Lipkind, Heather S. MD, MS1; Duzyj, Christina MD, MPH1; Rosenberg, Terry J. PhD2; Funai, Edmund F. MD1; Chavkin, Wendy MD, MPH3; Chiasson, Mary Ann DrPH2,3
Studies have explored potential medical and nonmedical influences on the cesarean delivery rate. Gould et al found that primary cesarean delivery varied directly with socioeconomic status.1 Recent data from Arizona suggest that individual hospital variations of primary cesarean delivery range from a low rate of 10.3% to a high rate of 34.2% depending on the hospital of delivery. The rate of cesarean delivery varies across regions in the United States by institutional type, patient’s race, and type of practitioner.2–4 Financial considerations and institution of delivery have been shown to play strong roles in influencing the mode of delivery. A study performed in the early 1990s shows that the increasing cesarean delivery rate in the 1970s and 1980s was related to insurance status, with privately insured women having the highest cesarean delivery rates.5
In the United States, there has been interest in decreasing the primary cesarean delivery rate for the last two decades, beginning with a National Institute of Health consensus conference in 1980.6 The U.S. Department of Health and Human Services Healthy People 2010 also developed the goal of a national primary cesarean delivery rate of 15% among low risk women, representing a decrease from the 1998 baseline of 18%.7 Despite these efforts, and a modest decrease in the overall cesarean delivery rate in the mid-1990s, the primary cesarean delivery rate has again begun to rise. The percentage of U.S. births delivered by cesarean has climbed 50 % over the last decade and reached an all time high of 29.1 % in 2004, the last national data to be reported.8 Previous studies of cesarean delivery have shown that there is an association between socioeconomic status and the primary cesarean delivery rate.1,9
There is also controversy as to whether the increasing cesarean delivery rate has actually led to improvement in neonatal outcomes. One study suggests that neonates delivered by cesarean may have a higher rate of respiratory morbidity compared with those delivered vaginally.10
New York City provides an excellent opportunity to investigate possible relationships between insurance status, hospital type and the primary cesarean delivery rate as the city has a large and economically diverse population as well as the largest municipal hospital system in the United States. The public hospital system in New York City, run by the Health and Hospitals Corporation, consists of 11 acute-care hospitals with birthing facilities. These hospitals serve primarily a Medicaid patient population. There are also 38 voluntary private hospitals, including not-for-profit voluntary hospitals and for-profit hospitals, which serve women with both private insurance and Medicaid.
The study reported here analyzes differences between the primary cesarean delivery rate for patients with private insurance and those with Medicaid in New York City in both public and private hospital settings. We also examine associations between the differing rates of mode of delivery and birth outcomes in different hospital settings.
MATERIALS AND METHODS
We analyzed the birth files of all singleton neonates born in New York City from 1996 through 2003 to nulliparous women who were delivered either by vaginal or cesarean delivery. Information from the confidential medical portion of the birth certificate was obtained from the New York City Department of Health and Mental Hygiene, Office of Vital Statistics.3,11
Data were collected from 321,308 nulliparous women delivering singleton neonates weighing between 500 and 6,000 g beyond 24 weeks of gestation. We restricted the analysis to first births to exclude pregnancies in which the outcome of the previous pregnancy may have influenced the decision to have a cesarean delivery. We included only women who were delivered by primary cesarean or spontaneous vaginal delivery and excluded all forceps and vacuum deliveries.
The 11 hospitals that were part of the New York City Health and Hospital System, a public-benefit corporation, were considered public hospitals. All other hospitals in New York City were considered private, including both voluntary not-for-profit and for-profit hospitals.
The patients then were analyzed by type of insurance, which was limited to private and Medicaid as captured on the New York City birth certificate without personal identifiers. Women who belonged to a health maintenance organization, who were self-pay, or who had unknown insurance status were excluded from the analysis. Three separate groups of patients were created and analyzed as defined by insurance status and hospital of delivery: public hospital/Medicaid insurance, private hospital/Medicaid insurance, and private hospital/private insurance. Owing to the small number of patients with private insurance delivering in the public hospital system (176 patients, less than 0.003%), these patients were excluded from the analysis.
Maternal age at delivery, race, prepregnancy weight, and maternal education level data were analyzed for the three provider groups. Maternal race and ethnicity categories were defined by the New York City Department of Health and Mental Hygiene on the birth file. The prepregnancy weight variable was used because information on Body Mass Index is not listed in the birth certificate and prepregnancy weight been used as a proxy in previous studies by our group.12,13
Obstetrical procedures and maternal and medical conditions have been consistently underreported in birth certificates.14,15 The New York City birth file also provides two separate areas for listing medical- and pregnancy-related complications. All medical complications include current illnesses and are indicated by specific boxes on the certificate. Pregnancy-related complications are noted in a separate area. Because specific complications were not mutually exclusive and some were rare or underreported, we created a categorical composite of one or more of the conditions for the different indications (pregnancy complications yes/no and medical complication yes/no) in an attempt to increase the accuracy of reporting maternal medical conditions.16
The maternal medical complications included renal disease, cardiac disease, pregestational and gestational diabetes, lung disease, herpes, and chronic hypertension.
Pregnancy complications include one or more of the following: eclampsia, cord prolapse, breech delivery, placenta previa, chorioamnionitis, preeclampsia, and oligohydramnios. We also looked at indications for cesarean delivery marked on the birth certificate in the three patient groups. In an attempt to examine the distinction between primary cesarean deliveries without trial of labor and those that occurred after a trial of labor, we looked at the rate of indication for cesarean delivery for breech presentations and the rate of cesarean delivery where all indications had been excluded on the birth certificate.
We reviewed unadjusted primary cesarean delivery rates. To control for possible confounding factors, cesarean delivery rates within the three groups were stratified by independent risk factors for primary cesarean delivery, including maternal age at delivery, race/ethnicity, prepregnancy weight, maternal education, birth weight, and medical-and pregnancy- related complications. We report the rates per 100 deliveries for all confounding factors by the three provider and insurance types.
The odds of primary cesarean delivery compared with vaginal delivery was then evaluated using χ2 tests for categorical variables. We tested for confounding using variables that were both a risk factor for the outcome and associated with the exposures of interest. Potential confounders were identified using a priori knowledge of the literature and included maternal demographic characteristics, pregnancy-related characteristics, maternal medical- and pregnancy-related problems, and hospital type and care provider. Covariates were considered for inclusion in the multivariable model if the bivariate χ2 P values were less than .05. We used the difference in the −2 log likelihood ratios to determine which variables improved model fit and removed covariates with inconsequential contribution to the model, retaining only those which significantly contributed to final model fit (P<.05). To examine the influence of changing practice patterns over time, we also entered a variable that included time into the original model. As it did not meet the criteria for inclusion into the model, it was not included in the final analysis.
We then looked at two measures of adverse neonatal outcomes in the immediate postpartum period. Neonatal intensive care unit (NICU) admissions and 5-minute Apgar score less than 7 were examined to determine whether there was a difference in outcomes by hospital type and insurance status. These measures have been used as outcome variables in multiple previous studies.17,18 Tabular findings were evaluated using χ2 tests. The multivariable model analysis was created using methods as described above.
SAS 9.1 for Windows was used for data analysis. The study was approved by Institutional Review Board at the New York City Department of Health and Mental Hygiene.
From 1996 through 2003, there were 321,308 primiparous singleton deliveries by either cesarean or normal spontaneous vaginal delivery in New York City to women with either Medicaid or private insurance. There were 51,682 and 269,626 women who delivered in public hospitals and private hospitals, respectively. Between 1996 and 2003, there were 79,442 primary cesarean deliveries. Patients with private insurance had the highest overall cesarean delivery rate (30.4%), followed by patients with Medicaid delivering in the private hospitals (21.2%), and Medicaid patients delivering in the public hospital system (20.5%). The cesarean delivery rate increased for all patient groups over time.
Table 1 shows maternal demographic characteristics that have been associated with the rate of primary cesarean delivery. Women with private insurance were more likely to be older than 35 years of age than were women in the two Medicaid populations. Patients with Medicaid in both public and private hospitals were more likely to be non-white than were women with private insurance. The three groups of women had similar prepregnancy weight distributions. Patients with private insurance had lower rates of pregnancy-related complications than did Medicaid populations at both types of hospital. Medicaid patients in the public hospital system had the highest percentage of both medical and pregnancy complications. It is important to note that, because this is a large database, some comparisons may be statistically significant that are not clinically relevant.
Table 2 shows the rates of primary cesarean delivery by hospital and insurance status per 100 births for maternal demographics and birth characteristics. Maternal age greater than or equal to 35 years old has been shown to be an independent risk factor for cesarean delivery, and age at delivery has been increasing in New York City.2,19,20 We found that as age increased the cesarean delivery rate increased for all three groups. The cesarean delivery rate was highest in all age categories for patients with private insurance.
All women were assigned to one of four mutually exclusive racial or ethnic groups: Hispanic, Asian, non-Hispanic white, and non-Hispanic African-American. Non-Hispanic African-Americans had the highest rate of primary cesarean delivery, and non-Hispanic whites had the lowest rate in all three patient populations. Within each racial/ethnic group, the rate of primary cesarean delivery was highest for women with private insurance over all racial/ethnic categories.
Although the proportion of women with pregnancy- and medical-related complications was lower for women with private insurance than for the two Medicaid groups, the rate of cesarean delivery was still significantly higher than in the Medicaid populations for women with medical- and pregnancy-related complications.
We examined the unadjusted and adjusted odds of cesarean delivery compared with vaginal delivery for hospital and insurance type, and adjusted for birth weight, gestational age at delivery, and maternal characteristics that have been associated with cesarean delivery in multiple studies.1,5,9,21 Maternal weight, race, medical complications and pregnancy complications were all associated with increased rates of cesarean delivery.
Table 3 shows the tabulated effects of the variables that significantly affected the rate of primary cesarean delivery. After adjusting for multiple potential confounders, patients with private insurance had significantly higher odds of having cesarean deliveries than did Medicaid patients delivering in a public hospital (adjusted odds ratio [OR] 1.57, 95% confidence interval [CI] 1.52–1.63). The Medicaid population delivering in private hospitals also had slightly increased odds of cesarean delivery compared with patients with Medicaid delivering in public hospitals (adjusted OR 1.12, 95% CI 1.08–1.15).
Patients with private insurance had a slightly higher rate of macrosomic neonates than did the Medicaid populations and slightly higher overall birth weights. Table 2 shows that the rate of primary cesarean delivery was highest among women who gave birth to very low birth weight neonates, lowest for normal-weight neonates, and increased again as the birth weight rose above 4,000 g. Patients with private insurance had a higher rate of cesarean delivery for both low and high birth weight neonates than did both Medicaid populations. Medicaid patients delivering in public hospitals had the lowest cesarean delivery rate for all birth weights.
The most frequent indications for cesarean deliveries in women with private insurance were failure to progress and cephalo-pelvic disproportion (40.9%) compared with women with Medicaid (34.6% and 36.3%) in public and private hospitals. Women with Medicaid were more likely to have cesarean deliveries for fetal distress in public and private hospitals (38.1% and 34.9%) than were women with private insurance (22.8%). Although it is not possible to distinguish on the birth certificate what percentage of the primary cesarean deliveries occurred with and without a trial of labor, we were able to examine the percentage of deliveries that were performed for a breech presentation and without any indications. Women with private insurance were more likely to have cesarean deliveries performed for breech presentation (14.7%) than were those with Medicaid in the public and private hospital setting (13.2% and 12.7%). Private patients were also more likely to have cesarean deliveries where no indication was noted on the birth certificate (19.6%) than were those with Medicaid in public and private hospitals (13.1% and 142%). This suggests that more deliveries were performed without a trial of labor in the private-patient population; however, our ability to judge this was limited.
A total of 27,279 neonates (8.0%) had NICU admissions regardless of mode of delivery, and 2,442 neonates (0.8%) had 5-minute Apgar scores less than 7. Table 4 shows the rates of neonatal outcomes for the three different patient groups stratified by mode of delivery. Patients with Medicaid delivering in public hospitals had higher rates of NICU admissions and 5-minute Apgar score less than 7 compared with both groups delivering in private hospitals for both vaginal and cesarean deliveries. After controlling for multiple factors including mode of delivery, patients with both private insurance and Medicaid delivering in private hospitals had significantly decreased adjusted odds of NICU admission than did patients delivering in the public hospital system (adjusted OR 0.48, 95% CI 0.46–0.51 and adjusted OR 0.59, 95% CI 0.57–0.62). After controlling for confounders including mode of delivery, neonates born to those with both private insurance and Medicaid delivering in private hospitals were also significantly less likely to have 5-minute Apgar scores less than 7 than were those born to patients delivering in the public hospital system (adjusted OR 0.59, 95% CI 0.51–0.68 and adjusted OR 0.73, 95% CI 0.65–0.82).
The proportion of all births by cesarean delivery has climbed 50% over the past decade.8 This statistic has concerned public health experts, accreditation bodies, and insurance companies, who worry that some cesarean deliveries might be performed without medical indications. Previous studies have shown that changes in obstetrical practice, including decreased forceps usage, increasing rates of primary cesarean delivery for breech, labor induction, epidural anesthesia, and delivery by an obstetrician all contribute to an increased cesarean delivery rate.22–26 An increasing concern for long-term neonatal morbidity also has been cited as a contributing factor.26
In this study we found that women with private insurance delivering in private hospitals had both the greatest increase in cesarean delivery rates over the study period as well as a higher rate of cesarean delivery than did Medicaid patients delivering in either private or public hospitals. This relationship held true after controlling for maternal age, education, race, neonatal birth weight, and other potentially confounding variables.
It has been demonstrated that fear of litigation has contributed to the increase in cesarean delivery.27–29 Privately insured patients may be perceived as posing a greater medical–legal risk, causing providers in private settings to proceed more cautiously with labor progress. This may be reflected in the higher rate of “failure to progress” indications for cesarean delivery in privately insured patients in our study.
The higher rate of cesarean deliveries for “failure to progress” in the private-patient population also may be attributed to an increase in induction of labor in the private-patient population. According to the National Center for Heath Statistics, in 2003, the rate of induction of labor in the United States was approximately 21.6%, an increase of more than 13% since 1989, the first year these data were collected. Half of the inductions were elective.27 This increase has been attributed to increasing availability and use of cervical ripening agents but also to patient pressure on physicians and physician convenience.
This raises the question of whether the lower cesarean delivery rate in Medicaid patients reflects receipt of less vigilant attention by their providers or more of a delay in action on fetal distress owing to less of a perceived threat of litigation. Laye et al in 2006 compared compliance with the American College of Obstetrics and Gynecology guidelines in “teaching” service patients and private patients. They show that patients on a “teaching” service were less likely to have primary cesarean deliveries and were more likely to be delivered in adherence to guidelines set forth by the American College of Obstetricians and Gynecologists.30 Moreover, this study reveals no difference in admission to the NICU or other neonatal outcomes between patient populations, potentially refuting the “negligence” theory.
It is controversial as to whether the increasing cesarean delivery rate actually has led to improvement in neonatal outcomes. Liston et al reviewed 142,929 deliveries in Nova Scotia between 1988 and 2002 and observed that neonates born via cesarean delivery in labor had a fourfold to fivefold rate of “depression” at birth (defined as delay in initiating respiratory effect, 5-minute Apgar score of less than 3, and hypoxic–ischemic encephalopathy) compared with those born from spontaneous deliveries.17 They also noted a twofold to threefold increased rate of all neonatal respiratory conditions. Kolas et al in 2007 noted increased NICU transfer rates and pulmonary disorders among planned cesarean deliveries as compared with planned vaginal deliveries.10 Odd et al recently correlated prolonged Apgar score less than 7 with low intelligence quotients at age 18, raising concerns about long-term sequelae.18 In our study, neonatal complication rates were higher for those delivered by cesarean than for those delivered vaginally, but neonatal outcomes also differed by hospital type regardless of insurance.
Both the private and Medicaid patients delivering neonates in private hospitals had better neonatal outcomes than did patients delivering in public hospitals in New York City. This was true even after controlling for cesarean delivery, race, and medical- and pregnancy-related complications, which were higher in the private-patient population. Howell et al examined the neonatal mortality rates for very low birth weight neonates in New York City hospitals. They found that African-American neonates in New York often were treated in hospitals with higher mortality rates and that white mothers were more likely to deliver in hospitals with lower mortality rates.31
There were several limitations to our study. Because this was a retrospective study based on birth certificate data, misclassification of data may have occurred. It is unlikely that misclassification would vary according to outcomes or exposures. We excluded women whose records were incomplete owing to missing data. The exclusion, however, may not have been differential across the three provider groups depending on how accurately the birth certificates were filled out in the different hospital settings. The New York State Department of Health performed a validation study in which they compared birth certificate data from four counties in New York State and sampled 440 medical records in 1999. They then compared the birth certificates with the medical records and looked at the sensitivity, specificity, and positive predictive values of the records. There was more than 95% accuracy for 9 out of 10 maternal demographic variables. Primary mode of delivery was 99.75% correct.14
Studies provide evidence that maternal demographic data and basic neonatal characteristics are accurately reported on birth certificates; however, these studies also show that obstetrical procedures and maternal and neonatal medical conditions have been underreported consistently, which can lead to underestimates of clinical events. In addition, the underreporting may be biased in that certain conditions that potentially lead to a cesarean delivery may be more heavily reported than conditions that are not related to the outcome of the pregnancy.14,15,32,33
Another limitation of birth certificate data is that it is difficult to make a distinction between a trial of labor and a primary cesarean delivery without a trial of labor with birth certificate data. Although more cesarean deliveries were performed for breech presentation and without an indication for cesarean delivery in the private-patient population, this is not a good proxy for a cesarean delivery without a trial of labor, which may have implications for neonatal outcomes. The new birth certificate for New York City includes a box for “elective” cesarean delivery, which may be helpful in examining the neonatal outcomes after a trial of labor compared with no trial of labor. Vital statistics data are known to be limiting.
This study focused on primary cesarean deliveries and neonatal outcomes in the New York City population, but it has important implications on a national level given the rising overall cesarean delivery rate. What we found in this study that was both surprising and concerning is that setting of care, which is intimately related to socioeconomic status, affects not only the cesarean delivery rate but also associated neonatal outcomes.
1. Gould JB, Davey B, Stafford RS. Socioeconomic differences in rates of cesarean section. N Engl J Med 1989;321:233–9.
2. Gregory KD, Ramicone E, Chan L, Kahn KL. Cesarean deliveries for medicaid patients: a comparison in public and private hospitals in Los Angeles County. Am J Obstet Gynecol 1999;180:1177–84.
3. Thorpe LE, Berger D, Ellis JA, Bettegowda VR, Brown G, Matte T, et al. Trends and racial/ethnic disparities in gestational diabetes among pregnant women in New York City, 1990-2001. Am J Public Health 2005;95:1536–9.
4. Tussing AD, Wojtowycz MA. The cesarean decision in New York State, 1986. Economic and noneconomic aspects. Med Care 1992;30:529–40.
5. Stafford RS, Sullivan SD, Gardner LB. Trends in cesarean section use in California, 1983 to 1990. Am J Obstet Gynecol 1993;168:1297–302.
6. Cesarean childbirth. NIH Consensus Statement 1980;3:1–30.
7. U.S. Department of Health and Human Services. Healthy people 2010: understanding and improving health. 2nd ed. Washington (DC): U.S. Government Printing Office; 2008.
8. Hamilton B, Martin JA, Ventura SJ, Sutton PD, Menacker F. Births: preliminary data for 2004. Natl Vital Stat Rep 2005;54:1–17.
9. Braveman P, Egerter S, Edmonston F, Verdon M. Racial/ethnic differences in the likelihood of cesarean delivery, California. Am J Public Health 1995;85:625–30.
10. Kolås T, Saugstad OD, Daltveit AK, Nilsen ST, Øian P. Planned cesarean versus planned vaginal delivery at term: Comparison of newborn infant outcomes. Am J Obstet Gynecol 2006;195:1538–43.
11. Frost F, Starzyk P, George S, McLaughlin JF. Birth complication reporting: the effect of birth certificate design. Am J Public Health 1984;74:505–6.
12. Rosenberg T, Garbers S, Chavkin W, Chiasson MA. Prepregnancy weight and adverse perinatal outcomes in an ethnically diverse population. Obstet Gynecol 2003;102:1022–7.
13. Rosenberg T, Garbers S, Lipkind H, Chiasson MA. Maternal obesity and diabetes as risk factors for adverse pregnancy outcomes: differences among 4 racial/ethnic groups. Am J Public Health 2005;95:1545–51.
14. Roohan PJ, Josberger RE, Acar J, Dabir P, Feder HM, Gagliano PJ. Validation of birth certificate data in New York State. J Community Health 2003;28:335–46.
15. Lydon-Rochelle MT, Holt VL, Cardenas V, Nelson JC, Easterling TR, Gardella C, et al. The reporting of pre-existing maternal medical conditions and complications of pregnancy on birth certificates and in hospital discharge data. Am J Obstet Gynecol 2005;193:125–34.
16. Zweifler J, Garza A, Hughes S, Stanich MA, Hierholzer A, Lau M. Vaginal birth after cesarean in California: before and after a change in guidelines. Ann Fam Med 2006;4:228–34.
17. Liston FA, Allen VM, O’Connell Jangaard KA. Neonatal outcomes with caesarean delivery at term. Arch Dis Child Fetal Neonatal Ed 2008;93:F176–82.
18. Odd DE, Rasmussen F, Gunnell D, Lewis G, Whitelaw A. A cohort study of low Apgar scores and cognitive outcomes. Arch Dis Child Fetal Neonatal Ed 2008;93:F115–20.
19. Cleary-Goldman J, Malone FD, Vidaver J, Ball RH, Nyberg DA, Comstock CH, et al. Impact of maternal age on obstetric outcome. Obstet Gynecol 2005;105:983–90.
20. Dulitzki M, Soriano D, Schiff E, Chetrit A, Mashiach S, Seidman DS. Effect of very advanced maternal age on pregnancy outcome and rate of cesarean delivery. Obstet Gynecol 1998;92:935–9.
21. de Regt RH, Minkoff HL, Feldman J, Schwarz RH. Relation of private or clinic care to the cesarean birth rate. N Engl J Med 1986;315:619–24.
22. Makris N, Xygakis A, Chionis A, Sakellaropoulos G, Michalas S. The management of breech presentation in the last three decades. Clin Exp Obstet Gynecol 1999;26:178–80.
23. Lieberman E, Lang JM, Cohen A, D’Agostino R Jr, Datta S, Frigoletto FD Jr. Association of epidural analgesia with cesarean delivery in nulliparas. Obstet Gynecol 1996;88:993–1000.
24. Michelson KA, Carr DB, Easterling TR. The impact of duration of labor induction on cesarean rate. Am J Obstet Gynecol 2008;199:299e1–4.
25. Coco AS, Gates TJ, Gallagher ME, Horst MA. Association of attending physician specialty with the cesarean delivery rate in the same patient population. Fam Med 2000;32:639–44.
26. Joseph KS, Young DC, Dodds L, O’Connell CM, Allen VM, Chandra S, et al. Changes in maternal characteristics and obstetric practice and recent changes in primary cesarean delivery. Obstet Gynecol 2003;102:791–800.
27. Penn Z, Ghaem-Maghami S. Indications for caesarean section. Best Pract Res Clin Obstet Gynaecol 2001;15:1–15.
28. Studdert DM, Mello MM, Sage WM, DesRoches CM, Peugh J, Zapert K, et al. Defensive medicine among high-risk specialist physicians in a volatile malpractice environment. JAMA 2005;293:2609–17.
29. Dubay L, Kaestner R, Waidmann T. The impact of malpractice fears on cesarean section rates. J Health Econ 1999;18:491–522.
30. Laye MR, Dellinger EH. Timing of scheduled cesarean delivery in patients on a teaching versus private service: adherence to American College of Obstetricians and Gynecologists guidelines and neonatal outcomes. Am J Obstet Gynecol 2006;195:577–82.
31. Howell EA, Hebert P, Chatterjee S, Kleinman LC, Chassin MR. Black/white differences in very low birth weight neonatal mortality rates among New York City hospitals. Pediatrics 2008;121:e407–15.
32. Schoendorf KC, Branam AM. The use of United States vital statistics in perinatal and obstetric research. Am J Obstet Gynecol 2006;194:911–5.
33. DiGiuseppe DL, Aron DC, Ranbom L, Harper DL, Rosenthal GE. Reliability of birth certificate data: A multi-hospital comparison to medical records information. Matern Child Health J 2002;6:169–79.
© 2009 by The American College of Obstetricians and Gynecologists.