Midwifery models of care are an increasingly important development in the future configuration of maternity services in the UK and internationally. Midwifery managed units are sometimes called Community Maternity Units or Birth Centres. Within this report they will be referred to as Community Maternity Units (CMUs). A Framework for Maternity Services in Scotland recognised the unique contribution that CMUs can make to maternity services in Scotland. Models of midwifery led care include midwife services that may be free-standing, operating with or without GP input, sited within an acute or community setting or alongside a large maternity unit. This section provides a consensus of the available evidence surrounding midwifery managed units.
Recent obstetric and paediatric guidelines suggest annual delivery rates of >3,000 as being crucial to ensuring maintenance of clinical skills and competencies of all staff and supporting an effective and efficient neonatal intensive care unit (RCOG/RCM 1999). Centralisation of neonatal facilities and the rationalisation of management and clinical services of individual trusts has driven many of the changes in Scotland and the UK (Walsh D, 2001). Whilst there is general consensus regarding the site of antenatal and postnatal care provision, there is debate about place of birth. Some argue for centralising all births at units with over 3,000 deliveries/year where neonatal and adult intensive care facilities are readily available, whilst others support the provision of local CMUs for women with low obstetric risk. Although the majority of the available evidence is in favour of CMUs, there is some concern that, due to the relatively small numbers involved, the evidence to support or refute the safety of these units is more difficult to establish (Graham 1997). Even if better outcome data were available, the studies of women at low risk of obstetric complications and their babies have too few adverse outcomes to allow significant differences between different forms of care to be detected.
There is a trend in England to turn some unsustainable consultant units into CMUs (Bournemouth Midwifery Unit, Edgware Birth Centre, Trowbridge Hospital, Grantham, Goole). While these have been evaluated, figures are small making generalisation difficult. Current levels of activity in 99/00 in Scottish CMUs range from 1 to 193 deliveries per annum.
Promoting normality
Promoting normality includes the use of available evidence to support care especially the use of skilled one-to-one midwifery care in labour. The recent study in North America to evaluate the effectiveness of continuous labour support by nurses demonstrated no difference in outcomes for women who received skilled one-to-one care or the 'normal' model of intrapartum care, however nurses in both arms of the trail followed hospital protocols which involved a high degree of intervention. The report concluded that reducing North American Caesarean Section rates and other interventions cannot be accomplished by only implementing a policy of continuous support. In order to achieve this aim all stakeholders must implement a comprehensive, multidisciplinary team approach to changing the environment in which care is provided (Hodnett et al. 2002)
Midwifery models of care
CMUs offer a cost-effective, safe and satisfying alternative for women who are experiencing normal pregnancy and childbirth (Rosser et al 2001), and offer an opportunity for the midwife to utilise her skills and fulfil her role. The Royal College of Midwives (RCM) has defined a midwife service as one that:
The international evidence highlights that outcomes for women who are cared for in Birth Centres (CMUs) are at least as good as for women in conventional units. The 1991 Birth Certificate data was examined for all singleton vaginal deliveries between 35 and 43 weeks gestation: after controlling for socio-demographic and medical risk factors, the outcomes for physicians and nurse-midwives were compared and demonstrated reduced CMU intervention rates and better neonatal outcomes (MacDorman, Marioan, et.al, 1998). The National Birth Centre Study (NBCS) in New York is the largest study to date and was a prospective observational survey of antenatal, intrapartum and postnatal care. It involved 84 free-standing birthing units (CMUs) and involved 12,000 women (Rooks et al, 1989, Rooks et al, 1992). A smaller prospective study of 2000 matched women of low obstetric risk, who gave birth in standard maternity hospitals, was undertaken: this study is limited due to the non-randomisation design and it included only 50% of the existing birth centres. Albers et al (1991) reviewed the outcomes of free-standing birth centres in the USA and differences in methodology, such as inclusion criteria, small samples and selection differences were identified. The paper concluded that non-hospital style birth settings have advantages to low risk women and that such settings were cheaper, there was less intervention in labour and outcomes were similar. Despite their large sample size, these studies, in keeping with other research of this nature, were not able to generalise or demonstrate significant trends.
The most recent publication of the systematic review of 'home-like versus conventional institutional settings for birth' (Hodnett, July 2002) involved 6 trials including 9000 women. Hodnett highlighted that substantial numbers of women were transferred to 'conventional' care antenatally, and this is consistent with other studies. Allocation to a home-like setting was associated with greater satisfaction with care, lower rates of intrapartum analgesia/anaesthesia, augmented labour, and operative delivery. There was a slight trend towards higher perinatal mortality in the home-like setting (odds ratio 1.49, 95% confidence interval 0.79 to 2.78), but this was not statistically significant. The review concluded that there are benefits from such models of care. It noted that increased support from care givers in labour may be particularly important in these settings.
United Kingdom evidence base
Although the majority of the available evidence supports CMUs (Hodnett 2002), many professionals are still cautious. There continues to be concerns about safety. These concerns are largely due to the deficiencies in the available information and the small numbers involved in many of the studies. Graham (1997) suggests that there has not yet been a trial of midwifery managed care with significant numbers to reliably demonstrate safety on the grounds of stillbirth or neonatal death (although this is also the case for other models of care). Some small studies highlighted the effectiveness and positive impact of small GP units in England (Young, 1987; Garrett et al, 1987; Lowe et al, 1987; Campbell 1997). Generally studies demonstrate reduced intervention, better satisfaction and similar outcomes to 'traditional' models of care. Qualitative research studies demonstrate some of the 'softer' outcomes of CMUs: women valued being in control, being able to have a safe birth, having good interpersonal relationships and being treated with dignity and respect. (Esposito, 1993). The advantages of CMUs identified from research are summarised in Table 4.1 ( Annex D).
Home birth
There is much debate about the safety and appropriateness of home birth. When Zander and Chamberlain (1999) stated that 'the assumption that hospital provides a safer environment for women at low risk as well as the high risk mothers is not evidence based' much debate ensued. In a published letter to the BMJ Drife (1999) argued, based on 3 international studies (Anderson & Murphy 1995, Bastian, Keirse, Lancaster 1998, Murphy & Fullerton 1998) but discounting available UK data, that hospital delivery is now three to four times safer than home delivery for the baby in cases of normal birth. This is countered by McFarlane (1999) and Bullock (1999). They argue that between the 4 datasets comprising the 3 international studies and the CEMD 1997, there was no consistency in the definitions of categories of death, or in the overall groups of births with which the deaths were compared. In their view this invalidates any attempt to use these data to make direct comparisons between the outcomes of births in these settings.
There is sufficient UK data to support the argument, that a planned home birth is a safe option for mother and baby (Northern Regional Perinatal Mortality Survey Coordinating Group, 1996)). Over the 14 years of the studies, the risk of death during delivery or in the first four weeks of life, in a baby of normal birth weight and without a lethal abnormality, was higher in those born to the 1% of women who had booked for a home delivery (1 in 538) compared with all other such births (1 in 810). However, during the last 10 years of that period, when the midwife was always the community lead professional, mortality in this subgroup was lower in those booking for home vs hospital delivery (1 in 1890 vs 1 in 904). The overall conclusion was that the perinatal hazard associated with planned home birth in the few women who exercised the option was low and, based on confidential enquiries, mostly unavoidable. Delays occasioned by the need to arrange and effect transfer probably contributed to only one neonatal death (Bulloch C 1999). Neither the National Birthday Trust Fund survey of 6044 planned home births in the United Kingdom (Chamberlain, Wraight, Crowley 1997) nor the prospective and retrospective studies in the former Northern Region of England (Davies, Hay, Reid &Young 1996, Northern Region Perinatal Mortality Survey Coordinating Group 1996) yielded results that would alter the key conclusion of Where to be Born?, which was that "there is no evidence to support the claim that the safest policy is for all women to give birth in hospital" (Campbell & Macfarlane 1994).
A Dutch study to investigate the relationship between the intended place of birth ( home or hospital) and perinatal outcome in women with low risk pregnancies (after controlling for parity and social, medical, and obstetric background) analysed prospective data from 97 midwives and 1836 women. The results concluded that there was no relation between the planned place of birth and perinatal outcome in primaparous women, when controlling for a favourable or less favourable background. In multiparous women, perinatal outcome was significantly better for planned home births than for planned hospital births, with or without control for background variables (Wiegers, Keirse van der Zee, Berghs, 1996).
A meta-analysis of 6 observational studies, including 24,092 primarily low-risk pregnant women, examined the safety of planned home birth compared with planned hospital birth (Olsen & Ole. 1997). The principal difference in outcomes were a reduced frequency of low Apgar scores and fewer medical interventions in the home birth group. The study concluded that "home birth is an acceptable alternative to hospital confinement for selected pregnant women, and leads to reduced medical interventions." A Cochrane review of available evidence concluded that there is no strong evidence to favour either planned hospital birth or planned home birth for low risk pregnant women (Olsen O, Jewell MD, 2002 last substantial amendment 1999).
Assessing risk and risk management strategies for CMUs
It is difficult to assess perinatal risk (Enkin, 1994; Institute of Medicine and National Research Council, 1992) and adopting a social model of childbirth will influence how risk is assessed. Nonetheless careful consideration must be given to using informed evidence and local population profiles to develop risk assessment and management strategies. In the Wormerveer study (which excluded women booked to consultant care in Holland), the perinatal mortality rate was highest in those transferred in labour; the authors interpreted this finding to support the notion of good selection rather than an unacceptable risk to the transferred group (Van Alten, Eskes, Treffers, 1989).
The most recent review of 'low tech' versus conventional care in labour highlighted that caregivers and clients in home-like settings require to identify signs of maternal and fetal complications (Hodnett, 2002). If any model of midwifery managed care is to be introduced, then risk factors associated with location must be reviewed. The 5 th CEMD Report UK (2001) stresses the importance of assessing, identifying and managing risk appropriately.
Inclusion and exclusion criteria for CMUs
Establishing the clinical criteria to determine appropriate booking and transfer guidelines is the subject of much debate. There is international evidence that, the longer centres are established, the more liberal criteria become (Waldenstrom 1998). In the UK this is reflected in areas such as Edgeware and Bournemouth, which have recently amended and expanded their entry criteria to its CMU adjacent to a consultant led obstetric unit. It is important that entry and exit criteria are evidence based, agreed locally and adhered to.
Transfer evidence for CMUs
In the Leicester and Aberdeen trials (Hundley et al, 1994 ) only 46% of the women, who were randomised to deliver in the midwife managed unit, actually delivered there compared with 62% of women who booked and delivered in Bournemouth. Campbell et al (1999) argue that the adjacent proximity of the obstetric unit in Aberdeen and Leicester could affect decisions to transfer women and increases the likelihood of transfer. A common reason for transfer is for analgesia during labour, especially epidural analgesia. Evidence consistently demonstrates that the outcomes of CMUs are at least equal to consultant obstetric units.
Table 4.2: Recent data from two free standing midwifery delivery units (Scottish Level 1b equivalent)
| No. of bookings | No. of births | No. of transfers |
antenatal | intrapartum | postnatal |
mother | baby |
Aberdare (Sept 97-Aug 98) | 249 | 209 | 40 (16%)* | 20 (10%)† | 3 (1)† | 2 (1)† |
Crowborough (Apr 97-Mar 98) | 331 | 179 | 109 (33%)* | 17 (9)† | 1 | 0 |
Source: Zander & Chamberlain 1999 * percentage of bookings
† percentage of births
An audit of women who delivered in a rural GP unit in Scotland (Stranraer) highlighted that the intrapartum transfer rate in women booked to deliver at the general practitioner unit was 12.8% (68/530); this included 18% of primiparous women (37) and 9% of multiparous women (31). The commonest reason for transfer was delay in the first stage of labour 21/68 (31%) transfers. Of the unplanned transfers in labour, 17 women had caesarean sections (25% of all transfers) and one a forceps delivery; the remaining 50 had normal deliveries. A further 68 women who were booked for consultant delivery elsewhere presented to this unit and required transfer by ambulance with midwife escort (120 km). Most of these women were considered to be in labour and represent the highest risk cases; they cause anxiety and have resource implications for the ambulance, midwife, and general practitioners (Baird et al, 1996). A study of a midwife managed unit in Norway (Holt et al, 2001) showed that of the 628 low risk women in the study, 152 (24.2%) were transferred antenatally, a further 41 (6.5%) were transferred in the intrapartum period. The study concluded that this model could be used in rural and remote areas as an alternative to centralisation of births.
Quality of care in CMUs
Crucial to quality in midwifery managed care is a culture and value shift by many midwives and doctors. If birth is not treated as normal, women may not feel empowered. A recent study of home birth in Scotland highlighted that some midwives adopted the same medical model at home as they did in hospital (Edwards 2000). Professional support and development, peer review and sound midwifery supervision is central to good care. Walsh ( 2001) argues that all midwives should have: