Outcomes of Planned Hospital Birth Attended by Midwives Compared with Physicians in British Columbia

ABSTRACT: Background: The impact of midwifery versus physician care on perinatal outcomes in a population of women planning birth in hospital has not yet been explored. We compared maternal and newborn outcomes between women planning hospital birth attended by a midwife versus a physician in British Columbia, Canada. Methods: All women planning a hospital birth attended by a midwife during the 2‐year study period who were of sufficiently low‐risk status to meet eligibility requirements for home birth as defined by the British Columbia College of Midwives were included in the study group (n =488). The comparison group included women meeting the same eligibility requirements but planning a physician‐attended birth in hospitals where midwives also practiced (n =572). Outcomes were ascertained from the British Columbia Reproductive Care Program Perinatal Registry to which all hospitals in the province submit data. Results: Adjusted odds ratios for women planning hospital birth attended by a midwife versus a physician were significantly reduced for exposure to cesarean section (OR 0.58, 95% CI 0.39–0.86), narcotic analgesia (OR 0.26, 95% CI 0.18–0.37), electronic fetal monitoring (OR 0.22, 95% CI 0.16–0.30), amniotomy (OR 0.74, 95% CI 0.56–0.98), and episiotomy (OR 0.62, 95% CI 0.42–0.93). The odds of adverse neonatal outcomes were not different between groups, with the exception of reduced use of drugs for resuscitation at birth (OR 0.19, 95% CI 0.04–0.83) in the midwifery group. Conclusions: A shift toward greater proportions of midwife‐attended births in hospitals could result in reduced rates of obstetric interventions, with similar rates of neonatal morbidity. (BIRTH 34:2 June 2007)     

Midwifery Workforce Density Moderates the Association Between Independent Practice and Pregnancy Outcomes

Introduction

Studies comparing pregnancy outcomes before and after state transition to independent midwifery practice have found little change in primary cesarean birth and preterm birth rates. One reason may be the failure to control for midwife density. The objective was to test if the local midwife density moderates the association between state independent midwifery practice and pregnancy outcomes.

Methods

Birth records were abstracted from the State Inpatient Databases for 6 states. The Area Health Resource File provided county variables. Midwife density was operationalized as no midwives, low midwife density (<4.5 per 1000 births), and high midwife density (≥4.5 midwives per 1000 births). Multivariate logistic regression models compared primary cesarean birth and preterm birth, controlling for maternal and county characteristics. Moderation was tested by including an interaction term (independent practice × density) to the regression models. The magnitude of association for the interaction was measured by stratifying the models.

Results

The study included 875,156 women; most (79.7%) resided in a county with low midwife density. Restricted midwifery practice was associated with increased odds of both primary cesarean birth and preterm birth. The interaction term was significant for both preterm birth and primary cesarean, indicating moderation. The largest magnitude of difference was the increased odds of preterm birth in counties with a high midwife density and restricted practice (odds ratio, 3.50; 95% CI, 2.43-5.06) compared with those with high midwife density and independent practice.

Discussion

Midwife density moderates the association between independent midwifery practice and primary cesarean birth and preterm birth. Moderation may explain why prior studies found small or no changes in outcomes when states adopted independent practice. Moderation models can improve testing for associations with independent practice. Both midwife independent practice and increasing midwifery workforce size can be strategies to improve state pregnancy outcomes.     

Trends in Midwife‐Attended Births in the United States, 1989‐2009

Introduction: Data on attendance at birth by midwives in the United States have been available on the national level since 1989. Rates of certified nurse‐midwife (CNM)–attended births more than doubled between 1989 (3.3% of all births) and 2002 (7.7%) and have remained steady since. This article examines trends in midwife‐attended births from 1989 to 2009. Methods: The data in this report are based on records gathered as part of the US National Standard Certificate of Live Birth from a public use Web site, Vital Stats ( http://www.cdc.gov/nchs/VitalStats.htm ), that allows users to create and download specialized tables. Results: Between 2007 and 2009, the proportion of all births attended by CNMs increased by 4% from 7.3% of all births to 7.6% and a total of 313,516. This represents a decline in total births attended by CNMs from 2008 but a higher proportion of all births because total US births dropped at a faster rate. The proportion of vaginal births attended by CNMs reached an all‐time high of 11.4% in 2009. There were strong regional patterns to the distribution of CNM‐attended births. Births attended by “other midwives” rose to 21,787 or 0.5% of all US births, and the total proportion of all births attended by midwives reached an all‐time high of 8.1%. The race/ethnicity of mothers attended by CNMs has shifted over the years. In 1990, CNMs attended a disproportionately high number of births to non‐white mothers, whereas in 2009, the profile of CNM births mirrors the national distribution in race/ethnicity. Discussion: Midwife‐attended births in the United States are increasing. The geographic patterns in the distribution of midwife‐attended births warrant further study.     

Perineal Injury in Nulliparous Women Giving Birth at a Community Hospital: Reduced Risk in Births Attended by Certified Nurse‐Midwives

Introduction : Our objective was to determine if there is a difference in rates of perineal injury sustained by nulliparous women attended by obstetricians compared with certified nurse‐midwives (CNMs) at a US community hospital. Methods : We analyzed retrospective data for 2819 women who spontaneously gave birth to singleton, vertex, term, live infants between 2000 and 2005. The independent variable was attendant type (obstetrician or CNM). The main outcome variables were intact perineum, episiotomy, and spontaneous perineal lacerations. Multivariate logistic regression was used to adjust for six potential confounders: macrosomia, maternal age, epidural anesthesia, oxytocin administration, medical insurance status, and ethnicity. Results : The odds ratios (ORs) for obstetrician‐attended births versus CNM‐attended births were significant for a spontaneous minor perineal laceration versus intact perineum (OR = 1.82; 95% confidence interval [CI], 1.33–2.48), spontaneous major laceration versus intact perineum (OR = 2.29; 95% CI, 1.13–4.66), and episiotomy use versus no perineal injury, with or without extension (OR = 2.94; 95% CI, 2.01–4.29). Discussion : We found that the prevalence and severity of perineal injury, both spontaneous and from episiotomy use, were significantly lower in CNM‐attended births. J Midwifery Womens Health 2010;55:243–249 c̊ 2010 by the American College of Nurse‐Midwives.     

“The Midwifery Two‐Step”: A Study on Evidence‐Based Midwifery Practice

Introduction: To date, there has been little documentation of how practice‐based midwifery networks in the United States might influence the transfer and development of knowledge in childbearing and women's health care. The first phase of this participatory action research project was to conduct a qualitative study with a community of midwifery practices to understand their perspectives on evidence‐based practice and how an organized network could facilitate their work. Methods: Midwives within the community of interest were invited by letter or e‐mail to participate in individual or small group interviews about knowledge transfer, primary concerns of evidence‐based practice, and potential for a midwifery practice‐based research network. Participatory action research strategies and organizational ethnographic approaches to data collection were used to guide qualitative interviews. Results: Eight midwifery practices enrolled in the study with 23 midwives participating in interviews. They attended births at 2 hospitals in the community. Two broad areas of discourse about evidence‐based practice were identified: 1) challenges from influential persons, finances and resources, and the cultural perception of midwifery, and 2) strategies to foster best practice in the face of those challenges. The midwives believed a research network could be useful in learning collectively about their practices and in the support of their work. Discussion: Evidence‐based practice is a goal but also has many challenges in everyday implementation. Practice‐based research networks hold promise to support clinicians to examine the evidence and form strong coalitions to foster best clinical practice. The second phase of this study will work with this community of midwives to explore collective strategies to examine and improve practice.     

Work and workload of Dutch primary care midwives in 2010

Objective

to re-assess the work and workload of primary care midwives in the Netherlands.

Background

in the Netherlands most midwives work in primary care as independent practitioners in a midwifery practice with two or more colleagues. Each practice provides 24/7 care coverage through office hours and on-call hours of the midwives. In 2006 the results of a time registration project of primary care midwives were published as part of a 4-year monitor study. This time the registration project was repeated, albeit on a smaller scale, in 2010.

Method

as part of a larger study (the Deliver study) all midwives working in 20 midwifery practices kept a time register 24 hours a day, for one week. They also filled out questionnaires about their background, work schedules and experiences of workload. A second component of this study collected data from all midwifery practices in the Netherlands and included questions about practice size (number of midwives and number of clients in the previous year).

Findings

in 2010, primary care midwives actually worked on an average 32.6 hours per week and approximately 67% of their working time (almost 22 hours per week) was spent on client-related activities. On an average a midwife was on-call for 39 hours a week and almost 13 of the 32.6 hours of work took place during on-call-hours. This means that the total hours that an average midwife was involved in her work (either actually working or on-call) was almost 59 hours a week. Compared to 2004 the number of hours an average midwife was actually working increased by 4 hours (from 29 to 32.6 hours) whereas the total number of hours an average midwife was involved with her work decreased by 6 hours (from 65 to 59 hours). In 2010, compared to 2001–2004, the midwives spent proportionally less time on direct client care (67% versus 73%), although in actual number of hours this did not change much (22 versus 21). In 2009 the average workload of a midwife was 99 clients at booking, 56 at the start of labour, 33 at childbirth, and 90 clients in post partum care.

Conclusion

the midwives worked on an average more hours in 2010 than they did in 2004 or 2001, but spent these extra hours increasingly on non-client-related activities.     

The motivation and capacity to go ‘above and beyond’: Qualitative analysis of free-text survey responses in the M@NGO randomised controlled trial of caseload midwifery

Objectiveto explore whether women allocated to caseload care characterise their midwife differently to those allocated to standard care.Designmulti-site unblinded, randomised, controlled, parallel-group trial.Settingthe study was conducted in two metropolitan teaching hospitals across two Australian cities.Populationwomen of all obstetric risk were eligible to participate. Inclusion criteria were: 18 years or older, less than 24 week’s gestation with a singleton pregnancy. Women already booked with a care provider or planning to have an elective caesarean section were excluded.Interventionsparticipants were randomised to caseload midwifery or standard care. The caseload model provided antenatal, intrapartum and postnatal care from a primary midwife or ‘back-up’ midwife; as well as consultation with obstetric or medical physicians as indicated by national guidelines. The standard model included care from a general practitioner and/or midwives and obstetric doctors.Measurements and findingsparticipants’ responses to open-ended questions were collected through a 6-week postnatal survey and analysed thematically. A total of 1748 women were randomised between December 2008 – May 2011; 871 to caseload midwifery and 877 to standard care. The response rate to the 6-week survey including free text items was 52% (n=901). Respondents from both groups characterised midwives as Informative, Competent and Kind. Participants in the caseload group perceived midwives with additional qualities conceptualised as Empowering and ‘Endorphic’. These concepts highlight some of the active ingredients that moderated or mediated the effects of the midwifery care within the M@NGO trial.Key conclusioncaseload midwifery attracts, motivates and enables midwives to go Above and Beyond such that women feel empowered, nurtured and safe during pregnancy, labour and birth.Implications for practicethe concept of an Endorphic midwife makes a useful contribution to midwifery theory as it enhances our understanding of how the complex intervention of caseload midwifery influences normal birth rates and experiences. Defining personal midwife attributes which are important for caseload models has potential implications for graduate attributes in degree programs leading to registration as a midwife and selection criteria for caseload midwife positions.     

Improving Quality and Safety in Maternity Care: The Contribution of Midwife‐Led Care

This article draws on findings from a recent Cochrane systematic review of midwife‐led care and discusses its contribution to the safety and quality of women's care in the domains of safety, effectiveness, woman‐centeredness, and efficiency. According to the Cochrane review, women who received models of midwife‐led care were nearly eight times more likely to be attended at birth by a known midwife, were 21% less likely to experience fetal loss before 24 weeks' gestation, 19% less likely to have regional analgesia, 14% less likely to have instrumental birth, 18% less likely to have an episiotomy, and significantly more likely to have a spontaneous vaginal birth, initiate breastfeeding, and feel in control. In addition to normalizing and humanizing birth, the contribution of midwife‐led care to the quality and safety of health care is substantial. The implications are that policymakers who wish to improve the quality and safety of maternal and infant care, particularly around normalizing and humanizing birth, should consider midwife‐led models of care and how financing of midwife‐led services can support this. Suggestions for future research include exploring why fetal loss is reduced under 24 weeks' gestation in midwife‐led models of care, and ensuring that the effectiveness of midwife‐led models of care on mothers' and infants' health and well‐being are assessed in the longer postpartum period.     

Risk factors and midwife-reported reasons for episiotomy in women undergoing normal vaginal delivery

Purpose

In Singapore, a developed Asian nation, a relatively high proportion of women undergo episiotomy. We assess risk factors and midwife-reported reasons for episiotomy among women undergoing normal vaginal deliveries (NVDs) conducted by midwives and ascertain the association between episiotomy and degree of perineal tear.

Methods

Participants included 77 midwives from a high-volume delivery unit in Singapore. The study had three sequential phases: (1) medical record review of women undergoing NVDs conducted by midwives over a 1-month period to document the proportion with episiotomy; (2) focus group discussions with midwives to form a checklist of reasons for episiotomy; (3) checklist-based documentation of midwife-reported reasons for episiotomy and data collection on maternal, neonatal, practice and midwife factors, and degree of perineal tear among women undergoing NVDs conducted by midwives over a 2-month period. Risk factors for episiotomy were assessed through logistic regression.

Results

Primiparity, advanced maternal age, Indian ethnicity, higher birth weight and older midwife age were associated with episiotomy. The most common midwife-reported reason for episiotomy among primiparous women was primiparity (55.1 %), and among multiparous women was fetal distress (20.0 %) and poor maternal effort (20.0 %). All women with episiotomy sustained at least a second-degree perineal tear versus 27.1 % among women without episiotomy.

Conclusion

Most midwife-reported reasons for episiotomy were not congruent with international practice guidelines. Women without episiotomy have lesser tears than those with episiotomy. Practice protocols and educational programs are needed to change episiotomy practice.     

Midwife-led care unit for ‘low risk’ pregnant women in a Japanese hospital

Objectives.?To examine the obstetric outcomes of our ‘low risk’ pregnant women under the midwife-led delivery care compared with those under the obstetric shared care.

Methods.?A retrospective cohort study compared outcomes of labor under midwife ‘primary’ care with those under obstetric shared care. The factors examined were: maternal age, parity, gestational age at delivery, length of labor, augmentation of labor pains, delivery mode, episiotomy, perineal laceration, postpartum hemorrhage, neonatal birth weight, Apgar score, and umbilical artery pH. In this study, pregnant women were initially considered ‘low risk’ at admission when they had no history of medical, gynecological, or obstetric problems and no complications during the present pregnancy.

Results.?There were 1031 pregnant women initially considered ‘low risk’ at admission. At admission, 878 of them (85%) requested to give birth under midwife care; however 364 of these women (42%) were transferred to obstetric shared care during labor. The average length of labor under the midwife ‘primary’ care was significantly longer than that under the obstetric shared care. However, there were no significant differences in the rate of prolonged labor (≥24?h). There were no significant differences in other obstetric or neonatal outcomes between the two groups.

Conclusions.?There was no evidence indicating that midwife ‘primary’ care is unsafe for ‘low risk’ pregnant women. Therefore, midwifery care is recommended for ‘low risk’ pregnant women.     

Intrapartum care at a tertiary hospital in Cambodia: A survey using the Bologna Score

Background

developing countries strive to reduce maternal- and child mortality, partly through establishing health centres/hospitals with skilled birth attendants. The aim of this study was to describe childbirth care, by the use of the Bologna Score at a tertiary hospital in Cambodia with approximately 8,500 births per year.

Methods

a prospective cross-sectional study. The Bologna Score instrument, which reflects the adaption of evidence-based care and attitudes of caregivers, was used for data collection and three study specific questions. The midwives collected data from 177 consecutive childbirths.

Results

all women were assisted by a skilled birth attendant, the majority by a midwife (63%) and the remaining women by a physician (35%) or midwife student under supervision. A spontaneous vaginal birth was planned for 82% of the women. All women seeking care at the hospital survived the childbirth. A full 5-point Bologna Score, suggesting evidence-based management for women with spontaneous vaginal birth, was not achieved for any of the women. The use of supine position and lack of an accompanying person in the birth room, were items responsible for loss of points. Partogram and skin-to-skin contact between baby and mother were items noted for three quarters of the planned vaginal births, and the item ‘Absence of labour augmentation’, was affirmed to a great extent. Little more than half of the women had an episiotomy and almost 16% of the children had an Apgar score <7 at 5 mins.

Conclusion

the Bologna Score was easy to use and pointed at items that could be improved. It was satisfying that all women survived, but alarming that 16% of the children had a low Apgar score. The findings suggest that childbirth care can be improved at the hospital.     

Being safe practitioners and safe mothers: A critical ethnography of continuity of care midwifery in Australia

Objective

to examine how midwives and women within a continuity of care midwifery programme in Australia conceptualised childbirth risk and the influences of these conceptualisations on women's choices and midwives' practice.

Design and setting

a critical ethnography within a community-based continuity of midwifery care programme, including semi-structured interviews and the observation of sequential antenatal appointments.

Participants

eight midwives, an obstetrician and 17 women.

Findings

the midwives assumed a risk-negotiator role in order to mediate relationships between women and hospital-based maternity staff. The role of risk-negotiator relied profoundly on the trust engendered in their relationships with women. Trust within the mother–midwife relationship furthermore acted as a catalyst for complex processes of identity work which, in turn, allowed midwives to manipulate existing obstetric risk hierarchies and effectively re-order risk conceptualisations. In establishing and maintaining identities of ‘safe practitioner’ and ‘safe mother’, greater scope for the negotiation of normal within a context of obstetric risk was achieved.

Key conclusions and implications for practice

the effects of obstetric risk practices can be mitigated when trust within the mother–midwife relationship acts as a catalyst for identity work and supports the midwife's role as a risk-negotiator. The achievement of mutual identity-work through the midwives' role as risk-negotiator can contribute to improved outcomes for women receiving continuity of care. However, midwives needed to perform the role of risk-negotiator while simultaneously negotiating their professional credibility in a setting that construed their practice as risky.     

Temporal changes in key maternal and fetal factors affecting birth outcomes: A 32-year population-based study in an industrial city

Background

In Australia and internationally, there is concern about the growing proportion of women giving birth by caesarean section. There is evidence of increased risk of placenta accreta and percreta in subsequent pregnancies as well as decreased fertility; and significant resource implications. Randomised controlled trials (RCTs) of continuity of midwifery care have reported reduced caesareans and other interventions in labour, as well as increased maternal satisfaction, with no statistically significant differences in perinatal morbidity or mortality. RCTs conducted in the UK and in Australia have largely measured the effect of teams of care providers (commonly 6–12 midwives) with very few testing caseload (one-to-one) midwifery care. This study aims to determine whether caseload (one-to-one) midwifery care for women at low risk of medical complications decreases the proportion of women delivering by caesarean section compared with women receiving 'standard' care. This paper presents the trial protocol in detail.

Methods/design

A two-arm RCT design will be used. Women who are identified at low medical risk will be recruited from the antenatal booking clinics of a tertiary women's hospital in Melbourne, Australia. Baseline data will be collected, then women randomised to caseload midwifery or standard low risk care. Women allocated to the caseload intervention will receive antenatal, intrapartum and postpartum care from a designated primary midwife with one or two antenatal visits conducted by a 'back-up' midwife. The midwives will collaborate with obstetricians and other health professionals as necessary. If the woman has an extended labour, or if the primary midwife is unavailable, care will be provided by the back-up midwife. For women allocated to standard care, options include midwifery-led care with varying levels of continuity, junior obstetric care and community based general medical practitioner care. Data will be collected at recruitment (self administered survey) and at 2 and 6 months postpartum by postal survey. Medical/obstetric outcomes will be abstracted from the medical record. The sample size of 2008 was calculated to identify a decrease in caesarean birth from 19 to 14% and detect a range of other significant clinical differences. Comprehensive process and economic evaluations will be conducted.

Trial registration

Australian New Zealand Clinical Trials Registry ACTRN012607000073404.     

Midwifery in Israel

Socialized medicine provides free and easily accessible prenatal and intrapartum care in Israel. Mothers also receive the equivalent of dollars 100 to dollars 300 after delivering in hospital. The infant mortality rate is 5.5 per 1000 live births. Uncomplicated births (80%) are attended by midwives. Midwives routinely rupture membranes, attach electronic fetal monitoring, give intravenous fluids and deliver in lithotomy position. Approximately 15% of women undergo induction, 20% augmentation, 40% epidural, 30% episiotomy, 17% cesarean section. In 2003, 1% of births were cesareans by maternal request, rather than medical necessity, and 0.1% of births were planned homebirths. The system processes 140,000 births per year with good medical outcomes, especially considering Israel is only 56 years old.     

Improved obstetric management after implementation of a scaled-up quality improvement intervention: A nested before-after study in three public hospitals in Nepal

Background

We assessed the change in obstetric management after implementation of a quality improvement intervention, the Nepal Perinatal Quality Improvement Package (NePeriQIP).

Methods

The Nepal Perinatal Quality Improvement Package was a stepped-wedge cluster-randomized controlled trial conducted in 12 public hospitals in Nepal between April 2017 and October 2018. In this study, three hospitals allocated at different time points to the intervention were selected for a nested before–after analysis. We used bivariate and multivariate analyses to compare obstetric management in the control vs intervention group.

Results

There were 25 977 deliveries in the three hospitals during the study period: 10 207 (39%) in the control and 15 770 (61%) in the intervention group. After adjusting for maternal age, ethnicity, education, gestational age, stage of labor at admission, complications during labor, and birthweight, the intervention group had a higher proportion of fetal heart rate monitoring performed as per protocol (adjusted odds ratio [aOR] 1.19, 95% confidence interval [CI] 1.12–1.27), shorter time intervals between each fetal heart rate monitoring (aOR 2.09, 95% CI 1.96–2.23), a higher likelihood of abnormal fetal heart rate being detected (aOR 1.53, 95% CI 1.25–1.68), progress of labor more often being recorded immediately after per vaginal examination (aOR 2.73, 95% CI 2.55–2.93), and partograph filled as per standards (aOR 3.18, 95% CI 2.98–3.50). The cesarean birth rate was 2.5% in the control group and 8.2% in the intervention group (aOR 3.12, 95% CI 2.64–3.68).

Conclusions

The NePeriQIP intervention has potential to improve obstetric care, especially intrapartum fetal surveillance, in similar low-resource settings.     

Outcomes Associated with Planned Home and Planned Hospital Births in Low‐Risk Women Attended by Midwives in Ontario,Canada, 2003–2006: A Retrospective Cohort Study

Background: Midwives in Ontario, Canada, provide care in the home and hospital and are required to submit data for all births to the Ontario Ministry of Health database. The purpose of this study was to compare maternal and perinatal/neonatal mortality and morbidity and intrapartum intervention rates for women attended by Ontario midwives who planned a home birth compared with similar low‐risk women who planned a hospital birth between 2003 and 2006. Methods: The database provided outcomes for all women planning a home birth at the onset of labor (n = 6,692) and for a cohort, stratified by parity, of similar low‐risk women planning a hospital birth. Results: The rate of perinatal and neonatal mortality was very low (1/1,000) for both groups, and no difference was shown between groups in perinatal and neonatal mortality or serious morbidity (2.4% vs 2.8%; relative risk [RR], 95% confidence intervals [CI]: 0.84 [0.68–1.03]). No maternal deaths were reported. All measures of serious maternal morbidity were lower in the planned home birth group as were rates for all interventions including cesarean section (5.2% vs 8.1%; RR [95% CI]: 0.64 [0.56, 0.73]). Nulliparas were less likely to deliver at home, and had higher rates of ambulance transport from home to hospital than multiparas planning home birth and had rates of intervention and outcomes similar to, or lower than, nulliparas planning hospital births. Conclusions: Midwives who were integrated into the health care system with good access to emergency services, consultation, and transfer of care provided care resulting in favorable outcomes for women planning both home or hospital births.