American Journal of Epidemiology Advance Access originally published online on January 11, 2008
American Journal of Epidemiology 2008 167(6):711-718; doi:10.1093/aje/kwm363
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ORIGINAL CONTRIBUTIONS |
Effect of Hospital Volume on Maternal Outcomes in Women with Prior Cesarean Delivery Undergoing Trial of Labor
1 Division of Epidemiology, Department of Community Health, School of Public Health, Saint Louis University, St. Louis, MO
2 Department of Obstetrics and Gynecology, School of Medicine, Washington University in St. Louis, St. Louis, MO
Correspondence to Dr. Jen Jen Chang, Department of Community Health, Saint Louis University School of Public Health, 3545 Lafayette Avenue, Suite 300, St. Louis, MO 63104 (e-mail: changdalton{at}gmail.com).
Received for publication August 14, 2007. Accepted for publication November 9, 2007.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The authors examined the association between hospital volume of vaginal birth after cesarean section (VBAC) and VBAC failure, uterine rupture, and maternal morbidity. This was a secondary analysis of data from a retrospective cohort study carried out from 1995 to 2000. Trained nurses extracted detailed information from the medical records of more than 25,000 women with a prior cesarean delivery from 17 community and tertiary-care hospitals in the northeastern United States. The study sample included 12,844 women with prior cesarean section who attempted vaginal delivery with a singleton birth. Annual hospital VBAC volume was divided into tertiles. Primary outcomes included VBAC failure, uterine rupture, and a composite measure of maternal morbidity. The authors used multivariable logistic regression to assess the association between hospital VBAC volume and adverse VBAC outcomes after controlling for confounders. The authors did not find evidence of an association between hospital VBAC volume and the likelihood of adverse outcomes in VBAC after adjustment for patient mix. Other risk factors consistent with prior research were identified, including induction of labor,
2 prior cesarean deliveries, preeclampsia, diabetes mellitus, and high birth weight. Prior vaginal delivery was protective against adverse VBAC outcomes. The risk of an adverse VBAC outcome in low-volume hospitals was comparable to that in high-volume hospitals.
cesarean section; obstetric labor complication; parturition; trial of labor; uterine rupture; vaginal birth after cesarean
Abbreviations: VBAC, vaginal birth after cesarean
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
In 2004, the rate of cesarean delivery in the United States reached a historical high of 29.1 percent (1). One of the objectives of Healthy People 2010 is to reduce the rates of both primary and repeat cesarean delivery in an effort to reduce maternal morbidity and mortality (2). It has been proposed that the increasing rate of cesarean delivery may be partly attributed to aging of the childbearing population, an increase in maternal medical risks, or a rise in the number of multiple pregnancies (3). While more expansive use of vaginal birth after cesarean section (VBAC) will not decrease the primary rate of cesarean delivery, increasing the proportion of women who attempt VBAC could reduce repeat rates and therefore overall rates of cesarean delivery. The inverse relation between trends in the VBAC rate and the cesarean rate between 1989 and 2004 supports the concept that increasing VBAC may reduce the rate of cesarean birth (4). Most importantly, increased utilization of VBAC could significantly reduce the rate of maternal and neonatal morbidity, because the rate of perioperative morbidity escalates with increasing number of cesarean sections in an individual (5).
Because of concern about maternal morbidity and complications such as uterine rupture in women who attempt a trial of labor, the rate of VBAC in women with a previous cesarean birth has fallen by 67 percent since 1996, from 28.3 per 100 births to 9.2 per 100 births, with subsequent rising rates of repeat cesarean delivery (1). Prior research has indicated that the rate of successful trial of labor among women who have had a previous cesarean delivery ranges from 56 percent to 82 percent and that these women's risk of uterine rupture is generally less than 1 percent (6). Researchers have sought to elucidate factors that may increase risk during a trial of labor. These factors include prior and current obstetric history (7, 8), birth weight (9), maternal weight (10), and nonclinical factors, such as hospital or physician characteristics (11, 12).
The relation between hospital volume and clinical outcome has been examined for a variety of surgical and high-risk medical procedures, including pancreatic cancer, esophageal cancer, abdominal aortic aneurysms, and pediatric cardiac problems. It is generally held that hospitals with high volumes of a certain procedure probably have better clinical outcomes. Although previous studies have demonstrated that hospitals which perform more surgical procedures or discharge more patients with certain medical conditions have lower rates of in-hospital mortality (13), the relation of hospital volume to VBAC outcome is less clear. Thus, our purpose in this study was to estimate the association of hospital VBAC volume with clinically relevant outcomes, such as VBAC failure, uterine rupture, and other major types of maternal morbidity, among women attempting a trial of labor after at least one prior cesarean delivery.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
We conducted a secondary analysis of data from a multicenter, retrospective cohort study that included 25,079 pregnant women with at least one previous cesarean section (identified via International Classification of Diseases, Ninth Revision, code) who were offered VBAC between 1995 and 2000. Subjects were excluded if they had a prior vertical (classical) or unknown type of uterine incision or had a pregnancy complicated by fetal anomalies. The details of the retrospective cohort study have been previously published (8), but a brief description follows. The study subjects were patients from 17 community and university-based hospitals in the northeastern United States. The local institutional review board at each center approved the study protocol. Trained abstractors used a standardized, closed-ended data collection form to review the medical records and collected data on demographic factors, medical history, current and prior obstetric history, birth weight, major and minor complications, treatment, and outcome of the index pregnancy. Our aim in the present study was to examine the effect of annual hospital volume on adverse VBAC outcomes among women who attempted a trial of labor after having at least one prior cesarean. Therefore, this analysis was limited to a cohort of 12,844 women who attempted a VBAC with a singleton birth. Twin and higher-order gestations were excluded because multiple gestations may have differential risks of VBAC outcomes than singleton gestations.
We used tertiles of the distribution of annual hospital VBAC volume (<70, 70–100, or >100) as the primary exposure of interest. Hospitals with more than 100 VBAC deliveries annually were chosen as the reference group for hospital volume. The primary outcomes of interest included VBAC failure, uterine rupture, and a composite variable pertaining to VBAC-related major maternal morbidity. Failure of VBAC was defined as the need for a cesarean section after an attempted trial of labor. Uterine rupture was the main outcome for the primary analysis and was strictly defined a priori as separation of the uterine scar determined at laparotomy, immediately preceded by either a nonreassuring fetal heart rate tracing or maternal hemodynamic instability (systolic blood pressure <70 mmHg, diastolic blood pressure <30 mmHg, pulse rate >120 beats/minute, or intraoperative identification of hemoperitoneum). This definition excluded asymptomatic uterine dehiscence and uterine "windows." Major complications were defined as uterine rupture; operative injury such as bladder, bowel, or ureter injury; and uterine artery laceration (14). A binary composite variable for major maternal morbidity was generated for patients with at least one major complication. Women with multiple major complications were counted once and coded as having a major complication; we did not distinguish between those with a single adverse event and those with multiple adverse events. This composite outcome variable was used to globally assess the most severe and clinically significant operative morbidity. Factors that were potentially associated with the outcomes on the basis of empirical data were also identified. They included maternal race/ethnicity, insurance type, medical illnesses (preeclampsia, gestational hypertension, diabetes mellitus), labor method (spontaneous, induced, or augmented), obstetric history (prior vaginal delivery, 2 prior cesarean deliveries), birth weight, and hospital setting and residency program affiliation (university hospital, community hospital with a residency program, or community hospital without a residency program). Preeclampsia was defined as pregnancy-associated hypertension with proteinuria requiring magnesium sulfate for seizure prophylaxis. Gestational hypertension was defined as acutely elevated blood pressure in the third trimester without proteinuria.
Statistical analysis
Differences in sample characteristics by tertile of annual VBAC hospital volume were assessed through proportion comparisons using the Pearson chi-squared test. We used multivariable logistic regression to examine relations between tertiles of annual hospital VBAC volume and the outcomes of interest. Because differences in patient mix may be an important source of variation in the occurrence of outcomes among different hospital VBAC volume groups, we adjusted for a number of patient characteristics in the multivariable regression analysis. They included maternal race/ethnicity, insurance type, preeclampsia, gestational hypertension, diabetes, labor method, prior vaginal delivery, 2 prior cesarean deliveries, birth weight, and hospital setting and residency program affiliation. Because gestational age may not be as reliably coded as birth weight, gestational age was not included in the multivariable analysis. We used the patient as the unit of analysis, with annual volume being measured at the hospital level. Patients were clustered by hospital. The standard errors, 95 percent confidence intervals, and associated significance levels for adjusted odds ratios based on logistic regression were adjusted for the nested data structure using the robust cluster variance estimator (15). All tests were two-tailed, and p < 0.05 was considered significant. All statistical analyses were performed with STATA, version 9.0 (Stata Corporation, College Station, Texas).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The characteristics of study participants are summarized by hospital VBAC volume in table 1. More than half of the VBAC deliveries in the study sample took place in hospitals in the highest tertile of annual VBAC volume. Hospitals in the first and second tertiles of annual VBAC volume accounted for 12 percent and 29 percent of total VBAC volume, respectively. Failed VBAC occurred slightly more frequently among hospitals in the highest tertile of VBAC volume (25.4 percent), followed by those in the first tertile (23.4 percent) and the second tertile (22.7 percent) (
2 = 10.47, p < 0.01). The proportion of uterine rupture was higher in the second tertile of VBAC volume than in the other two tertiles (1.3 percent vs. 0.9 percent), although this difference was of borderline significance (2 = 5.67, p = 0.06). Composite major maternal morbidity occurred most frequently in hospitals in the upper two tertiles of hospital VBAC volume (2 = 6.91, p = 0.03). Compared with hospitals in the higher tertiles of VBAC volume, patients from hospitals in the lowest tertile were more likely to be Caucasian, to have health maintenance organization or private insurance coverage, to have spontaneous or augmented labor, and to deliver a baby with a birth weight of 4,000 g or more. Patients from hospitals in the second tertile of VBAC volume were more likely to have Medicaid coverage, to have a prior vaginal delivery, and to deliver a baby with a birth weight of 2,500–3,999 g. Hospitals in the highest tertile of VBAC volume had a higher proportion of patients who were Hispanic, had induced labor, and delivered a low birth weight baby (<2,500 g).
|
Table 2 details selected characteristics of the 17 hospitals participating in the study. Hospitals in the lowest tertile of VBAC volume tended to be community hospitals without a residency program, whereas hospitals in the second tertile were more often community hospitals with a residency program and hospitals in the highest tertile were more likely to be university hospitals. Compared with hospitals in the upper tertiles, hospitals in the lowest tertile were less likely to have 24-hour in-house anesthesia, blood bank support, an obstetrician/gynecologist, a neonatal intensive care unit, maternal and fetal medicine staffing, and more than 2,000 deliveries per year.
|
Table 3 shows the unadjusted rate of trial of labor by hospital type for all women with a prior cesarean delivery. The rates of attempting VBAC varied by individual hospital type among patients who had a prior cesarean delivery, from a low of 42 percent to a high of 71 percent. Nonetheless, the proportions of VBAC-eligible women who actually underwent a trial of labor were relatively similar, with rates of 55.2 percent, 51.2 percent, and 57.3 percent for hospitals in the first, second, and third tertiles of annual VBAC volume, respectively.
|
The crude analysis showed that annual hospital volume was not associated with the odds of failed VBAC, uterine rupture, or maternal morbidity (table 4). Because of the possible differences in patient mix and the distribution of risk factors across tertiles of hospital VBAC volume, the results were adjusted for variables potentially associated with outcomes of interest. After adjustment for covariates, annual hospital VBAC volume was not statistically significantly associated with the likelihood of failed VBAC, uterine rupture, or maternal morbidity (table 5). However, hospitals with lower VBAC volume had a borderline-significant protective effect against composite major maternal morbidity (adjusted odds ratio = 0.53, 95 percent confidence interval: 0.28, 1.02). This analysis corroborated the findings of prior studies performed within this cohort, which indicated that maternal race, number of prior cesarean sections, prior vaginal delivery, labor method, preeclampsia, diabetes, and birth weight were consistently associated with an increase or decrease in risk for one or more VBAC-related complications (table 5).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
In general, in this study we did not find evidence of a relation between annual hospital VBAC volume and failed VBAC or uterine rupture among women with a prior cesarean section undergoing a trial of labor. We did, however, observe a borderline statistically significant protective effect against VBAC-related major maternal morbidity among women undergoing a trial of labor in hospitals in the lowest tertile. The latter finding was surprising and somewhat counterintuitive, as table 2 indicated that hospitals in the highest tertile arguably had better facilities with which to respond to potentially adverse outcomes during delivery. The patient mix characteristics, nonetheless, showed that patients from hospitals in the highest tertile, as compared with those in the lowest tertile, were more likely to be of minority race/ethnicity, to be diabetic, to have labor induced, and to have Medicaid coverage (table 1). Those factors could have an effect on either complications of VBAC or VBAC success. In other words, the observed protective effect of undergoing a trial of labor in hospitals in the lowest tertile may reflect residual confounding from a lower-risk patient population as compared with hospitals in the higher tertiles. Another explanation could be that resources and staff in hospitals in the higher tertiles are so stretched or overworked as to lead to deterioration in outcomes.
The volume-outcome relation has been observed for several surgical procedures and medical conditions, mainly focusing on mortality and long-term patient survival (13). Several explanations have been proposed for the association of higher hospital volume with improved outcomes in those hospitals. They include the effect of "learning by doing" and maintenance of skills ("practice makes perfect"), economies of scale attributable to increased specialization within a hospital, and selective referral of patients to hospitals with better outcomes (16). However, publication bias—that is, a greater tendency to publish studies with positive outcomes than studies with negative outcomes—cannot be ruled out. Similarly, investigators who fail to demonstrate significant associations may choose not to pursue or report their analyses.
Prior research that has examined the volume-outcome relation in maternity care has shown mixed results. Using a large clinical sample, Klein et al. (17) did not find evidence of a relation between physician delivery volume and maternal complications at delivery. In a study based on birth certificate data, LeFevre (18) observed that physician volume in obstetrics was not statistically significantly associated with perinatal and neonatal death. On the contrary, Heaphy and Bernard (19) analyzed hospital discharge data and showed that hospitals with higher delivery volume were significantly less likely to have higher rates of maternal complications in normal labor and delivery. Similarly, Garcia et al. (20) observed a dose-response relation between hospital volume and complications of delivery in which the likelihood of complications decreased as volume increased. Bouvier-Colle et al. (21) surveyed three regions of France and reported that a low volume of deliveries (<500 births per year) was associated with substandard care in cases of severe morbidity due to obstetric hemorrhage. Conversely, in a recent study based on a large administrative database, Grobman et al. (22) reported that a higher annual obstetric volume was associated with higher risk of maternal lacerations and injury at delivery.
Among the few studies that have examined the volume-outcome relation in VBAC, a positive relation between high-volume obstetric units and better outcomes has been reported. In a large retrospective cohort study of Canadian women with a prior cesarean delivery, Wen et al. (23) showed that relative to women who elected repeat cesarean delivery, those undergoing a trial of labor in low-volume hospitals had a nearly twofold increased risk of uterine rupture compared with those who delivered in high-volume hospitals. In an Australian population study, Stone et al. (24) reported that higher hospital volume was associated with a higher likelihood of women with prior cesarean undergoing a trial of labor, but obstetric volume was not predictive of VBAC success. In another study, McMahon et al. (25) showed that Canadian women with a prior cesarean section undergoing a trial of labor in smaller community hospitals had risks of failed VBAC that were increased by 50 percent relative to a trial of labor in tertiary-care facilities. On the basis of a MEDLINE database search, a manual search of published abstracts from annual meetings of the Society for Maternal-Fetal Medicine (2000–2007), and the limited number of completed multicenter observational studies with sample sizes adequate to address this question, we feel that the likelihood of significant publication bias affecting this body of literature is minimal (23–27). Thus, our negative study results contradicting prior research efforts are more likely to be explained by population differences or perhaps selection or information bias than publication bias.
Prior studies that adjusted for patient mix more adequately were less likely to report a positive effect of hospital volume on outcome than were studies that did not adjust for patient mix differences in the sample (28). Unmeasured differences in patient population case mix between hospitals with different volumes could result in misleading results due to confounding. The findings in the present study failed to demonstrate a significant association between hospital volume and adverse VBAC outcomes. The observed negative findings may be due to inadequate adjustment for patient mix in the study population, the existence of maternity care policies and procedures at low-volume hospitals, or the presence of a lower-risk study population in low-volume hospitals. Also of relevancy is the issue of whether a volume threshold exists, above which outcomes are better but do not continue to improve with further volume increases (28). Prior research on evaluating hospital volume and medical procedures or conditions has typically employed widely varying definitions of high and low volume (28). The variations in cutoff points for hospital volume categories may explain the differential findings regarding the effect of hospital volume across studies. In addition to patient and hospital system factors (shown in tables 1 and 2), physician-level characteristics are also likely to influence the success of VBAC (28). They include the skills and experience of the various relevant clinicians and litigation pressure. Such information was unavailable in the present study. In the United States, a team management approach is adopted for patient care. A given pregnancy and labor and delivery are typically managed by a team of physicians and health-care personnel to optimize maternal and fetal outcomes. As a result, it is difficult to delineate a specific effect of physician-level characteristics on VBAC patient outcomes.
In the interpretation of our study findings, the following limitations must be considered. First, all of the information in this study was derived from medical records for the index pregnancy, including information on clinical and obstetric history. Variability in the interpretation and coding of obstetric complications is possible, which could have resulted in misclassification errors for these variables. We believe that such misclassification would not have been systematic and may have biased our results towards the null. Second, selection bias could be present because of the retrospective study design. Referral patterns, local practice in counseling, and selection of patients for a trial of labor may vary by physician and hospital type. Our analysis indicated that the VBAC rates between tertiles of annual hospital VBAC volume were similar, which suggests that the potential effect of selection bias on the observed effect estimate may be minimal (table 3). Our previous work in the same study population showed that the odds of trial of labor varied by maternal age, race/ethnicity, gravidity, number of previous cesarean deliveries, and Medicaid status (12). Greenland et al. (29) suggested that if selection factors are antecedents of both the outcome and the exposure under study and are measured in all study subjects, selection bias may be controlled as if it were a confounding factor within the study analysis. Accordingly, in the present study we included in the multivariable analysis the variables that were found to be predictive of differential VBAC attempt rates.
Our study had several strengths as well. First, it was based on a large and diverse population, which makes our findings more generalizable. Second, detailed information was obtained for each patient, which provided us with relatively comprehensive data on patient characteristics for patient mix adjustment. Third, we collected our data directly from patient records, rather than relying on birth certificate data or other administrative data sets to minimize measurement error. Fourth, the data quality for this study was enhanced by trained abstractors who followed strict, detailed, a priori definitions of the outcome variables and potentially confounding variables, minimizing information bias.
In conclusion, our study suggests that the risk of adverse VBAC outcomes in low-volume hospitals was comparable to that in high-volume hospitals. This information can aid clinicians in choosing care providers for women with a prior cesarean section who are considering a trial of labor. Women with a prior cesarean section who are served by small community hospitals should be considered for a trial of labor as long as the facility and patient meet the standard selection criteria described by the American College of Obstetrics and Gynecologists (30). In future studies, investigators need to clarify the existence of a volume threshold and specify the volume level required to maintain proficiency in VBAC delivery once it has been acquired. Future researchers should also aim to gain a better understanding of those critical physician-level factors and organizational processes that affect VBAC outcomes, which may be differential across high- and low-volume hospitals. Such knowledge could provide a blueprint for improvement in obstetric care and management for patients who are candidates for VBAC.
|
ACKNOWLEDGMENTS |
|---|
This study was supported by a K12 Multidisciplinary Clinical Research Career Development Program Scholar award from the National Institutes of Health.
Conflict of interest: none declared.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
- Hamilton BE, Ventura SJ, Martin JA, et al. Preliminary births for 2004. (2004) Hyattsville, MD: National Center for Health Statistics.
- Public Health Service, US Department of Health and Human Services. Healthy People 2010. (2000) Washington, DC: US Department of Health and Human Services.
- Depp R. Cesarean delivery. In: Obstetrics: normal and problem pregnancies—Gabbe SG, Niebyl JR, Simpson JL, eds. (2002) 4th ed. Philadelphia, PA: Churchill-Livingstone. 539–606.
- Martin JA, Hamilton BE, Sutton PD, et al. Births: final data for 2004. (National vital statistics reports, vol 55, no. 1). (2006) Hyattsville, MD: National Center for Health Statistics.
- Silver RM, Landon MB, Rouse DJ, et al. Maternal morbidity associated with multiple repeat cesarean deliveries. Obstet Gynecol (2006) 107:1226–32.[CrossRef][Web of Science][Medline]
- Chauhan SP, Martin JN, Henrichs CE, et al. Maternal and perinatal complications with uterine rupture in 142,075 patients who attempted vaginal birth after cesarean delivery: a review of the literature. Am J Obstet Gynecol (2003) 189:408–17.[CrossRef][Web of Science][Medline]
- Marchiano D, Elkousy M, Stevens E, et al. Diet-controlled gestational diabetes mellitus does not influence the success rates for vaginal birth after cesarean delivery. Am J Obstet Gynecol (2004) 190:790–6.[CrossRef][Web of Science][Medline]
- Macones G, Hausman N, Edelstein R, et al. Predicting outcomes of trials of labor in women attempting vaginal birth after cesarean delivery: a comparison of multivariate methods with neural networks. Am J Obstet Gynecol (2001) 184:409–13.[CrossRef][Web of Science][Medline]
- Elkousy M, Sammel M, Stevens E, et al. The effect of birth weight on vaginal birth after cesarean delivery success rates. Am J Obstet Gynecol (2003) 188:824–30.[CrossRef][Web of Science][Medline]
- Carroll CS Sr, Magann EF, Chauhan SP, et al. Vaginal birth after cesarean section versus elective repeat cesarean delivery: weight-based outcomes. Am J Obstet Gynecol (2003) 188:1516–20. discussion 1520–2.[CrossRef][Web of Science][Medline]
- Hashima J, Eden K, Osterweil P, et al. Predicting vaginal birth after cesarean delivery: a review of prognostic factors and screening tools. Am J Obstet Gynecol (2004) 190:547–55.[CrossRef][Web of Science][Medline]
- Dunsmoor-Su R, Sammel M, Stevens E, et al. Impact of sociodemographic and hospital factors on attempts at vaginal birth after cesarean delivery. Obstet Gynecol (2003) 102:1358–65.[CrossRef][Web of Science][Medline]
- Dudley RA, Bae RY, Johansen KL, et al. When and how should purchasers seek to selectively refer patients to high quality hospitals? In: Interpreting the volume-outcome relationship in the context of health care quality—Hewitt M, ed. (2000) Washington, DC: National Academy Press. 79–94. (Proceedings of a National Academy of Sciences workshop, May 11, 2000).
- McMahon MJ, Luther ER, Bowes WA, et al. Comparison of a trial of labor with an elective second cesarean section. N Engl J Med (1996) 335:689–95.
[Abstract/Free Full Text] - Williams RL. A note on robust variance estimation for cluster-correlated data. Biometrics (2000) 56:645–6.[CrossRef][Web of Science][Medline]
- Hobar JD, Badger GJ, Lewit EM, et al. Hospital and patient characteristics associated with variation in 28-day mortality rates for very low birth weight infants. Pediatrics (1997) 99:149–56.
[Abstract/Free Full Text] - Klein MC, Spence A, Kaczorowski J, et al. Does delivery volume of family physicians predict maternal and newborn outcome? CMAJ (2002) 166:1257–63.
[Abstract/Free Full Text] - LeFevre ML. Physician volume and obstetric outcome. Med Care (1992) 30:866–71.[CrossRef][Web of Science][Medline]
- Heaphy PE, Bernard SL. Maternal complications of normal deliveries: variation among rural hospitals. J Rural Health (2000) 16:139–47.[Web of Science][Medline]
- Garcia FA, Miller HB, Huggins GR, et al. Effect of academic affiliation and obstetric volume on clinical outcome and cost of childbirth. Obstet Gynecol (2001) 97:567–76.[CrossRef][Web of Science][Medline]
- Bouvier-Colle MH, Ould El Joud D, Varnoux N, et al. Evaluation of the quality of care for severe obstetrical haemorrhage in three French regions. BJOG (2001) 108:898–903.[CrossRef][Medline]
- Grobman WA, Feinglass J, Murthy S. Are the Agency for Healthcare Research and Quality obstetric trauma indicators valid measures of hospital safety? Am J Obstet Gynecol (2006) 195:868–74.[CrossRef][Web of Science][Medline]
- Wen SW, Rusen ID, Walker M, et al. Comparison of maternal mortality and morbidity between trial of labor and elective cesarean section among women with previous cesarean delivery. Am J Obstet Gynecol (2004) 191:1263–9.[CrossRef][Web of Science][Medline]
- Stone C, Halliday J, Lumlery J, et al. Vaginal births after caesarean (VBAC): a population study. Paediatr Perinat Epidemiol (2000) 14:340–8.[CrossRef][Web of Science][Medline]
- McMahon MJ, Luther ER, Bowes WA Jr, et al. Comparison of a trial of labor with an elective second cesarean section. N Engl J Med (1996) 335:689–95.
[Abstract/Free Full Text] - Macones GA, Peipert J, Nelson DB, et al. Maternal complications with vaginal birth after cesarean delivery: a multicenter study. Am J Obstet Gynecol (2005) 193:1656–62.[CrossRef][Web of Science][Medline]
- Landon MB, Hauth JC, Leveno KJ, et al. Maternal and perinatal outcomes associated with a trial of labor after prior cesarean delivery. N Engl J Med (2004) 351:2581–9.
[Abstract/Free Full Text] - Halm EA, Lee C, Chassin MR. Is volume related to outcome in health care? A systematic review and methodologic critique of the literature. Ann Intern Med (2002) 137:511–20.
[Abstract/Free Full Text] - Greenland S. Basic methods for sensitivity analysis of biases. Int J Epidemiol (1996) 25:1107–16.
[Abstract/Free Full Text] - ACOG Committee on Practice Bulletins–Obstetrics, Porter TF, Zelop CM. Vaginal birth after previous cesarean delivery. (ACOG Practice Bulletin no. 54). Obstet Gynecol (2004) 104:203–12.[Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||