Mark A. Clapp

A Multi-State Analysis of Early-Term Delivery Trends and the Association With Term Stillbirth.

OBJECTIVE: To investigate whether reduction in early-term deliveries was associated with increasing rates of term stillbirth. METHODS: This is a retrospective descriptive analysis of variation in term delivery timing and stillbirth from 2005 to 2011 based on birth certificate and fetal death data. Early-term deliveries (37 0/7–38 6/7 weeks of gestation) as a percentage of total term delivery and term stillbirth rates were calculated for each state, both overall and for low- and high-risk women. We analyzed whether state-level changes in early-term deliveries and term stillbirth were correlated using Pearson correlation coefficients. States were also categorized as high or low reduction (above or below the national average) and changes in stillbirth rates for these groups were analyzed using a Cochrane-Armitage test for linear trend. RESULTS: There was a decline in early-term deliveries across the United States: 1,123,467 of 3,533,233 term, singleton births occurred in the early term in 2005 (31.8%) as compared with 978,294 of 3,429,172 (28.5%) in 2011. Reductions varied widely by state. There was no change in the term stillbirth rate (123/100,000 births in 2005 compared with 130/100,000 in 2011; P=.189) nor change in the high reduction states alone. There was no correlation between state-level changes in early-term deliveries and term stillbirth. There was an increase in term stillbirths among women with diabetes (from 238/100,000 to 300/100,000 births; P=.010), independent of changes in early-term delivery timing. CONCLUSION: The reduction in early-term deliveries across the United States between 2005 and 2011 was not associated with an increase in the rate of term stillbirth. LEVEL OF EVIDENCE: II

Obstetrician volume as a potentially modifiable risk factor for cesarean delivery.

OBJECTIVE: To examine the relationship between an obstetricians delivery volume and a patients risk for cesarean delivery. METHODS: This retrospective cohort study examined patient-level and obstetrician-level data between 2000 and 2012 at a large academic hospital. All laboring patients who delivered viable, liveborn, singleton newborns (N=58,328) were included. We measured the association of delivery volume and cesarean delivery using a multivariate logistic regression. We also assessed the association of volume by calculating adjusted cesarean delivery rates using the least squares means method. These analyses were performed on the subset of nulliparous patients with term, singleton, vertex-presenting fetuses. In addition, the association of obstetrician experience was compared against delivery volume. RESULTS: There was a twofold increase in the odds of cesarean delivery for patients whose obstetricians performed fewer than the median (60) number of deliveries per year (quartile 1: odds ratio 2.00, 95% confidence interval 1.68–2.38; quartile 2: odds ratio 2.73, 95% CI 2.40–3.11) as compared with quartile 4. The adjusted cesarean delivery rate decreased from 18.2% to 9.2% from the highest to lowest volume quartile (P<.001). Compared with the volume effects, an obstetricians experience had a smaller effect on a patients risk of cesarean delivery. CONCLUSION: Patients delivered by obstetricians with low delivery volume are at significantly increased risk for cesarean delivery after controlling for patient and obstetrician characteristics. In contrast, obstetrician experience had a less significant effect. These findings may prompt discussions regarding the role of volume in credentialing and practice models that direct patients to obstetricians with high delivery volume. LEVEL OF EVIDENCE: II

The effect of adolescence and advanced maternal age on the incidence of complete and partial molar pregnancy

OBJECTIVE To compare the age-specific incidence of complete (CM) and partial molar (PM) pregnancy in a large tertiary care center in the United States. METHODS Incidence rates of CM and PM per 10,000 live births were calculated using databases from Brigham and Womens Hospital, between 2000 and 2013. Age-specific rates were calculated for women younger than 20 years old (adolescents), 20-39 years old (average age), and 40 years and older (advanced maternal age). Pearson χ(2) test was used to evaluate potential differences among groups. Rate ratios (RR) and 95% confidence intervals (CI) were used to compare risk of molar pregnancy among average age women with that of adolescents and women of advanced age. Holm-Bonferonni adjustment was used to correct for multiple comparisons. RESULTS Between 2000 and 2013, there were 255 molar pregnancies (140 CM and 115 PM) and 105,942 live births, corresponding to a molar pregnancy rate of 24 per 10,000 live births (95% CI 21-27). Rates of CM and PM were 13 (95% CI 11-16) and 11 (95% CI 9-14) per 10,000 live births respectively. The incidence of CM differed significantly among maternal age groups (p<0.001). Compared to average age women, adolescents were 7.0 times as likely to develop CM (95% CI 3.6-8.9, p<0.001), and women with advanced maternal age were nearly twice as likely (1.9, 95% CI 1.8-4.7, p=0.002). The rate of PM did not vary significantly among age groups (p=0.26). CONCLUSIONS Adolescence and advanced maternal age were associated with increased risk of complete mole, but not partial mole.

Effect of race/ethnicity on risk of complete and partial molar pregnancy after adjustment for age

OBJECTIVE To quantify the effect of race/ethnicity on risk of complete and partial molar pregnancy. METHODS We conducted a cross-sectional study including women who were followed for complete or partial mole and those who had a live singleton birth in a teaching hospital in the northeastern United States between 2000 and 2013. We calculated race/ethnicity-specific risk of complete and partial mole per 10,000 live births, and used logistic regression to estimate crude and age-adjusted relative risks (RR) of complete and partial mole. RESULTS We identified 140 cases of complete mole, 115 cases of partial mole, and 105,942 live births. The risk of complete mole was 13 cases per 10,000 live births (95% confidence interval [CI] 11-16) and that of partial mole was 11 cases per 10,000 live births (95% CI 9-13). After age-adjustment, Asians were more likely to develop complete mole (RR 2.3 95% CI 1.4-3.8, p<0.001) but less likely to develop partial mole (RR 0.2; 95% CI 0.04-0.7, p=0.02) than whites. Blacks were significantly less likely than whites to develop partial mole (RR 0.4; 95% CI 0.2-0.8, p=0.01) but only marginally less likely to develop complete mole (RR 0.6; 95% CI 0.3-1.0, p=0.07). Hispanics were less likely than whites to develop complete mole (RR 0.4; 95% CI 0.2-0.7, p=0.002) and partial mole (RR 0.4; 95% CI 0.2-0.9, p=0.02). CONCLUSION Race/ethnicity is a significant risk factor for both complete and partial molar pregnancy in the northeastern United States.

The Relationship Between Intertwin Membrane Separation and Pregnancy Outcome.

OBJECTIVE: To evaluate the association of intertwin membrane separation and pregnancy outcome. METHODS: This is a retrospective cohort study of women with dichorionic twins who were diagnosed with spontaneous intertwin membrane separation between 2004 and 2013 at a large tertiary care maternity hospital. Control participants were selected as the next two sets of dichorionic twins that delivered at the study institution after a case participant delivered and that did not have an intrauterine procedure. Maternal, fetal, and delivery characteristics were compared using Wilcoxon rank-sum tests. Logistic regressions were used to assess the association of membrane separation and preterm delivery. RESULTS: Among the 27 cases of spontaneous intertwin membrane separation, the median gestational age at diagnosis was 28 weeks (interquartile range 25.5–28.8) and the median gestational age at delivery was 37 weeks (interquartile range 35.3–37.0). The rate of preterm delivery, our primary outcome, was 48% for the case group and 76% for the control group (odds ratio [OR] 0.29, P=.01). The rate of spontaneous preterm delivery was also lower for the case group (19% compared with 44%; OR 0.25, P=.26) as was the rate of neonatal intensive care unit admission (37% compared with 61%; OR 0.37, P=.04). CONCLUSION: Our retrospective cohort study demonstrates that intertwin membrane separation is not associated with adverse outcomes in dichorionic twin pregnancies. Thus, it is reasonable to manage these pregnancies expectantly. LEVEL OF EVIDENCE: II

The relationship between the rising cesarean delivery and postpartum readmission rates

Objective:This study seeks to determine if the increasing rate of postpartum readmissions is related to the increasing rate of cesarean delivery.Study Design:Readmitted patients were identified in the State Inpatient Databases of California, Florida and New York from 2004 to 2011. Relevant maternal comorbidities, pregnancy complications and intrapartum events were collected using ICD-9 diagnosis and procedure codes. The effects of cesarean delivery were first examined via univariate logistic regression to calculate the odds of readmission by year for patients who had delivered via cesarean section. Then, we used multivariate logistic regression models to isolate the effect of mode of delivery on the odds of readmission by adjusting for the effects of patient demographics, hospital characteristics and maternal comorbidities.Results:Nearly one million deliveries were identified each year, and ~600 000 deliveries per year met inclusion criteria. During this time, the readmission rate increased from 1.72 to 2.16%, and the cesarean delivery rate increased from 30.4 to 33.9%. The odds of readmission for patients delivered via cesarean section decreased yearly, from 1.343 (95% CI: 1.295 to 1.392) in 2004 to 1.046 (95% CI: 1.012 to 1.108) in 2011. In a multivariate model, the odds based on year were 1.032 (95% CI: 1.030 to 1.035), demonstrating an increased odds of readmission over time. When cesarean delivery was added to the model, this odds estimate did not change (OR: 1.031, 95% CI: 1.028 to 1.035), suggesting it did not account for the increased odds of readmission over time, even though cesarean delivery rates increased. However, when maternal comorbidities were added to the model, the odds ratio for year became insignificant (OR: 1.001, 95% CI: 0.998 to 1.005), suggesting that they accounted for the increasing rate of readmissions.Conclusions:The increasing cesarean delivery rate does not explain the increasing rate of postpartum readmissions. Rather, the increasing postpartum readmission rate appears to be related to maternal comorbidities.

Hospital-Level Variation in Postpartum Readmissions

Readmission rates are used as a quality indicator and linked to reimbursement for certain medical and surgical conditions.1 Obstetric maternal readmissions have not been rigorously studied as a quality measure, though their use has been proposed.2 The goal of this study was to determine the potential utility of this metric and its ability to accurately reflect quality by quantifying (1) the variance in hospital postpartum readmission rates and (2) the percentage of the variance that was attributed to the effect of the hospital after controlling for case mix.

A Multi-State Analysis of Postpartum Readmissions in the United States [26].

INTRODUCTION: To describe the trends in postpartum readmissions over time, to characterize the common indications for readmissions, and to determine maternal, delivery, and hospital characteristics associated with readmission. METHODS: Postpartum readmissions (n=114,748) occurring within the first six weeks after delivery in California, Florida, and New York were identified between 2004–2011 in State Inpatient Databases. We calculated the rates of readmissions and their indications by state and over time. The characteristics of the readmission stay were compared among the diagnoses. Odds ratios were calculated using a multivariate logistic regression to determine the predictors of readmission. RESULTS: The readmission rate increased from 1.72% in 2004 to 2.16% in 2011. Readmitted patients were more likely to be publicly insured (54.3% vs 42.0%, P<.001), Black (18.7% vs 13.5%, P<.001), have comorbidities, and to have had a cesarean delivery (37.2% vs 32.9%, P<.001). The most common indications for readmission were infection (15.5%), hypertension (9.3%), and psychiatric illness (7.7%). Readmission day varied by diagnosis: day three for hypertension, day five for infection, and day nine for psychiatric disease. Maternal comorbidities were the strongest predictors of postpartum readmissions: psychiatric disease (OR 2.542 [95% CI 2.448–2.600]), substance use (OR 2.016 [95% CI 1.958–2.075]), seizure disorder (OR 1.989 [95% CI 1.873–2.113]), hypertension (OR 1.886 [95% CI 1.839–1.934]), and tobacco use (OR 1.859 [95% CI 1.800–1.921]). CONCLUSION/IMPLICATIONS: Understanding the risk factors, etiologies, and cause-specific timing for postpartum readmissions may aid in the development of new quality metrics in obstetrics and targeted strategies to curb the rising rate of postpartum readmissions in the United States.

Effects of mitochondrial disease/dysfunction on pregnancy: A retrospective study

A retrospective survey assessed the gynecologic, obstetric and fertility history of 103 women with mitochondrial disease (MD)/dysfunction (Md) aged 16 to 75 who had previously been pregnant. Most participants (34%) had a mitochondrial myopathy and there were 248 combined pregnancies with live deliveries (average 3.6 pregnancies/woman). In general, pregnancy in those with MD/Md appears to exacerbate some constitutional and neurological symptoms and may be more frequently associated with common obstetric complications, but this did not appear to result in worse pregnancy outcomes. Most women carried their pregnancy to term, but their neonates tended to have more congenital anomalies than expected.

The relative effects of patient and hospital factors on postpartum readmissions

ObjectiveTo determine the relative effects of patient and hospital factors on a hospital’s postpartum readmission rate.Study designThis retrospective cohort study was conducted using State Inpatient Databases from California, Florida, and New York between 2004 and 2013. We compared patient and hospital characteristics among hospitals with low and high readmission rates using χ2 tests. Risk-adjusted 30-day readmission rates were calculated for patient, delivery, and hospital characteristics to understand factors affecting readmission using fixed and random effects models.ResultsPatients in hospitals with low readmission rates were more likely to be white, to have private insurance and higher incomes, and to have fewer comorbidities. The patient comorbidities with the highest risk-adjusted readmission rates included hypertension (range, 2.14–3.04%), obesity (1.78–2.94%), preterm labor/delivery (2.50–2.60%), and seizure disorder (1.78–3.35%). Delivery complications were associated with increased risk-adjusted readmission rates. Compared to patient characteristics, hospital characteristics did not have a profound impact on readmission risk.ConclusionObstetric readmissions were more attributable to patient and demographic characteristics than to hospital characteristics. Readmission metric-based incentives may ultimately penalize hospitals providing high-quality care due to patient characteristics specific to their catchment area.