Development and Validation of a Risk Prediction Model for Cesarean Delivery After Labor Induction.

Objective: The goal of the study was to develop and validate a prediction model for cesarean delivery after labor induction that included factors known before the start of induction, unlike prior studies that focused on characteristics at the time of induction. Materials and Methods: Using 17,370 term labor inductions without documented medical indications occurring at 14 U.S. hospitals, 2007-2012, we created and evaluated a model predicting cesarean delivery. We assessed model calibration and discrimination, and we used bootstrapping for internal validation. We externally validated the model by using 2122 labor inductions from a hospital not included in the development cohort. Results: The model contained eight variables-gestational age, maternal race, parity, maternal age, obesity, fibroids, excessive fetal growth, and history of herpes-and was well calibrated with good risk stratification at the extremes of predicted probability. The model had an area under the curve (AUC) for the receiver operating characteristic curve of 0.82 (95% confidence interval 0.81-0.83), and it performed well on internal validation. The AUC in the external validation cohort was 0.82. Conclusion: This prediction model can help providers estimate a woman's risk of cesarean delivery when planning a labor induction.

Impact of the Agency for Healthcare Research and Quality's Safety Program for Perinatal Care.

BACKGROUND: The Safety Program for Perinatal Care (SPPC) seeks to improve safety on labor and delivery (L&D) units through three mutually reinforcing components: (1) fostering a culture of teamwork and communication, (2) applying safety science principles to care processes; and (3) in situ simulation. The objective of this study was to describe the SPPC implementation experience and evaluate the short-term impact on unit patient safety culture, processes, and adverse events. METHODS: We supported SPPC implementation by L&D units with a program toolkit, trainings, and technical assistance. We evaluated the program using a pre-post, mixed-methods design. Implementing units reported uptake of program components, submitted hospital discharge data on maternal and neonatal adverse events, and participated in semi-structured interviews. We measured changes in safety and quality using the Modified Adverse Outcome Index (MAOI) and other perinatal care indicators. RESULTS: Forty-three L&D units submitted data representing 97,740 deliveries over 10 months of follow-up. Twenty-six units implemented all three program components. L&D staff reported improvements in teamwork, communication, and unit safety culture that facilitated applying safety science principles to clinical care. The MAOI decreased from 5.03% to 4.65% (absolute change -0.38% [95% CI, -0.88% to 0.12%]). Statistically significant decreases in indicators for obstetric trauma without instruments and primary cesarean delivery were observed. A statistically significant increase in neonatal birth trauma was observed, but the overall rate of unexpected newborn complications was unchanged. CONCLUSIONS: The SPPC had a favorable impact on unit patient safety culture and processes, but short-term impact on maternal and neonatal adverse events was mixed.

Relationship Between Labor and Delivery Unit Management Practices and Maternal Outcomes.

OBJECTIVE: To define, measure, and characterize key competencies of managing labor and delivery units in the United States and assess the associations between unit management and maternal outcomes. METHODS: We developed and administered a management measurement instrument using structured telephone interviews with both the primary nurse and physician managers at 53 diverse hospitals across the United States. A trained interviewer scored the managers' interview responses based on management practices that ranged from most reactive (lowest scores) to most proactive (highest scores). We established instrument validity by conducting site visits among a subsample of 11 hospitals and established reliability using interrater comparison. Using a factor analysis, we identified three themes of management competencies: management of unit culture, patient flow, and nursing. We constructed patient-level regressions to assess the independent association between these management themes and maternal outcomes. RESULTS: Proactive management of unit culture and nursing was associated with a significantly higher risk of primary cesarean delivery in low-risk patients (relative risk [RR] 1.30, 95% CI 1.02-1.66 and RR 1.47, 95% CI 1.13-1.92, respectively). Proactive management of unit culture was also associated with a significantly higher risk of prolonged length of stay (RR 4.13, 95% CI 1.98-8.64), postpartum hemorrhage (RR 2.57, 95% CI 1.58-4.18), and blood transfusion (RR 1.87, 95% CI 1.12-3.13). Proactive management of patient flow and nursing was associated with a significantly lower risk of prolonged length of stay (RR 0.23, 95% CI 0.12-0.46 and RR 0.27, 95% CI 0.11-0.62, respectively). CONCLUSION: Labor and delivery unit management varies dramatically across and within hospitals in the United States. Some proactive management practices may be associated with increased risk of primary cesarean delivery and maternal morbidity. Other proactive management practices may be associated with decreased risk of prolonged length of stay, indicating a potential opportunity to safely improve labor and delivery unit efficiency.

Public Health Data in Action: An Analysis of Using Louisiana Vital Statistics for Quality Improvement and Payment Reform.

Introduction In 2012, the Louisiana (LA) Department of Health and Hospitals revised the LA birth certificate to include medical reasons for births before 39 completed weeks' gestation. We compared the completeness and validity of these data with hospital discharge records. Methods For births occurring 4/1/2012-9/30/2012 at Woman's Hospital of Baton Rouge, we linked maternal delivery and newborn birth data collected through the National Perinatal Information Center with LA birth certificates. Among early term births (37-38 completed weeks' gestation), we quantified the reasons for early delivery listed on the birth certificate and compared them with ICD-9-CM codes from Woman's discharge data. Results Among 4353 birth certificates indicating delivery at Woman's Hospital, we matched 99.8% to corresponding Woman's administrative data. Among 1293 early term singleton births, the most common reasons for early delivery listed on the birth certificate were spontaneous active labor (57.5%), gestational hypertensive disorders (15.3%), gestational diabetes (8.7%), and premature rupture of membranes (8.1%). Only 2.7% of births indicated "other reason" as the only reason for early delivery. Most reasons for early delivery had >80% correspondence with ICD-9-CM codes. Lower correspondence (35 and 72%, respectively) was observed for premature rupture of membranes and abnormal heart rate or fetal distress. Discussion There was near-perfect ability to match LA birth certificates with Woman's Hospital records, and the agreement between reasons for early delivery on the birth certificate and ICD-9-CM codes was high. A benchmark of 2.7% can be used as an attainable frequency of "other reason" for early delivery reported by hospitals. Louisiana implemented an effective mechanism to identify and explain early deliveries using vital records.

Integrated Approach to Reduce Perinatal Adverse Events: Standardized Processes, Interdisciplinary Teamwork Training, and Performance Feedback.

OBJECTIVE: To improve safety practices and reduce adverse events in perinatal units of acute care hospitals. DATA SOURCES: Primary data collected from perinatal units of 14 hospitals participating in the intervention between 2008 and 2012. Baseline secondary data collected from the same hospitals between 2006 and 2007. STUDY DESIGN: A prospective study involving 342,754 deliveries was conducted using a quality improvement collaborative that supported three primary interventions. Primary measures include adoption of three standardized care processes and four measures of outcomes. DATA COLLECTION METHODS: Chart audits were conducted to measure the implementation of standardized care processes. Outcome measures were collected and validated by the National Perinatal Information Center. PRINCIPAL FINDINGS: The hospital perinatal units increased use of all three care processes, raising consolidated overall use from 38 to 81 percent between 2008 and 2012. The harms measured by the Adverse Outcome Index decreased 14 percent, and a run chart analysis revealed two special causes associated with the interventions. CONCLUSIONS: This study demonstrates the ability of hospital perinatal staff to implement efforts to reduce perinatal harm using a quality improvement collaborative. Findings help inform the relationship between the use of standardized care processes, teamwork training, and improved perinatal outcomes, and suggest that a multiplicity of integrated strategies, rather than a single intervention, may be essential to achieve high reliability.

Comparing expectant management and spontaneous labor approaches in studying the effect of labor induction on cesarean delivery.

PURPOSE: Evidence of the impact of labor induction on cesarean delivery (CD) remains inconclusive because of differing methodological approaches. A spontaneous labor comparison group describes patterns retrospectively, whereas an expectant management comparison group prospectively evaluates a decision to induce. We examined the influence of comparison group on the association between labor induction and CD. METHODS: We studied 166,559 mother-newborn dyads from 14 National Perinatal Information Center member hospitals, 2007-2012. We included singleton births 34-42 completed weeks' gestation and excluded women with contraindications to vaginal delivery. We calculated risk ratios (RR) adjusted for hypertensive and diabetic disorders, intrauterine growth restriction, parity, and maternal age. RESULTS: When comparing induction to spontaneous labor, induction had significantly lower risk for CD at weeks 34-35 (adjusted RR [95% confidence interval (CI)]: 0.6 [0.5, 0.7] for week 34 and 0.7 [0.6, 0.8] for week 35) and higher risk at weeks 37-41 (adjusted RRs [95% CIs]: 1.8 [1.6, 2.1], 2.1 [1.9, 2.2], 1.8 [1.7, 1.9], 1.9 [1.8, 2.0], and 1.6 [1.5, 1.7], respectively). When comparing induction to expectant management, adjusted RRs [95% CIs] were significantly below 1.0 for week 34 (0.8 [0.7, 0.9]), week 36 (0.9 [0.8, 0.9]), and week 37 (0.9 [0.8, 0.9]), and were only elevated at week 40 (1.4 [1.3, 1.4]) and week 41 (1.4 [1.3, 1.5]). CONCLUSIONS: Using two different methodological approaches with the same sample, we confirm that comparing labor induction to spontaneous onset of labor, instead of expectant management of pregnancy, does not fully inform clinical practice and may lead to an exaggerated estimate of the risk of CD.

Hospital differences in special care nursery use for newborns of gestational diabetic mothers.

OBJECTIVE: Relatively healthy newborns of mothers with gestational diabetes mellitus (GDM) sometimes receive unwarranted surveillance. We studied the relationship between hospital characteristics and special care nursery use and total length of stay among GDM deliveries. METHODS: We identified GDM deliveries at 44 USA member hospitals of the National Perinatal Information Center from 2007 to 2011. To study low risk, relatively healthy newborns with presumed discretion in special care nursery use, we analyzed 43 444 singleton newborns with only minor or moderate complications and WHO were not preterm or low birthweight. RESULTS: Among eligible newborns, 6% received special care, but this ranged from 1% to 16% across 44 hospitals studied. Unadjusted associations suggested special care nursery use was highest in academic teaching hospitals, the Midwest, hospitals with ≥40% Medicaid births, and hospitals with a high supply of special care nursery beds. However, after controlling for clustering within hospitals, there were no significant associations between hospital characteristics and special care nursery use or length of stay. CONCLUSIONS: Hospital-level variation in special care nursery use and length of stay of relatively healthy newborns of mothers with GDM is unexplained by hospital characteristics and suggests other operational or management factors impacting utilization of newborn care resources.