282 (7.6) | 21 (7.9) | 303 (7.6) | 690 (7.6) | 486 (6.6) | 1,176 (7.2) |
Table 3 provides crude and adjusted relative risks for hospital and patient-level risk characteristics. After adjustment for the hospital and individual-level risk factors shown, hospitals with AAP NICU levels III & IV were more likely to not meet the target for Black patients than hospitals with NICU levels I & II. This is despite no difference between these hospital groups for any of the risk factors (rates of older age, obesity, hypertension, diabetes, low education and prenatal care). Hospitals with higher NTSV cesarean birth rate in white patients were also more likely to not meet the target for Black patients. However, the association between white NTSV cesarean rate and meeting the target for Black patients were only seen among level I & II hospitals. Table 4 illustrates crude and adjusted odds ratios for cesarean delivery indications among Black patients delivering at low and high NTSV rate hospitals. Black patients in hospitals that did not meet the national target had substantially and significantly higher NTSV cesarean rates for labor dystocia, fetal concern with spontaneous labor, as well as for no labor cesarean (largely suspected macrosomia).
Table 3.
Relative risk for the rate of hospital- and aggregated individual-level characteristics comparing hospitals NOT meeting the national NTSV cesarean birth rate target for Black gravidas to hospitals meeting the target*
| AAP NICU level | |||
|---|---|---|---|
|
| |||
| All Hospitals | Level I & II | Level III & IV | |
|
| |||
| Hospital-level factors and aggregated individual-level characteristics | aRR† (95% CI) | aRR† (95% CI) | aRR† (95% CI) |
| Hospital-level characteristics | |||
| >50% Medicaid in NTSV deliveries [yes vs no] | 1.30 (0.97, 1.73) | 2.21 (0.94, 5.21) | 1.24 (0.98, 1.57) |
| >20% of vaginal births attended by CNM [yes vs no] | 0.97 (0.72, 1.30) | 1.27 (0.53, 3.06) | 0.84 (0.64, 1.09) |
| Hospital Ownership [Public or Integrated Health System vs Private] | 1.03 (0.81, 1.32) | 1.43 (0.36, 5.78) | 1.08 (0.86, 1.35) |
| Average Annual Delivery Volume, 2018-2020 [ref= < 1,500] | |||
| 1,500-2,499 births | 1.23 (0.88, 1.73) | 0.96 (0.42, 2.22) | 1.33 (0.98, 1.81) |
| >=2500 births | 1.18 (0.86, 1.62) | 1.04 (0.44, 2.42) | 1.18 (0.88, 1.58) |
| AAP NICU level [Level III&IV vs Level I & II] | 1.56 (1.18, 2.05) | -- | -- |
| NTSV cesarean rate in White gravidas | 1.17 (1.06, 1.30) ‡ | 2.39 (1.47, 3.87) ‡ | 1.06 (0.97, 1.16)‡ |
| % of Black gravidas among all NTSV births | 1.00 (0.91, 1.10)‡ | 0.79 (0.53, 1.19)‡ | 1.03 (0.98, 1.10)‡ |
| 1.04 (0.96, 1.13)‡ | 1.13 (0.76, 1.69)‡ | 1.01 (0.95, 1.07)‡ | |
|
| |||
| Aggregated individual-level characteristics | |||
| % of Black NTSV gravidas age 35 and older | |||
| % of Black NTSV gravidas with pre-pregnancy BMI 30 and higher | 1.06 (0.97, 1.15)‡ | 1.01 (0.78, 1.30)‡ | 1.08 (1.00, 1.16)‡ |
| % of Black NTSV gravidas with high school or less education | 1.00 (0.95, 1.05)‡ | 1.09 (0.95, 1.26)‡ | 0.96 (0.92, 0.99)‡ |
| % of Black NTSV gravidas started prenatal care at the first trimester | 1.02 (0.97, 1.07)‡ | 1.01 (0.86, 1.18)‡ | 1.00 (0.95, 1.05)‡ |
| % of Black NTSV gravidas with preeclampsia or chronic hypertension | 1.00 (0.95, 1.05)‡ | 1.06 (0.76, 1.47)‡ | 1.00 (0.96, 1.03)‡ |
| % of Black NTSV gravidas with Preexisting or gestational diabetes | 0.96 (0.83, 1.10)‡ | 0.92 (0.45, 1.88)‡ | 1.02 (0.92, 1.14)‡ |
Analysis restricted to 135 hospitals with at least 30 NTSV births among both Black and white gravidas. Hospitals not meeting the target for Black gravidas were the outcome referent.
Adjusted RR and 95% CI estimated using robust poisson regression model including all the hospital-level factors and aggregated individual-level characteristics listed in the Table.
Risk of cesarean relative to 5 percentage point increase for each characteristic
Table 4.
Crude and adjusted odds of NTSV cesarean birth (CD) indication using vaginal birth as the outcome referent, comparing Black gravidas in hospitals that met the Healthy People 2020 target rate of 23.9% for Black gravidas to those in hospitals that did not meet the target*
| Hospital Group | ||||
|---|---|---|---|---|
| Met Target for Black | NOT Meet Target for Black Gravidas | Odds Ratios of CD by indication | ||
| Number of NTSV Black gravidas | 3,990 | 16,399 | ||
| NTSV CD indications | N (% CD) | N (%CD) | cOR† (95% CI) | aOR† (95% CI) |
|
| ||||
| Labor Dystocia (Spontaneous labor) | 175 (4.4) | 999 (6.1) | 1.54 (1.31, 1.82) | 1.62 (1.34, 1.96) |
| Labor Dystocia (Induced labor) | 242 (6.1) | 1,251 (7.6) | 1.40 (1.21, 1.61) | 1.36 (1.15, 1.63) |
| Fetal Concern (Spontaneous labor) | 197 (4.9) | 1,180 (7.2) | 1.62 (1.39, 1.89) | 1.70 (1.42, 2.04) |
| Fetal Concern (Induced labor) | 140 (3.5) | 596 (3.6) | 1.15 (0.95, 1.39) | 1.13 (0.90, 1.42) |
| No labor (e.g. macrosomia) | 76 (1.9) | 627 (3.8) | 2.23 (1.75, 2.84) | 2.32 (1.76, 3.07) |
| Other indications (e.g. abruption) | 26 (0.7) | 157 (1.0) | 1.63 (1.08, 2.48) | 1.44 (0.89, 2.34) |
Analysis restricted to 135 hospitals with at least 30 NTSV births among both Black and white gravidas.
Crude and adjusted OR and 95% CI estimated from multinomial logistic regression model. Adjusted model included the following factors: 1) Individual-level maternal demographic and clinical factors: age at delivery, pre-pregnancy BMI, education levels, insurance, prenatal care onset, and presence of comorbidities including preeclampsia or chronic hypertension, and preexisting or gestational diabetes; and 2) Hospital-level characteristics: >20% of vaginal births attended by CNM, hospital ownership, average annual delivery volume, and AAP NICU level.
Considering characteristics in hospitals with NTSV cesarean rate in Black patients fall around the cutoff point of 23.9% may be similar and may mask the difference between hospitals with higher and lower rates, we conducted sensitivity analysis excluding hospitals with rates between 22.9% and 24.9%, and the results were similar (Supplementary Table s3 and s4). We also conducted sensitivity analysis using 25.0% and 26.0% as the target cutoff points. The difference of the fetal concern with induced labor rate between hospitals meeting and not meeting the target was larger when using higher rate as the cutoff point, while the difference of the rate of the other CD indications became smaller (Supplementary Table s4).
Comment
Principal Findings:
There was large variation in hospital NTSV cesarean birth rates among all racial and ethnic groups with the greatest variation among Black patients. Despite the population-level NTSV cesarean birth rate being well above the 23.9% target for Black patients, 22% of hospitals were able to meet this goal for their Black patients. This led us to explore if these hospitals with low NTSV cesarean birth rates for their Black patients had different patient or hospital characteristics than those with high NTSV cesarean birth rates for Black patients. No significant differences between the two groups of hospitals were observed. There were university affilaited, community-based, public-system, and integrated health system hospitals that acheived low NTSV Cesarean birth rates for their Black patients while others from the same categories had high rates. Futhermore, hospitals with low NTSV cesarean rates for their Black patients had significantly lower rates for all major cesarean indications: labor dystocia, fetal concern, and cesarean without labor.
Results in the Context of What is Known:
These findings suggest that the driver for higher NTSV cesarean birth rate among Black patients does not lie with patient or hospital characteristics. This is consistent with other studies examining hospital and provider variation in cesarean birth rates-which indicate that a substantial portion of NTSV cesarean birth rate is driven by unit culture and clinician attitudes.25,26,27 It is now well documented that the large variation among hospitals appears to be spread among the major primary cesarean indications: fetal intolerance of labor and failure to progress in labor.16,17 Each involves clinical decisions with high degrees of subjectivity with the potential for high provider variation. Several prior studies of racial disparities compared cesarean indications between Black and white patients and the indication of fetal concern was the most disparete.28,29 A recent multi-hospital study within the Maternal-Fetal Medicine Unit Network identified that the main drivers for higher NTSV cesarean birth rates among Black patients, when compared with white patients, were both fetal intolerance of labor and labor dystocia.30 We also noted a significant disparity among patients with no-labor cesareans, albeit at a lower absolute rate. We pose that these relatively subjective indications are opportunities for unconscious bias to be activated. Our study provides a different perspective by comparing Black patients in hosptials with low rates to Black patients in hospitals with high rates of NTSC cesarean births.
Clinical Implications:
It is very concerning that 62 “high-performing” hospitals that met the national target with a mean NTSV cesarean birth rate of 21.4% for their white patients had strikingly higher average NTSV cesarean birth rate of 29.5% for their Black patients. In conversations with several of these hospitals, it was disclosed that the facilities had never analyzed their cesarean birth rates stratified by race and ethnicity and were surprised by their disparities. This highlights important opportunities for monitoring and quality improvement actions. As an example of a recent quality improvement project, Hamm and colleagues studied the introduction of a standardized induction protocol, and noted a significant reduction in the cesarean rate in Black patients managed with the induction protocol when compared with Black patients in the non-protocol observational group.31 This approach undescores the importance of standardizing care to overcome the subjectivity and unconscious bias that can interfere with care.
These results provide support to reject the “mother blame” narrative (attributing outcomes to patient characteristics) for higher cesarean birth rates among Black patients. This narrative neglects to address key factors that impact perinatal care including equity, quality of care and social determinants of health.32 Besides implicit bias, there are other reasons hospital culture may be supportive or obstructive for Black patients during labor. Key elements in a successful birth culture include labor support, shared decision making, and lack of microaggressions. Many Black patients identify these elements lacking in their birth experience which leads to mistrust in the healthcare system.33 In perinatal care, the history of structural and systemic racism has led to Black women in the U.S. having the highest rates of adverse maternal outcomes. Drivers of health disparities in the U.S. are rooted in “racism not race.”34 There is a gap in rigorous literature naming the impact of racism and provider practices on maternal outcomes, and without this explicit discussion, dismantling disparities in health outcomes cannot be achieved and clinicians are not held accountable.35 More policy implementation needs to be done to improve cultural competence and reduce healthcare provider bias through training and education and a push for a more racially diverse clinical workforce.36
Research Implications:
The extreme variation noted in Black NTSV cesarean birth rates strongly supports the call for a quality measure that stratifies NTSV cesarean birth rates by race at the hospital level.37 It can be accomplished in all states as all required data elements are present on the birth certificate including self-identified race/ethnicity. By examining disparities at the local level, hospitals can determine how to focus their quality improvement efforts. The current report indicates that disparities are not uniform among hospitals, and some have managed to eliminate them completely, potentially providing lessons for others. Examining hospital variation in rates in addition to population-level rates may be a more useful approach to understanding racial disparities. A formal qualitative analysis of culture and attitudes at high disparity versus low disparity hospitals is planned but was not performed for this report.
Strengths and Limitations:
Strengths of this study include large numbers of hospitals and large numbers of patients of multiple racial/ethnic identities. While administrative data was used, key data elements (cesarean delivery and race and ethnicity) were of high quality. The successful California Supporting Vaginal Birth Initiative showed that NTSV CD rates are not fixed and all racial groups have potential for reduction. Limitations include the relative lack of racial concentration among California hospitals where few hospitals have more than 25% Black patients (none have more than 35%) and the relatively low average number of Black NTSV births per hospital (151). While many potential drivers for disparity were controlled for, other explanations are possible and would require larger sample sizes.
Conclusions:
The central finding of this report is that many hospitals were able to meet the national NTSV cesarean birth target for their Black patients even though the state average for Black patients was still significantly elevated above that target. Furthermore, hospitals with low rates for Black patients were not different from hospitals with high rates by institutional or patient characteristics, suggesting that all hospitals should be able to meet NTSV cesarean targets for Black patients.
Supplementary Material
Tweetable Statement:
Hospitals with low cesarean birth rates among Black patients have patient and hospital characteristics no different from hospitals with high rates, stressing the importance of care practices in disparities.
AJOG at a glance:
Why was this study conducted?
Cesarean birth rates among Black patients have risen much faster than other racial/ethnic groups in the last 15 years and the reason is poorly understood.
Key Findings:
Analysis of nulliparous term singleton vertex cesarean rates for 238 California hospitals over 3 years reveals substantially higher rates and significantly greater hospital variation for Black patients compared to other racial groups. 29 hospitals were identified with consistent cesarean rates for Black patients under the Healthy People 2020 target of 23.9%. These hospitals included examples from all healthcare practice environments. Patient characteristics of these hospitals did not differ from facilities with high cesarean rates for Black patients.
What does this add to what is known?
These findings indicate that low rates for Black patients can be achieved in all types of birthing facilities and suggests important roles for unit culture and unconscious bias.
Funding Support acknowledgements:
California Health Care Foundation NIH R01 NR020335
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
The authors report no conflicts of interest.
Citations
- 1.Osterman MJK, Martin JA. Trends in low-risk cesarean delivery in the United States, 1990–2013. National vital statistics reports; vol 63 no 6. Hyattsville, MD: National Center for Health Statistics. 2014. [PubMed] [Google Scholar]
- 2.MacDorman MF, Menacker F, Declercq E. Cesarean birth in the United States: epidemiology, trends, and outcomes. Clin Perinatol 2008;35:293–307. doi: 10.1016/j.clp.2008.03.007 [DOI] [PubMed] [Google Scholar]
- 3.Martin JA, Hamilton BE, Osterman MJK, Driscoll AK. Births: Final Data for 2019. National Vital Statistics Reports; vol 70 no 2. Hyattsville, MD: National Center for Health Statistics. 2021. doi: 10.15620/cesareanc:100472 [DOI] [PubMed] [Google Scholar]
- 4.Miller ES, Hahn K, Grobman WA; Society for Maternal-Fetal Medicine Health Policy Committee. Consequences of a primary elective cesarean delivery across the reproductive life. Obstet Gynecol. 2013. Apr;121(4):789–797. doi: 10.1097/AOG.0b013e3182878b43. [DOI] [PubMed] [Google Scholar]
- 5.Silver RM. Implications of the first cesarean: perinatal and future reproductive health and subsequent cesareans, placentation issues, uterine rupture risk, morbidity, and mortality. Semin Perinatol 2012;36:315–23. doi: 10.1053/j.semperi.2012.04.013 [DOI] [PubMed] [Google Scholar]
- 6.Fingar KR, Hambrick MM, Heslin KC, Moore JE. Trends and Disparities in Delivery Hospitalizations Involving Severe Maternal Morbidity, 2006–2015. 2018 Sep 4. In: Healthcare Cost and Utilization Project (HCUP) Statistical Briefs. Rockville (MD): Agency for Healthcare Research and Quality (US); 2006. Feb–. Statistical Brief #243. [PubMed] [Google Scholar]
- 7.Leonard SA, Main EK, Scott KA, Profit J, Carmichael SL. Racial and Ethnic Disparities in Severe Maternal Morbidity Prevalence and Trends. Annals of Epidemiology. 2019. 33:30–36 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Petersen EE, Davis NL, Goodman D, Cox S, Mayes N, Johnston E, Syverson C, Seed K, Shapiro-Mendoza CK, Callaghan WM, Barfield W. Vital Signs: Pregnancy-Related Deaths, United States, 2011-2015, and Strategies for Prevention, 13 States, 2013-2017. MMWR Morb Mortal Wkly Rep. 2019. May 10;68(18):423–429. doi: 10.15585/mmwr.mm6818e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Kern-Goldberger AR, Booker W, Friedman A, Gyamfi-Bannerman C. Racial and Ethnic Disparities in Cesarean Morbidity. Am J Perinatol. 2021. Dec 10. doi: 10.1055/s-0041-1739305. [DOI] [PubMed] [Google Scholar]
- 10.Getahun D, Strickland D, Lawrence JM, Fassett MJ, Koebnick C, Jacobsen SJ. Racial and ethnic disparities in the trends in primary cesarean delivery based on indications. Am J Obstet Gynecol. 2009. Oct;201(4):422.e1–7. doi: 10.1016/j.ajog.2009.07.062. [DOI] [PubMed] [Google Scholar]
- 11.Salahuddin M, Mandell DJ, Lakey DL, Eppes CS, Patel DA. Maternal risk factor index and cesarean delivery among women with nulliparous, term, singleton, vertex deliveries, Texas, 2015. Birth. 2019. Mar;46(1):182–192. doi: 10.1111/birt.12392. [DOI] [PubMed] [Google Scholar]
- 12.Bryant AS, Washington S, Kuppermann M, Cheng YW, Caughey AB. Quality and equality in obstetric care: racial and ethnic differences in caesarean section delivery rates. Paediatr Perinat Epidemiol. 2009. Sep;23(5):454–62. doi: 10.1111/j.1365-3016.2009.01059.x. [DOI] [PubMed] [Google Scholar]
- 13.Okwandu IC, Anderson M, Postlethwaite D, Shirazi A, Torrente S. Racial and Ethnic Disparities in Cesarean Delivery and Indications Among Nulliparous, Term, Singleton, Vertex Women. J Racial Ethn Health Disparities. 2021. Jul 12. doi: 10.1007/s40615-021-01057-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Cáceres IA, Arcaya M, Declercq E, Belanoff CM, Janakiraman V, Cohen B, Ecker J, Smith LA, Subramanian SV. Hospital differences in cesarean deliveries in Massachusetts (US) 2004-2006: the case against case-mix artifact. PLoS One. 2013;8(3):e57817. doi: 10.1371/journal.pone.0057817. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Main EK, Moore D, Farrell B, Schimmel LD, Altman RJ, Abrahams C, Bliss MC, Polivy L, Sterling J. Is there a useful cesarean birth measure? Assessment of the nulliparous term singleton vertex cesarean birth rate as a tool for obstetric quality improvement. Am J Obstet Gynecol. 2006. Jun;194(6):1644–51doi: 10.1016/j.ajog.2006.03.013. [DOI] [PubMed] [Google Scholar]
- 16.Barber EL, Lundsberg LS, Belanger K, Pettker CM, Funai EF, Illuzzi JL. Indications contributing to the increasing cesarean delivery rate. Obstet Gynecol. 2011; 118(1): 29–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.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. Obstet Gynecol. 2012; 120(5): 1181–1193. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Nouh A, Amin-Hanjani S, Furie KL, Kernan WN, Olson DM, Testai FD, Alberts MJ, Hussain MA, Cumbler EU; American Heart Association Stroke Council; Council on Arteriosclerosis, Thrombosis and Vascular Biology; Council on Cardiovascular and Stroke Nursing; Council on Clinical Cardiology; and Council on Lifestyle and Cardiometabolic Health. Identifying Best Practices to Improve Evaluation and Management of In-Hospital Stroke: A Scientific Statement from the American Heart Association. Stroke. 2022. Apr;53(4):e165–e175. doi: 10.1161/STR.0000000000000402. [DOI] [PubMed] [Google Scholar]
- 19.Goitein L, James B. Standardized Best Practices and Individual Craft-Based Medicine: A Conversation About Quality. JAMA Intern Med. 2016. Jun 1;176(6):835–8. doi: 10.1001/jamainternmed.2016.1641. [DOI] [PubMed] [Google Scholar]
- 20.Beate D. Probabilistic record linkages for generating a comprehensive epidemiological research file on maternal and infant health http://www.health-info-solutions.com/Probabilistic%20Record%20Linkages%20may%202002.pdf. Accessed April 4, 2023.
- 21.Rosenstein MG, Chang S-C, Sakowski C, Markow C, Teleki S, Lang L, Logan J, Cape V, Main EK. Hospital Quality Improvement Interventions, Statewide Policy Initiatives and Rates of Nulliparous Term Singleton Vertex Cesarean Deliveries in California. JAMA 2021. Apr 27;325(16):1631–1639. doi: 10.1001/jama.2021.3816. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Specifications Manual for Joint Commission National Quality Measures (v2022A1), Perinatal Care Measure PC-02. https://manual.jointcommission.org/releases/TJC2022A1/MIF0167.html Accessed April 4, 2023.
- 23.American Academy of Pediatrics Committee on Fetus And Newborn. Levels of neonatal care. Pediatrics. 2012. Sep;130(3):587–97. doi: 10.1542/peds.2012-1999. [DOI] [PubMed] [Google Scholar]
- 24.Derrick B; Ruck A; Toher D; White P Tests for equality of variances between two samples which contain both paired observations and independent observations. Journal of Applied Quantitative Methods. 2018. 13 (2): 36–47. [Google Scholar]
- 25.White VanGompel E, Main EK, Tancredi D, Melnikow J. Do provider birth attitudes influence cesarean delivery rate: a cross-sectional study. BMC Pregnancy Childbirth. 2018. May 29;18(1):184. doi: 10.1186/s12884-018-1756-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.White VanGompel E, Perez S, Datta A, Wang C, Cape V, Main E. Cesarean overuse and the culture of care. Health Serv Res. 2019. Apr;54(2):417–424. doi: 10.1111/1475-6773.13123. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Metz TD, Allshouse AA, Gilbert SAB, Doyle R, Tong A, Carey JC. Variation in primary cesarean delivery rates by individual physician within a single-hospital laborist model. Am J Obstet Gynecol. 2016. Apr;214(4):531.e1–531.e6. doi: 10.1016/j.ajog.2016.01.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Stark EL, Grobman WA, Miller ES. The Association between Maternal Race and Ethnicity and Risk Factors for Primary Cesarean Delivery in Nulliparous Women. Am J Perinatol. 2021. Mar;38(4):350–356. doi: 10.1055/s-0039-1697587. [DOI] [PubMed] [Google Scholar]
- 29.Washington S, Caughey AB, Cheng YW, Bryant AS. Racial and ethnic differences in indication for primary cesarean delivery at term: experience at one U.S. Institution. Birth. 2012. Jun;39(2):128–34. doi: 10.1111/j.1523-536X.2012.00530.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Yee LM, Costantine MM, Rice MM, Bailit J, Reddy UM, Wapner RJ, Varner MW, Thorp JM Jr, Caritis SN, Prasad M, Tita ATN, Sorokin Y, Rouse DJ, Blackwell SC, Tolosa JE; Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) Maternal-Fetal Medicine Units (MFMU) Network. Racial and Ethnic Differences in Utilization of Labor Management Strategies Intended to Reduce Cesarean Delivery Rates. Obstet Gynecol. 2017. Dec;130(6):1285–1294. doi: 10.1097/AOG.0000000000002343. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.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]
- 32.Scott KA, Britton L, McLemore MR. The Ethics of Perinatal Care for Black Women: Dismantling the Structural Racism in “Mother Blame” Narratives. J Perinat Neonatal Nurs. 2019;33(2):108–15. [DOI] [PubMed] [Google Scholar]
- 33.Altman MR, McLemore MR, Oseguera T, Lyndon A, Franck LS. Listening to Women: Recommendations from Women of Color to Improve Experiences in Pregnancy and Birth Care. J Midwifery Womens Health. 2020;65(4):466–73. [DOI] [PubMed] [Google Scholar]
- 34.Chokshi DA, Foote MMK, Morse ME. How to act upon racism--not race--as a risk factor. JAMA Health Forum. 2022;3(2):e220548. doi: 10.1001/jamahealthforum.2022.0548 [DOI] [PubMed] [Google Scholar]
- 35.Boyd RW, Lindo EG, Weeks LD, McLemore MR. On racism: A new standard for publishing on racial health inequities. Health Affairs July 2, 2020. 10.1377/forefront.20200630.939347/. Accessed on April 4, 2023. [DOI] [Google Scholar]
- 36.Prather C, Fuller TR, Marshall KJ, Jeffries WL. The Impact of Racism on the Sexual and Reproductive Health of African American Women. J Womens Health (Larchmt). 2016;25(7):664–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Bryant AS, Washington S, Kuppermann M, Cheng YW, Caughey AB. Quality and equality in obstetric care: racial and ethnic differences in caesarean section delivery rates. Paediatr Perinat Epidemiol. 2009. Sep;23(5):454–62. doi: 10.1111/j.1365-3016.2009.01059.x. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.