Factors influencing utilisation of 'free-standing' and 'alongside' midwifery units for low-risk births in England: a mixed-methods study

Method

Telephone survey of all NHS trusts in England with maternity services.

Data collection

Our data collection was aided by information provided by BirthChoiceUK and the consumer organisation Which? Both of these companies provide web-based information about maternity service provision across the UK. BirthChoiceUK holds a database containing details of maternity unit configurations, which was supplied to Which? for the development of the Which? Birth Choice website. 56 Which? also audits MU provision and utilisation across the UK. We entered into a data agreement with Which? for it to share the details of maternity units and configurations along with information it had collected about birth numbers in MUs in England. We developed our own data collection pro forma after consulting both the Birthplace in England mapping data collection tool57 and pages on the Which? Birth Choice website relating to maternity units, and populated it from the Which? Birth Choice data. Heads of midwifery (HoMs) in the 134 trusts across England were then sent the survey. We then telephoned the HoMs, who confirmed, clarified or completed missing entries in the survey for their current maternity services. These telephone interviews, which lasted up to 30 minutes, took place over a 3-month period between March and May of 2016. Actual annual number of births was completed using the Which? Birth Choice data and updated in the telephone interviews, as required.

Ethics

This first stage of the research was classed as service evaluation and thus did not require Research Ethics Committee (REC) approval.

Sample

One hundred and thirty-four HoMs, representing all NHS trusts providing all publicly funded maternity care in England, were contacted by the research staff. Home birth was excluded from our data.

Analysis

Descriptive summary statistics and narrative description of configuration, organisation and operation of AMUs and FMUs was undertaken.

Method

A systematic search of media reports, content and discourse analysis was carried out guided by framing theory. 58

Data collection

The LexisNexis database was searched systematically to identify all media reports directly relating to the 14 FMUs in England that had closed between 2008 and 2015. Multiple FMUs from the same NHS trust were considered as one site, as were two units from different but nearby trusts that had closed after recommendations from the same service review. After screening, two services (Mid Yorkshire Hospitals NHS Trust, Wakefield Birth Centre; and Wiltshire Primary Care Trust, St Peter’s Maternity Unit, Shepton Mallet) were found to have no relevant articles, leaving a remaining eight services. The number of articles about each site ranged from 2 to 65. A total of 190 articles were included in the analysis. These included local newspaper articles (n = 175), transcripts of local television news items (n = 5) (all from one site) and articles from national publications (industry publications and national newspapers) (n = 11).

Analysis

Data were extracted onto a bespoke template in Microsoft Excel^® (Microsoft Corporation, Redmond, WA, USA), which provided a framework for analysis, working on two levels, to (1) identify and analyse the relevant content using a structured approach and (2) identify and analyse the discourse present in the media reporting in each case. The approach enabled us to capture the range of voices and to compare perspectives, identifying who the authors and speakers were in each report and relating this to extracted content and discourse. The aim of this was not to triangulate in terms of validation but to recognise that media analysis captures events not as ‘facts’ but as representations of them. The template was drafted in relation to the study objectives and discussed and revised within the research team. It was intended to be used flexibly, allowing for the amendment of categories when needed, and some categories on the template were combined in the light of the initial data extraction work. Two team members extracted data for each site independently of each other, using a separate sheet for each media report, and any areas of difference were discussed to reach consensus. Using a purpose-built macro, the data were then collated into a single spreadsheet for each site and a thematic approach was used to synthesise the findings within and then across all services.

Once the emergent themes had been identified, framing theory58,59 was used to guide the synthesis of themes in relation to discourse.

Method and justification

Case study methodology was the most appropriate approach for these aims, as it facilitates the exploration of large, complex organisations, such as maternity services. It does this by combining a range of data collection methods, including interviews and focus groups, with a variety of sampling techniques, to achieve data saturation and thus gain an in-depth understanding of the social processes and organisational culture within each study site. 60 Specifically, this project explored why organisational change, premised on a relatively strong evidence base, had not been achieved across England in relation to the provision and utilisation of MUs. Within each site, both individual interviews and focus groups were used to produce a detailed and rounded analysis of the case. A comparative case study approach enabled the identification of common and differentiating determinants of the successful implementation of MUs. 61 In-depth exploration of each study site was necessary to gain an understanding of the complex set of inter-related factors that played a role in the local uptake and implementation of MUs. The case studies provided sufficiently detailed descriptions of the local social processes and contextual factors to develop an understanding of the potential for individual findings to be sensitively generalised to other sites. 62

Using the data from stage 1, six case study sites were purposively sampled to identify:

• two high-performing services with both an AMU and a FMU, achieving ≥ 20% of total births

• two low-performing services with one or more MUs (AMU or FMU), achieving ≤ 10% of total births

• two services with no MUs.

Within each of the three categories, study sites were chosen to represent both rural and urban settings to enhance diversity. Six sites (two from each category above) were chosen to obtain a cross-section of local maternity care services that might approximate typical adoption, integration and utilisation of MU services.

Theoretical/conceptual framework

The theoretical and conceptual frameworks underpinning the case study methodology are described below.

This research sought to identify and analyse factors that help to explain variations in MU provision, with a view to identifying routes to overcoming barriers to change. Findings of the potential economic efficiency of MUs over OUs46 suggest that financial drivers alone do not explain the inequitable provision of MUs across the country. In recent years, there has been a widespread focus on the gap between research evidence and practice. 63,64 Prior research supported by the National Institute for Health Research describes multifaceted barriers to and drivers of change in the NHS, including contextual conditions, the nature of innovations and the processes of implementation. 65,66 Considering this work, a wide range of factors may be thought to facilitate or create barriers to the establishment and utilisation of MUs. Previous research has identified a number of important institutional factors shaping maternity services and potentially hindering innovation, including issues of professional power and the midwifery–obstetrics relationship,67,68 resource constraints and workforce issues,69 the environment of litigation and risk management,70,71 as well as inaccurate media portrayals of birth. 72 Currently, however, the prevalence and impact of potential barriers to the creation and utilisation of MUs remains unclear. In addition, a large body of research on innovation diffusion and implementation has demonstrated that innovations, particularly those that involve complex organisational change, do not move directly from evidence into practice, but are adapted, and in some cases transformed, through the process of implementation. 67,73 To explicate this process, our study utilised Damschroder et al. ’s74 Consolidated Framework for Implementation Research (CFIR), supplemented by theory that considers the wider institutional forces shaping health-care organisations and organisational change. 75,76

The CFIR provides useful insights for investigating why implementations fail, or are only partially successful, to guide future implementation efforts. 77 This was selected to provide a conceptual framework to analyse the contexts of MU implementation and systematically identify the factors influencing the extent to which MUs are adopted into practice and utilised. The CFIR was developed through meta-review of previous implementation literature and theories, and identifies 39 underlying constructs previously found to have an impact on the course and extent of implementation. The CFIR has now been used by a number of studies to identify factors affecting implementation, including the implementation of complex service initiatives,78–80 and has been found to be an effective tool to identify barriers to and facilitators of implementation. 81 Within the CFIR, constructs are divided between five major domains:

1. characteristics of the intervention itself, including the essential ‘core’ and adaptable ‘periphery’ elements, its design, complexity, cost and degree of flexibility

2. outer context of change, including pressure from patients, partners and other stakeholders

3. inner context of change, including the features of the local organisation, its culture, structure, practices and resources

4. individuals involved in change, their preferences, beliefs and identities

5. process of change itself, including details of planning, engaging and executing.

This framework is not intended to be applied wholesale, but rather offers a long list of constructs to consider for inclusion when investigating particular instances of implementation. We drew on CFIR to identify relevant constructs in early stages of the study to inform data collection and subsequent approaches to analysis, as outlined in the sections below. The current research sought to explore both the variation in initial engagement with MUs as well as subsequent implementation in sites that do engage. Therefore, the CFIR, geared towards understanding specific instances of implementation, was supplemented by theory geared towards consideration of the factors affecting MU adoption at a more general level. A large volume of recent work on health-care organisations has identified regulative, normative and cultural features of the health-care institutional environment75 that shape the potential for innovation and change. 76 Waring et al. 82 suggested the guiding categories of power, organisation, culture and knowledge, which have previously been adopted to make sense of the complex set of inter-related social factors known to affect processes and outcomes of health-care organisations. These categories guided the analysis of the more macro elements of the ‘outer context’ of change.

Data collection

Data collection consisted of in-depth telephone interviews and focus groups of midwives and service users, held locally (Figure 2).

FIGURE 2.

Case studies.

The interview schedule was developed by the research team, drawing on the findings from the first stage of the study, existing literature regarding the organisation of maternity services and literature on implementation and organisational change, and was reviewed by the service user reference group and the advisory group. Questions were focused on discussing the evolution of services in each of the sites.

Analysis plan and revisions

The process for analysis of qualitative case study data was as follows.

Both DW and CG attempted to each code a women’s focus group discussion according to the CFIR, but it became clear from early on in this exercise that the transcript codes did not map across to the CFIR domains and constructs, except in very selected areas. This was because the women’s transcripts reflected their personal experience of care and not the broad sweep of a major service implementation project. Following a team discussion, it was decided that the women’s focus group discussion would be open-coded and analysed thematically, utilising the principles of thematic analysis as outlined by Braun and Clarke,83 supported by NVivo software (QSR International, Warrington, UK). This approach follows six distinct stages: (1) familiarisation with data, (2) generating initial codes, (3) searching for them, (4) reviewing themes, (5) defining and naming themes and (6) writing up findings.

All other interviews and the midwives focus groups (MFGs) were analysed using the CFIR. First, in the pilot analysis, five members of the research team (DW, DC, HS, CG and SB) coded the same three interviews to the constructs in the CFIR, which are grouped into six domains, mentioned above. This coding was placed within NVivo, with nodes and subnodes created to match the domains and constructs of the CFIR. These team members then met to compare coding and discuss how data related to each of the constructs. A second pilot round was then undertaken with the same team members coding an additional two transcripts. Once it had been agreed that a shared understanding of CFIR constructs had been achieved, case study sites were then divided between the project members involved in coding (DC two sites, DW two sites, SB one site and HS one site). The remaining interviews and MFGs were then coded according to the constructs within the CFIR. It was considered appropriate for an individual coder to complete analysis of a whole site in order to develop an in-depth understanding of the case. During this stage, ‘open’ codes were also added within CFIR constructs to further identify how the data related to the broader construct. For example, data relating to the construct ‘patient needs and resources’ were also coded by the nature of specific patient needs and resources referred to, such as ‘patient need for choice’ or ‘patient need for convenient locale’.

At this stage, a number of issues with coding data to the constructs in the CFIR were encountered. One issue that arose was that there were differences in the number of open codes created by each coder, with some coders adding a larger number of codes with greater specificity in relation to the data. In addition, data from the MFGs and the individual interviews often related to multiple constructs, and were therefore coded in several places, under the framework. One of the consequences of this was the production of many additional nodes within NVivo, which made handling the full data set cumbersome; it also meant that it was difficult to test inter-rater reliability following this coding stage.

To address this, following this coding, a wider team of co-investigators (DW, HS, SB, CMc and LC) who had previously undertaken case study analysis met to start the process of collapsing codes into themes. In addition, the research fellow who had not participated in the coding, but was very familiar with the data scrutinised, reorganised and reduced the codes within each construct to reflect the important themes of the data and relate this to the logic of the CFIR construct in which it was located, in a process akin to axial coding.

This process of reduction revealed six cross-cutting themes that were relevant under multiple CFIR constructs (see Table 22). We then had to make a decision on how to report the findings: whether to give priority to the cross-cutting themes or to the CFIR domains as the organising headings. If we chose the former, it rendered the theoretical underpinning of the CFIR redundant, superseded by these six primary themes. However, the CFIR, as well as being empirically validated in multiple organisational research projects,74 had provided us with a heuristic template via its five headline domains of (1) intervention characteristics, (2) outer setting, (3) inner setting, (4) individual characteristics and (5) process. We therefore decided to locate the cross-cutting themes within these domains.

In addition, narrative summaries and within-site comparisons of each case study site were produced to tell the ‘story’ from each site and to tease out areas of unanimity and dissonance between the respondents. These sections utilised and synthesised elements of the broad domains of the CFIR (inner setting, outer setting, characteristics of individuals, intervention characteristics and process) to shape each story. This facilitated more congruence in the reporting of the data collected in each site, which lost some of its distinctiveness and rootedness when coded as one data set into NVivo.

A ‘within-case’ analysis proceeded to ‘cross-case’ comparisons. Waring’s84 categories of power, organisation, culture and knowledge assisted in providing a macro-lens to inform this level of analysis. In this way, a comprehensive picture of both the ‘inner’ and ‘outer’ context for change assisted in distilling the key barriers to and facilitators of MUs developed (see Chapter 6).

The CFIR assisted analysis of interviews and MFGs were juxtaposed with the thematically analysed women’s focus group at the level of the cross-case analysis.

Data collection

Data were captured by field notes and on flip charts throughout the day, which were then transcribed. Through this process of debate and discussion, further service guidance was co-developed by stakeholders and the research team.

Sample

The workshop was attended by 56 people. Our objective was to include all groups representing those involved in the maternity services, including service users, maternity care professionals, NHS trust managers, maternity policy representatives, service providers, service commissioners and politicians, and academics with an interest in maternity care.

Stakeholder workshop attendees were:

• four members of the service user reference panel

• head of policy of the National Childbirth Trust

• representative of the chief executive of the Royal College of Midwives (RCM)

• midwifery representative from the Department of Health and Social Care

• three co-investigators on the Birthplace in England study3

• national clinical director for Maternity and Women’s Health in NHS England

• director of nursing, Commissioning and Health Improvement

• chairperson of NICE Intrapartum Guideline Group

• knowledge mobilisation research fellow (maternity services)

• two representatives of former local supervising authority midwifery officers

• two consultant midwives

• neonatologist

• consultant obstetrician

• three representatives of Maternity Transformation, NHS England

• chief executive officer (CEO) of Birthrights

• two doulas

• midwifery clinical leads from an AMU and a FMU

• two representatives of the Midwifery Unit Network

• two representatives from two maternity and neonatal networks in England.

Ethics

This third stage of the research was classed as service evaluation and thus did not require REC approval.

Analysis

The workshop data were synthesised by HS and reviewed by DW and LC.

Staffing

We recorded whether or not AMUs had core staff (i.e. designated midwives assigned to permanently staff the AMU, rather than midwives allocated from the OU for each shift or some other variant). We then compared trusts achieving > 20% of all trust births in their AMUs with those trusts achieving 10% to 20% and those under 10% (Table 3).

Eighty-six AMUs had core staff and there was no difference between higher-performing and lower-performing trusts.

The vast majority of AMUs have a designated midwife clinical lead, but lower-performing AMUs are less likely to (Table 4).

Community midwives staff approximately 60% of AMUs on a shift system (Table 5).

Staff movement to obstetric-led unit

Nearly all (97%) AMU midwives are moved into OUs to work for part or all shifts regularly. This means that they may often be caring for low-risk women on OUs who are suitable for MUs (Table 6).

Closures

Related to the movement of staff out of AMUs was the finding that one-third of AMUs experience closures, with one-fifth closing frequently (more than once per month) (Tables 7 and 8). The closure lasts for the duration of a shift in most cases. Lower-performing AMUs close three times more often than higher-performing AMUs. Movement of staff out of AMUs to OUs and closure of AMUs were expedient and a response to shortage of midwives on OUs and/or workload on OUs. Movement was almost always asymmetrical, in the direction of OU.

Capacity

Higher-performing AMUs had more birth rooms (five to eight), whereas lower-performing AMUs had fewer birth rooms (two to four). The latter had fewer births per room (< 200/year), although even the larger AMUs underutilised their room capacity. Thus, all AMUs could absorb more births per year.

Access and physical location

Data were collected on whether trusts had an opt-out or opt-in model for women accessing their AMU (Table 9). Opt-out implies that low-risk women wanting a hospital birth were sent to the AMU unless they specifically requested to labour in the OU (i.e. a standard midwife-led or low-risk pathway). Opt-in means that women are sent to the OU unless they ask for the AMU. In addition, a small number of services operate a hybrid model that, for example, leaves the decision to the admitting midwife.

Fifty-seven per cent of AMUs had an opt-out policy and this appeared to be associated with higher utilisation. Of the higher-performing AMUs, 73% had an opt-out policy, whereas only 14% of the lower-performing AMUs had an opt-out policy.

Eighty-one per cent of trusts used NICE recommendations91 to ascertain women’s eligibility for AMU care, though > 50% of trusts adapted or broadened these recommendations to allow other women access to their AMU (Table 10).

We also asked if women who did not meet the local eligibility criteria were sometimes allowed access to AMU (Table 11).

High-performing trusts were more likely to allow this eligibility.

Alongside midwifery units were co-located in three ways: (1) 63% were on the same floor as the main labour ward, (2) 36% were on a different floor and (3) 1% were in a different building (Table 12).

There may be an association between smaller MU birth numbers and the location of the AMU on the same floor, but the numbers were too small to be conclusive.

Facilities

Staffing

Approximately half of the trusts with FMUs had core midwifery staff in their FMUs (Table 15). The differences between the three groups (< 10%, 10–20% and > 20%) must be interpreted with caution, as the FMUs did not figure strongly in overall birth proportions, particularly in trusts with both AMUs and FMUs.

For FMUs with no core staff, community midwives must staff them when they accompany women for birth.

Just over 50% of FMUs were staffed 24 hours per day, whereas the remainder were either open during daytime hours only (e.g. for clinics) or open when women booked there went into labour. Nearly all FMUs had a midwife clinical lead. Community midwives played a major role in staffing, mostly as the second midwife for births. Twelve trusts ran their FMU by a fully integrated model of core staff and community midwives.

Staff movement

Over half of FMU staff were moved within or between shifts, most often to the host OU (Table 16).

Closures

Nearly 40% of FMUs were closed more than once every 6 months (Table 17).

Access

Unlike AMUs, only 5% of FMUs had an opt-out policy, whereas nearly 50% operated an opt-in policy (Table 18).

Just over one-third of trusts had stricter eligibility criteria for their FMU than for their AMUs (Table 19).

Facilities (birthing pools)

Almost all FMUs (98%) had rooms with birthing pool facilities and most had more than one (Table 20).

Voices supporting closures

Our content analysis revealed that very little of the media reporting focused on safety per se and made very little reference at all to clinical evidence. However, the assumption that being closer to obstetric care was safer, despite this being contrary to the evidence, was implicit in the arguments in support of the closure of FMUs in favour of AMUs:

We feel that a midwife-led unit close to the maternity department would provide the best of both worlds – the possibility of a natural birth without medical intervention, but the security of knowing an obstetrician is close by in case of emergencies.

Trust chief executive, Scarborough Evening News97

Free-standing midwifery units were often portrayed as undesirable or inferior places to give birth:

The vast majority [of women in Grantham] opt to give birth at other hospitals where they can get:

None of which they can currently get at the Midwifery Led Birthing Unit at Grantham Hospital.

Journalist, Grantham Journal. 98

Reproduced with permission from Marie Bond, Editor, Grantham Journal

Trust CEOs, commissioners and some journalists asserted that the problem was that there was a lack of demand for FMUs from women. Women were assumed to have a free and informed choice about where they plan their births and they were not choosing FMU care:

The midwives do a great job but women are clearly choosing to go elsewhere for their care.

Clinical Commissioning Group accountable officer, Grantham Journal. 98

Reproduced with permission from Marie Bond, Editor, Grantham Journal

Chief executive officers and commissioners also used the articles to explain how much money their FMU was costing their service, quoting the specific amount of money that they perceived was ‘lost’ each year. In some cases, the financial losses were attributed directly to women’s decisions on place of birth:

Only 274 out of 5440 new mothers from Brent gave birth there in 2006, along with 18 from Harrow, leading to a deficit of £300,000 a year.

First-time mother, Kentish Gazette99

In Brent, where the FMU had been opened on the site of a closed OU, the CEO presented his decision to close the unit as a moral one:

I have a responsibility to make the best use of taxpayers’ money and the Brent Birthing Centre is losing £300,000 a year. As an accountable officer, that is not something I can sustain.

CEO, Harrow Times100

Articles deployed a combination of managerialist and consumerist discourses whereby managers were depicted as sharing out limited resources for the greater good, whereas women exercised a ‘free, consumer choice’ not to birth in FMUs:

It’s important we provide the best care possible for every pound spent, for everyone in Derbyshire.

Medical director, Derby Evening Telegraph101

It’s about how you provide the best model of maternity service for the area. We can’t continue to run the whole service in this way and something has to change.

Hospital trust CEO, Hull Daily Mail102

Often the proposed changes involved the closure of the FMU and the opening of an AMU in another of the trust’s hospitals. This was at times framed as a ‘relocation’ of the FMU, rather than a closure, and in the case of the Brent Birth Centre the new AMU was given the same name as the closed FMU.

Closures were frequently framed as ‘gains’, for example the opening of a new MU or more money to invest in other services:

What do the proposals mean for women living in North Tyneside? When a woman becomes pregnant, no matter whether she is considered to be low or high risk she will be able to choose to deliver her baby at either Northumbria Specialist Emergency Care Hospital or Royal Victoria Infirmary, Newcastle.

Commissioner, Chronicle Live103

I hope that most of the journal’s readers will be reassured to know that the changes that have been announced this week will bring about an expansion in the services on offer at Grantham Hospital and an improvement in the treatment and care given to local people.

Nick Boles, MP, Grantham Journal104

Voices opposing closure

Service users and campaigners privileged women’s right to birth locally, expecting that trusts had a duty to provide and support women’s opportunities to choose these places of birth.

In Kent, campaign messages emphasised the importance of women being able to birth locally to avoid long journeys in labour and to retain birth within the local community. Changes were questioned by service users using highly emotive language that drew on images of family, home and community:

The move would force Canterbury mothers-to-be into long journeys to hospital and mean Kent’s only city would disappear from birth certificates [our emphasis].

Kentish Gazette 105

Both of my children were born there as were their mother and me. I was therefore hoping to see a third generation eventually come into the world at Buckland.

Philip Moore, East Kent Mercury106

Service users and campaigners argued that the FMUs remained desirable places to give birth, but that many women were prevented from using FMUs through eligibility guidelines or a lack of information:

I think one of the fundamental issues is the lack of publicity the centre gets from GPs and midwives at booking in appointments. Two friends of the family who are expecting didn’t know the Jubilee Birth Centre existed until I told them!

Half the pregnant ladies were not given the chance to use it. They were told, you have to go to Lincoln.

Campaigner, Yorkshire and Lincolnshire Regional News and Weather107

For those opposing the closures, the arguments were more likely to be framed in terms of a ‘loss’. The loss of the unit or of local services and possible implications for the economic well-being of a local community, and other local services:

So North Tyneside General . . . will soon have no A&E [accident and emergency] and likely no maternity at all after already being degraded. What is to become of it? A cottage hospital where people go to die in old age? The demise of hospital services for residents . . . is disgraceful.

Claire Louise Keys, service user, Evening Chronicle108

However, campaigners supporting the Jubilee Birth Centre in Hull were reported as satisfied with the trust’s proposal to ‘relocate’ the FMU to the main hospital to become an AMU:

Jubilee Supporters' Group campaigner Sian Alexander, said: ‘While I am sad to see the Jubilee close permanently, I'm glad the trust at least acknowledges the importance of midwifery-led care and I hope the spirit of the Jubilee will be able to live on, albeit in another place’.

Hull Daily Mail 109

Underutilisation

Seven were said to be ‘undercapacity’, although no justification exists in the literature of a threshold for utilisation that equates to viability. Some FMUs open only for intrapartum care, remaining closed the rest of the time, but the argument is made that utilities, such as heating, lighting and water, are required continuously. However, there still exists a significant number (53) of FMUs with 50–250 births per year, mostly in rural areas. Some units provide antenatal and postnatal care and other community-based services for a rural area.

Distance from host obstetric-led unit

This rationale for closure applies exclusively to rural units, where it was judged that journeys of > 15 miles, or time taken to transfer of > 30 minutes, may impact on safety. It was interesting that none of the closures had been precipitated by clinical incidents, at least according to the HoMs interviews, and this was also reflected in our analysis of media content and discourses, which focused on economic and staffing issues, with little reference to clinical safety concerns.

Popularity of alongside midwifery units and rationalisation of obstetric-led units

A number of the interviewees mentioned that AMUs were opened in virtually all trusts where FMUs were closed. They were seen as a more popular alternative to FMUs, but this appears to be a professional view, rather than a service user view. Certainly, their capacity is greater, sometimes by as much as 10 times. However, this greater volume of births is primarily derived from women who would have previously accessed the main OU. Interestingly, at two of the closed FMU sites, home birth rates went up, suggesting that some local women chose to birth at home over attending the neighbouring OU/AMU.

There were some interesting examples of closures that did not fit the above categories. One FMU closed despite having > 300 births per year, although the rationale given was underutilisation. A metropolitan FMU closed because its local population of mainly immigrant women perceived it as a second-class service, because there were no doctors on staff. The FMU was built in an industrial estate that meant low visibility in the local community. More than one comment was made that small FMUs represented a ‘Rolls Royce’ service that few women could benefit from.

The next chapter describes in detail the findings from the six case study sites.

Leadership

Interviews and focus group data revealed a commonly held view that the reconfiguration developed from extremely strong midwifery leadership and widespread, multilayered and multidisciplinary consultation and engagement with both internal and external stakeholders:

. . . it’s crucial to have an inspirational leader. If you don’t have somebody at the very top who is passionate about it happening, it won’t happen. And they must cascade, get everybody onboard.

AMFG2

One of the important characteristics of a leader that midwives and service user representatives identified was charisma:

. . . a charismatic leader to kind of bring it together . . . unless you’ve got that then I think it’s quite hard to bring it to fruition.

AInt9

The longstanding, active and ongoing support of the trust’s MUs by key senior obstetricians was also viewed as very important. A senior obstetrician who was interviewed was a very strong advocate of MUs and normal birth. She described an enduring commitment to collaboration with midwives and to challenging the medicalised practice of service:

. . . ultimately I suppose it’s with them [women] having that better experience and it’s reducing C-section [caesarean] rates and the medicalisation of birth and bringing it back to something a bit more of the emotional, social, psychological side of birth.

AInt5

The importance of continuity of vision when new service leaders came into post was referred to by the senior obstetrician, the midwives and the service users. They suggested that recent changes in midwifery leadership had resulted in a reduction in staffing on the MUs and less service user engagement. This was perceived to contribute to falling numbers of births in both the large FMU and the AMU:

I think the birth centres are being used less at the moment, and that does seem to coincide with a change of leadership.

AInt9

Culture and beliefs about midwifery-led units

The midwives from each of the MUs identified strongly with their work context and organisation. They saw themselves as normality advocates, grounded in a social/midwifery model of childbirth:

I think that the birth centre philosophy is a social model of care . . . This is where we wanted to work with the community midwives who wanted the normality as well.

AMFG1

The midwives talked of the importance of a strong team culture in the MUs, with the need to trust your colleagues, especially in an emergency. This was facilitated by frequent, random unannounced skills drills, which the senior midwives set up.

In both the FMU and the AMU, there was a potent sense of ownership and a commitment to promoting their units and a strong commitment to the integrated MU community model:

. . . the midwives very much take ownership of the [AMU], and you don’t see a lot of higher management on the unit here, where you might do in some of the other areas.

AMFG1

The FMU midwives engaged in considerable promotional activity within the local community (e.g. providing information stands at local shops) and at the unit itself (e.g. Christmas parties for the local community to join):

So you have to do a lot of positive promotion, you have to get out there. And you’re almost selling a product. And that’s how we saw it. So we did lots of promotional events, and got lots in the press, about the positive things that we were doing . . .

AMFG1

The AMU staff had also held fundraising events to purchase equipment/’extras’, which had helped build the staff culture as well as increase the unit’s visibility:

. . . it’s like the fundraising [name] just mentioned, we did use to do that quite often, to promote, to raise money for us to use on the birth centre [AMU]. To decorate our rooms.

AMFG2

Change

The reconfiguration was not without resistance from staff, with two main contributors to this: first, a loss of workplace identity and culture for the staff (obstetric, neonatal and midwifery) from the OU which closed; second, the adjustment to a new model of practice for midwives:

[Staff] will tell you themselves that it wasn’t without challenge, etc. . . . the way they had to totally change the culture and the way they worked, even down to the language that was used rather than . . . they don’t use language around ‘delivery’ now, it’s about ‘birth’, and yeah, I think it was met with some resistance from staff initially . . .

AInt1

The development of the MUs met more resistance from the neonatologists, who were not supportive of a FMU initially:

As a team of neonatologists we clearly always see the extreme end of the spectrum in terms of, you know, babies who get into difficulties because those are the ones who we come in contact with, so obviously there were anxieties.

AInt9

The neonatologists’ engagement with the process was through the development of protocols. They strongly argued for the presence of a neonatal life support (NLS)-trained midwife on every shift at the FMU and this was implemented.

Organisation and staffing

From the outset, community midwives and a small team of core midwives staffed the MUs. This integrated staffing model was described very positively by those working in it, with there being a strong recommendation that the integrated model is desirable for an effectively functioning MU:

If you have an integrated model you develop a really highly skilled, flexible workforce, which can move to where the activity is.

AMFG2

The midwives valued the relational continuity of care, however limited, that the integrated model provided. They experienced this if they cared for women in labour whom they had previously met during the antenatal period.

This site demonstrated that strategic planning, visionary leadership and widespread sign-up from stakeholders set the conditions for successful MU development. In addition, developing a sense of identity and ownership in the MU staff facilitated MU promotion and optimum utilisation.

Leadership

In contrast to Armidale, there was an apparent lack of maternity service leadership during the reconfigurations at Borodale. A maternity service manager at the time of data collection commented that she was distracted from promoting the FMU as she was addressing other priorities in the trust:

And sometimes you have to go all through that before you get the thinking time to think actually, are we doing it the right way? Do you know what I mean? But when you’re in the middle of it, this is what you’ve got to do . . .

You’re firefighting almost?

Yeah.

Other staff interviews expressed uncertainty about whether or not the problems with leadership existed at trust or maternity unit level:

We are constrained by the trust and so you know, if people did say, hang on this isn’t working, we’ve got to get everything on one site, the trust won’t countenance that. So it’s difficult for me to know whether it’s maternity leadership or whether it’s them [the trust].

BMFG1

Culture and beliefs about the intervention

Across the trust and midwifery management of Borodale, there were a number of conflicting views regarding the value of MUs. Illustrative of those with negative views was one trust manager who suggested that the midwives:

. . . believe that you have to have padlocks, bolted door with machine guns and guard dogs between a unit that’s midwife run and a unit that has obstetricians.

BInt5

However, other trust managers were supportive of AMUs:

They are recommended by national guidelines – I don’t see an issue . . .

BInt4

There had been poor service user engagement in the trust. One of the commissioners was very positive about MUs and referred to a recently developed place of birth tool for midwives to use when discussing birthplace options with women. She was open to innovation in both organisation and provision of MUs, if they met the needs of women, and was keen to engage service users who reflected the trust’s population of childbearing women in this process:

We need to really reconsider how we’re doing engagement with women. The maternity service liaison committee is not representative of our service users. And they’re not current service users, but white middle class woman who gave birth years ago.

BInt6

The midwifery and obstetric management at the OU without an AMU wanted to open one and optimise its utilisation. They were keen for the MU to be opt-out and for the OU to be only for women receiving obstetric care:

. . . it should be an opt-out, which was something we’ve sort of been driving for over the past year . . .

BInt7

They also wanted a new AMU to be on a different floor, to properly distinguish the MU from the OU. This was in contrast to the lead obstetrician at the OU that already had an AMU, who, while believing AMUs were safe, insisted they needed to be on the same floor and as close as possible to the OU:

. . . an alongside midwifery-led unit needs to be on the same floor because the lift is, you know, takes too long if you have a cord prolapse.

Bint9

A senior manager spoke about outright hostility from midwives working in the OU that was closed and replaced with a FMU:

. . . they were completely horrified at the idea of having a standalone midwifery-led unit.

BInt3

She went on to say that most of them were redeployed to work in the new FMU, despite their resistance. This had contributed to a gradual fall in numbers of women birthing there and, at the time of our data collection, there was discussion about it being no longer viable.

Relationships and communication between midwifery staff working in the FMU and those working in the local community were reportedly poor, even though the community midwives were based in the same building as the FMU. For example, when the prospect was raised that community midwives might be required to do shifts in the FMU, it appeared that they were then even less likely to promote the FMU to women, because this increased the chances of being called in to work there:

. . . there have been situations where the community midwives don’t promote a unit because it increases their workload.

BMFG2

Poor relationships between AMU and OU midwives were also described:

. . . really sad I think, because obviously you acknowledge the [OU] midwives and the high-risk care they give, but equally you have to acknowledge the low-risk care that low-risk midwives are giving, they’re all important but there is a sort of ‘them and us’.

BMFG1

Overall, it was apparent that Borodale lacked a consistent strategy for MU development. There were a number of points of fragmentation in the organisational structure with contradictory and conflicting views about MUs’ value and how they should be operationalised. This was contributing to poor interunit relationships and underutilisation of the FMU in particular.

Resources and priorities

Financial constraints were cited as playing a major part in the delays in developing MUs, with the trust currently in ‘financial special measures’. In this context, plans that did not involve generating income were low priority. Attracting women across borders from other trusts was unlikely to work because neighbouring trusts were already well served with MUs.

One difficulty was quantifying the financial benefit of MUs according to a midwifery manager:

Reducing the caesarean section rate for example, yes that’s great, but actually it doesn’t really make that much difference because we’re paid for maternity with a tariff, so in terms of actually getting more money for doing less caesarean sections it just doesn’t happen. So from a financial perspective it’s a difficult business case to write.

CInt6

(It should be noted that the tariff system has been updated recently to eliminate this perverse incentive.)

The maternity management team had successfully put the case to the trust board on the basis of ‘patient satisfaction’ and instituting a midwife-led pathway of care, with the focus on women and responding to national guidelines, rather than money. The proposal was accepted by the trust on that basis, but then floundered later and was never implemented. Cliffdale was very proud of as its high home birth rate and this was mentioned many times in staff interviews:

We’ve also got a really, really good home birth service with a high birth rate, so we try and encourage, try and give women that choice, and facilitate that choice.

CInt6

One manager expressed uncertainty about how opening a MU would affect this:

So I’ve often thought, if we did have the midwifery-led unit would our home birth rate go down and then once that goes down it gets less easy to support and promote.

CInt8

The women’s focus groups also suggested there might be a trade-off between home and MUs:

I think we’ve got the highest number of kind of home births in the country. I think a lot of those women, if they had a choice, might consider going to MUs.

CWFG1

Leadership

An additional barrier to the development of MUs had been the high turnover of both trust and maternity service leadership. The trust had had six CEOs in the last 10 years. This had contributed to the proposal for a MU being revised at least three times.

Leadership changes in midwifery also during this period had also brought ‘rethinks’ of proposals, slowing momentum, and resulted in the postponement of MU development. The current midwife manager, who was about to leave to work in another trust, was keen on developing an AMU on the site of the smaller OU. She had been influential in the previously approved AMU not proceeding:

The problem with the current hospital at [OU 2] is that, you know, it’s an old build, creaking at the seams, you know, so there was never the ideal place to create the right sort of environment. That fell through because a new midwifery manager turned up at the last minute and said, actually this space isn’t big enough.

CInt8

The midwifery manager had been very active in shaping the development of a proposal for an AMU on the site of the smaller OU:

I have been part of the initial discussions and been asked, ‘could we just make this room the MLU’? And I’ve said, ‘No, we cannot put a sticking plaster over it. If we want to do this we’ve got to do it properly’. It’s got to be co-located, so I think we all agree that that’s what we want . . . we want the midwifery and obstetric team to embrace it for it to work.

CInt2

However, as the CEO left during the development of the proposal and, as reported in interviews, the trust experienced financial difficulties, the MU proposal was shelved. One of the midwifery managers was also opposed to developing a FMU outside the city, which had been suggested by some service users. She was strongly committed to an AMU as the best form of MU. Her rationale for this included the likelihood that an AMU would be more popular with service users, the fact that FMUs were situated in nearby trusts and that it might ‘improve obstetric and midwifery relationships’ as a result of medical students and junior doctors observing normal births.

Women in the focus group felt that the neighbouring FMU was inaccessible, which was compounded by the fact that transfers from the FMU go to an OU even further away. It was not seen as a reasonable option for most women in the trust:

. . . the FMU outside of trust is a long way away and you end up at another trust if something happens. So it isn’t really viable for many women, and then the only other option is home birth which obviously won’t suit everyone.

CWFG2

This site demonstrated how perceptions about finances in the wider trust, tariffs and cross-boundary complexity can impinge on MU plans. When combined with frequent turnover of senior management posts, these factors can frustrate the realisation of a cohesive, longstanding plan for establishing MUs.

Culture and beliefs about the intervention

Although this site had the longest history of MUs of all the sites, staff working in the MUs generally felt unsupported, particularly at the FMU, which was under threat of closure during the data collection period. The FMU midwives reported poor levels of consultation by the host unit’s midwifery managers regarding policy changes related to the FMU and said that the trust obstructed their FMU promotional activities:

Well they [trust management] always pay me lip service . . . the staff said when they met with the managers, ‘but you don’t promote us’. And they said, ‘well we do’ but we’ve been fighting for a year to get a video on the trust website, of a tour of our birth centre – there’s promotion and there’s promotion!

DMFG3

They reported experiencing hostility when transferring women to the linked OU and believed that neighbouring area community midwives were advising women against booking at the FMU because of the likelihood of transfer. They summed up their feelings by commenting:

You do feel like the poor relation. And it seems so bizarre, because the women love it, but they [management] don’t seem to care what the women want. They just seem to care what works for them.

DMFG3

Not surprisingly, a fairly strong oppositional culture between midwives in the different work contexts (MUs and OUs) was expressed in each of the three MFGs:

It’s a very ‘them and us’ sort of culture. And it’s not fair, it’s not nice, because we’re all in the same profession.

MFG. 4.1

The two city-based women’s focus groups, which were mainly composed of women of Asian ethnicity, both stated that there was a lack of information about the AMUs:

To be fair, I didn’t even know there were other choices. And I thought the one in [FMU], for some reason, was privatised. So I thought it was something that you’d go to and pay for if you wanted it.

DWFG1

They believed that the trust had a responsibility to inform women about MUs.

In contrast, a senior midwifery manager reported that community midwives do inform women about the range of birthplace options and also:

Do an assessment at 36 weeks . . . and review the notes again, we’ve got a chance to offer home birth and renew the AMU information as well.

DInt2

However, she did not mention whether or not this included the FMU option.

The MFGs revealed that the trust had recently updated the maternity care notes to include the choice of AMUs but not the FMU, which had to be selected under the category of ‘other’ and requested specifically.

Organisation and staffing

The larger and longer-established AMU functioned more effectively than the smaller AMU. There have been ongoing problems, according to the MFGs, in establishing the AMU’s autonomy at the smaller AMU, which is located more centrally in the OU. The dividing doors were rarely closed, and the obstetric staff enter it and monitor activity:

The other birth centre, just feels like an alcove of the main delivery suite, and it doesn’t have much of a sitting area. The rooms are lovely, but it just doesn’t feel as separate when you walk in.

DMFG2

It had only recently established core staff and still did not have its own clinical lead. Staff from the smaller AMU are also moved to the OU more frequently than those based in the larger AMU, disrupting the functioning of the unit and the sense of identity of the midwives with the AMU:

. . . they’ll ring from delivery suite and say we need a midwife. And because I’m rotational, it’s me that has to go. And then you’re like, all huffy about it, and I don’t even know why because I work on delivery suite a lot, so I know delivery suite. But it’s that mentality. I’m on the birth centre so I can get pulled off there . . .

DMFG1

The midwives also reported that the AMUs were used as overflow rooms for the OU, with high-risk women receiving care there:

The other day we came in, and we might as well have been an OU. We’d got 23 weekers, and 30. There wasn’t one lady in there who was low risk.

DMFG2

Leadership

Dewsdale, although categorised as a high-performing MU site in terms of percentage of women birthing in a MU, was hindered by poor support and communication mechanisms between the MUs and the rest of maternity services. In contrast to the other high-performing sites, no mention was made of there being strong or effective midwifery leadership. Staff in the MUs felt marginalised and ignored, especially in the FMU. There was also dissonance between management perceptions of MU choice and women’s experience of it.

The relatively recent opening of a second AMU sits uneasily alongside a provisional plan at the trust to close the OU where it is based, possibly replacing it with a FMU, and situating a very large AMU at the other OU with combined births at that hospital of > 10,000. This plan also includes the closure of the existing rural FMU. The new AMU may have a capacity of > 2000 births per year, making it the largest AMU in England.

Culture and beliefs about the intervention

There was an overarching shared vision at strategy level from both the trust and maternity managers for improving the utilisation of the MU:

The vision is to increase the numbers, so we have the quality boards, and we have the numbers, and we are aiming to increase the deliveries in the midwifery-led unit, and home deliveries . . . we are continuously striving to increase it.

EInt4

Actually, a home birth service was well established and thriving, with its own case load team. It had a very good reputation beyond the boundaries of the trust, as described by a woman in a focus group:

And the trust home birth service has got a really good reputation, so I’d heard on social media how good the service was. And you get to have a midwife that comes to your house, so you see the same midwife. And she was the one that actually delivered.

EWFG1

The home birth service was the only example in the case study sites of a functioning case load model. It had also established a popular and well-used Facebook page.

Managers did not perceive that there would be any trade-off between the MU and home birth, unlike Cliffdale:

. . . for a lot of women it’s [AMU] that halfway house, you know, home birth feels a little bit too out there, but they don’t want to be on an obstetric-led ward.

EInt5

Although the staff believed that there was a commitment to developing MU services, women in the focus groups experienced a different agenda. There was still an impression of dominance by medical staff:

However, there is still this perceived hierarchy of doctors. Doctors know best, got to do as they say, and sometimes the language that’s used and the way things are presented, women feel . . . what I hear at reunions and when I speak to the women who have babies there, they feel that there wasn’t much choice. We had to do what the doctor said, or, we had to have this intervention because . . .

EInt3

Organisation and staffing

The workload on the OU took priority over the MU, resulting in MU midwives frequently being transferred there. From time to time, this meant that the AMU had to be closed:

When delivery is busy the first port of call is let’s grab a birth centre midwife. So of course you’re already then depleting your two midwives leaving potentially one on, or zero if they’re that busy, and so that service is then shut, it’s closed, you know, you can’t then offer that for the women.

EMFG2

The closure of the AMU filtered back to women, who became wary of booking there. They interpreted this to mean that the AMU was an optional extra to the main service provision on the OU:

Basically if you had a baby there you were lucky . . . it got round to women and every now and then I still meet a woman that says oh, but is it open?

EWFG1

When I did visit the trust hospital, they [the midwives] said that the midwife-led unit can shut if the labour ward is busy, they have to pull the midwives out of there. So the labour ward will always take precedent over it.

EWFG2

Resources and priorities

The trust undertook a number of promotional activities for the AMU, including the 36-week antenatal appointment in the AMU to encourage women to choose the AMU for birth. However, midwives commented that 36 weeks seemed a little late for planning place of birth, because many women will have made a decision before then:

. . . it’s difficult because that choice isn’t made until 36 weeks and I understand why, let’s wait until 36-week appointment and then assess the suitability for birth centre. But I think by then women have already sort of got their head around where they’re going to have their baby. It’s too late.

EMFG2

Some midwives described having very little time to discuss birthplace options with women. They believed that low-risk women should be encouraged to express their preference in a care plan:

I just think we don’t give women enough time and education to make proper informed choices . . . they get told where to go and what to do, and that’s it. But it’s never mentioned to them along the way, if you would like to have a more low-risk environment.

EMFG1

The trust also had monthly ‘early pregnancy evening’ promotional events in the maternity unit which included tours of the AMU.

EMFG1

Eisedale was struggling to overturn a poor reputation within the wider trust, but had a vision for MUs alongside its existing and popular case-loading home-birth team. However, an obstetric focus undermined its ability to operationalise MU care because the OU demands always took precedence over those of the AMU. It also operated an opt-in policy for accessing the AMU. Together, these factors resulted in the frequent transfer of AMU midwives and closures of the service.

Culture and beliefs about the intervention

Within Ferndale, the culture and beliefs were particularly focused on reducing risk. It was also evident that the culture was loaded towards a preference for obstetric-led care. One corollary of this was a pervasive paternalism according to many respondents:

It’s very, very medicalised. But I think a lot of it is obviously within the hearts and minds of the midwives . . . what I know from other things that I’ve been involved in is that as midwives we don’t realise we use a lot of closed questioning of women. So we talk about choice, but we don’t actually mean choice, we mean choice for ourselves and what’s, you know . . . what choice suits us.

FInt6

Normal labour and birth was perceived to be the exception and was therefore commented on when it occurred:

I was walking out of the delivery suite and an obstetrician was walking next to me and there was a mum and new baby just leaving to go to a postnatal ward. And he said, ‘oh isn’t it lovely to see a nice normal birth’? And it was like, oh goodness me, you know, we don’t need to be hearing and saying those sorts of thing . . . said in complete innocence, but actually when you think about what they were saying in a way it’s like what we aren’t able to offer women.

FInt4

Some staff took the view that there was no need for a MU because low-risk care was already provided on the OUs:

But what’s quite interesting is that the current obstetric service can accommodate low-risk deliveries. So I mean if you look at both sides of the city there’s about 30% of all the women come through never see a doctor, so they could deliver anywhere really, the place of delivery is actually a lot less critical.

Int3

However, others felt that the strong team ethos between obstetricians and midwives was on the obstetrician’s terms:

I’m really impressed by how supportive the obstetricians are of the midwives; they are incredibly supportive, you know. It’s really fantastic team working. But I just feel it’s all meeting the obstetric point of view and we are midwives.

FInt7

Newly qualified midwives who trained locally were able to handle complications with confidence because that was what they were regularly exposed to, but they felt deskilled about normal birth:

I think, that because everyone has worked in such a high-risk environment, you become deskilled to an extent, and feel a bit apprehensive about trusting that women can have babies who are low risk.

FMFG1

Leadership

Once again, leadership came up as key reason why MUs had not been developed:

I think it’s lack of leadership and influence that’s probably blocking change because you need that leadership and vision. And as I keep telling people, the modern NHS doesn’t appoint leaders it appoints managers . . .

FInt2

. . . if we had that leadership and we had that real definite, yeah we’re going to do this, we’re going to . . . you know, walls could be knocked down and things could be done.

FMFG1

One of the new senior midwifery managers wanted to develop MUs, but felt that midwives would resist it because they would not have the confidence to work in them. She did not believe it would be possible to open an AMU owing to the obstetric dominance there, and favoured recruiting an entirely new team of midwives with a commitment to FMUs:

. . . the midwives would find it very difficult to sort of just be left alone because they quite like the way that they work at the moment. And I think that we need different talent group of midwives going into a standalone with a different philosophy of care.

FInt3

Resources and priorities

Some staff believed that maternity services reconfiguration depended, first of all, on how neonatal services would be rationalised. Until final decisions were made about this, no decision could be made about obstetric services and staff lamented the loss of maternity independence over recent years:

And if you look back in history it was obstetricians that developed neonatal services but now paediatricians have taken over neonatal services so they control them and we have to follow them.

FInt8

There were also concerns about the cost of furnishing a new FMU, as the trust had significant financial challenges:

I think the resistance comes from, how can we financially afford this? The trust is hugely in debt.

FInt7

Ferndale demonstrated how an embedded culture can frustrate change profoundly, using a range of strategies, including believing that the desired outcome (midwife-led care) is already being achieved, or by rationalising why the change will not work in this unique setting, or by projecting the rationale for no change onto another cause.

Environment of midwifery-led unit

The environment is a very important aspect of women’s decision to go to a MU.

The look and feel of the unit was the most significant feature for participants. This included it feeling homely, comfortable and private and facilitating a sense of relaxation or calm:

I remember picturing me and my partner in that really nice room, and it made the whole prospect of giving birth, which is scary when it’s your first, a lot easier to deal with.

EWFG2

. . . it’s like being home but with midwives, so they’ve got like, lovely flowery duvet covers or whatever, like it feels homely and cosy and that was really appealing.

CWFG2

Birth pools were the most frequently wanted facility, with lighting, kitchenette, music playing and balls less frequently mentioned.

Familiarity with the midwifery-led unit

When women were familiar with the MU – its environment, staff, location and accessibility – it assisted them to plan to give birth there.

When a MU (particularly a FMU) offered more than a birthplace (e.g. antenatal classes, postnatal checks or breastfeeding support), it was valued by women and often contributed to them planning to use the facility:

. . . so for us it was kind of ‘oh yeah, we’ve already been a few times’ [for antenatal care], you know where you’re going, it’s easier to drive there [FMU], because you’re already there, you know about the parking . . . you know your facilities and stuff like that, so you’re a bit more familiar with it. So it’s not as scary.

AInt1

After going to our antenatal classes at [FMU] I actually changed my mind and decided that I loved it there and wanted to go there.

AInt1

Continuity of midwifery care

Women expressed their desire for continuity of care, experience or lack of continuity. They indicated that continuity of midwifery care is important to, and wanted by, women. It was only experienced by women planning home birth in one site, or informally due to integrated FMU/community care in two sites, where women may see a midwife they know from antenatal care if she is rostered in FMU when they labour. Women in a site without a MU thought that continuity would be part of the package. If MUs offered continuity of care they would attract more women:

Because my midwife was my community midwife who also worked at [FMU], so for me I wanted that.

BWFG1

. . . And if I could have been at least partially guaranteed that I could have had the same midwife at [FMU] that I was seeing, and I would see the same person through the pregnancy, that would really encourage me to go to the MU rather than the [OU/AMU] hospital, because I would have felt like I had that advocate.

AWFG2

Influential people in birthplace decision-making

Women mentioned midwives as positively influential in their MU birthplace decision-making, more often than others, most often as a result of their practice and/or their promotion of the MU. Some women experienced midwives as not providing any information, making assumptions or not promoting MUs:

I think because the staff have been there [FMU] so long, they’re so positive about what they have. And the minute you speak to them about it then it just reassures you. But their experience and their passion for the place just sort of encourages you to stay there really.

DWFG3

I think the problem with it is, they make you aware that it’s there, but they tell you about the negatives, not the positives of it . . .

BWFG2

Other influences on the women (not always for a MU birth) were antenatal classes, family and community, and then, much less frequently, friends, their partner and a doctor:

I had already made up my mind, but then NCT [National Childbirth Trust antenatal class] definitely, like yeah I’ve made the right decision, I definitely need to go to the [MU], it’s the best, it’s amazing, and all of that.

AWFG2

And there was just a lot of fear amongst everyone [in antenatal class] about how to cope with giving birth. And I think that sort of reinforced my feeling that I would need a medicalised birth.

CWFG1

And then my mum and stepdad are paramedics, and so they convinced me, I suppose, in a way, to pick [OU/AMU]. So, because they know, they blue light people over there.

AWFG2

Role of facilities and services in decision

Many women planned to give birth in an OU, or in an AMU adjacent to one, because there are specialist facilities and services available there:

Just the idea of being in a hospital room and the doctor just being outside the door. Even though they could be busy, and there might be delays that way, I think just the safety of knowing that they’re there.

FWFG2

The belief that birth is inherently risky and unpredictable was widespread for women who planned an OU or AMU birth, where they believed that they and their baby would be ‘safe’:

. . . [regarding AMU] they’ve got a delivery suite [OU] underneath, because to me that would take away everything for me. If the risk was my only panic, that’s me, it’s mitigated, it’s gone.

BWFG1

A much smaller proportion of women had confidence in the birth process and the safety of a FMU. Some women mentioned their desire to avoid the risk of (unnecessary) intervention in the OU as part of their decision to plan to give birth in a MU:

I thought it’s very, very unlikely that something’s going to happen where literally I need medical intervention like this within the next 10 minutes, so to speak. And so because of that I thought I’m prepared to take that risk that if I need to be transferred, that transfer will be safe.

BWFG1

Location of midwifery-led unit

The location of a birthplace was an important factor in many women’s decision-making. For some, it was related to how close it was to their home, for access in labour. For others, the unit’s location relative to the referral OU, for access on transfer, took priority. The close location of AMUs to an OU was frequently mentioned as optimal:

. . . so even that, even if it’s a 5-minute drive, to me, [OU/AMU] like less than 5 minutes. But it just felt like hours just getting there, because you’re in so much pain. So to me, [FMU town] just like way far.

DWFG2

So you’re half an hour down the motorway then, haven’t you, and if you have a poorly baby you don’t – you want to be 30 seconds in next door, not 30 minutes down the motorway.

AWFG1

In contrast, the distant location of FMUs from the OU was an issue for some women, although not mentioned as frequently as the distance the FMU was from their home for access during labour (particularly when the FMU is rurally located and the women live in a city with an OU ± AMU):

I wanted to go to [FMU] because it was literally 2 minutes from my house. And I really didn’t . . . because trying to get to [OU/AMU] from my house, if you got stuck in traffic, it would just be . . . and I had my second one on the toilet at home.

BWFG1

Because my anxiety is that if anything goes wrong it’s a long way to hospital, which if you’re at [FMU], for me, that’s how I felt about it.

DWFG3

Cross-trust boundary issues were a factor for some women, with some being discouraged from booking to give birth in another trust and others denied the option:

I got told ‘no’ to [nearby trust town].

BWFG2

I think it depends on your postcode actually, so possibly someone that lived more in central [trust area] might not be able to go to [nearby trust].

EWFG1

Booking process and timing

Women expressed uncertainty about the booking process, such as whether or not it is necessary, how it is done, if it is possible to change your mind and if it is for births or scans (where notes are to be held).

Most sites complete the booking early in pregnancy, at the booking visit, with those appearing to have less discussion and flexibility. Others have some flexibility, although women reported experience of different processes within the same trust.

Some of those sites that formalise the booking later postpone the decision for MU birth until ‘low-risk’ status is confirmed in the third trimester. Some of these are well women who cannot plan a MU birth ‘just in case’ they develop a risk factor that will make them ineligible, whereas others are women who have a risk factor, such as a low-lying placenta, which requires review in the third trimester to see if it has resolved:

I wasn’t aware that you had to decide before you went in for your booking appointment, so I was asked on the spot and I didn’t know. But the midwife said that you have to choose now because they have to book the hospital in advance.

BWFG2

Yeah, because maybe sometimes you feel like well I’ve made my decision now, so that’s what I need to do. As opposed to knowing that it is actually changeable.

DWFG2

Eligibility criteria

Some women had poor knowledge of eligibility criteria, whereas others challenged them or were frustrated by them and others accepted them. There was a sense that the criteria were very risk averse and overly restrictive. Women also found variation in the way they were interpreted or implemented. The ability to have a review with a senior midwife or obstetrician was valued by some, but felt unnecessarily bureaucratic by others. There was a sense that a ‘fight’ was necessary to get a reasonable compromise:

At first they said with her she was going to be too big. Then they scanned me and said she was too small. So yeah, it was literally on border whether I could go or not.

BWFG1

[Regarding FMU] it’s so low risk though, if you’ve got any slight thing and it’s just took off you.

BWFG1

Capacity

Midwifery-led unit capacity is an issue for women, either the whole unit being full or closed, or an unstaffed FMU not able to be opened for a birth owing to no staff being available. Women had also experienced OUs being full. They found it problematic to plan to give birth somewhere when there was uncertainty about its availability when they were in labour and tried to book somewhere where this was least likely to happen, which sometimes meant changing the location or type of maternity unit they booked. Some women commented on the stress that those working in the NHS are under, with particular focus on midwives, who were seen as having high work pressures, limited time and providing reduced or minimal care:

What is scary is the number of rooms, you think if you get there and it’s full . . . and you’ve planned that for 9 months in your head . . . Like the day we went in there [AMU] for the tour, all the rooms were full . . . because they’d have to overflow into the labour one, I think god, that would have been really upsetting.

EWFG2

So I was thinking of going to, to the [nearby trust], purely because there’s a Which? Birth Choices thing [website], which said that the [AMU] had been closed so and so many times in the previous year due to understaffing. And I thought I’m not going to risk getting there on the day and ending up in labour ward [OU].

EWFG1

And I don’t think that’s the fault of the midwives concerned at all, I just think there aren’t enough of them. But that’s the thing, there aren’t enough of them.

DWFG3

External policy and incentives

The macro context of national maternity policy in England has a number of intersecting priorities that frustrate commissioner and provider strategic planning for MUs in local settings, and examples of this are given in this section.

Although the evidence base of MUs is quite well disseminated and national guidelines recommend them for every trust, a number of other macro factors impinge on developing them. These include the medicalisation of birth, contributing to greater obstetric surveillance that has overseen a shift to birthing on busy obstetric-led labour wards, the increased complexity of risks in the childbearing population, such that fewer women are eligible for MUs, and a risk-averse culture that militates against the use of MUs.

The provision of choice around place of birth has been policy in maternity services for > 20 years, but local services are not audited on how they operationalise this regarding MUs, and hence our data show considerable inconsistency.

Finally, financial constraints on the NHS were spoken about ubiquitously. As one respondent expressed it, ‘we’re in a period of austerity now’ (AInt4).

The medicalisation of birth, so that the wider population views giving birth as safe only on a labour ward, was highlighted in a number of interviews and focus groups:

I think also as we’ve been culturally accustomed to birth taking place in hospital since the 1970s, and to step outside of that model you’ve got to face your family and your peers and actually have a good reason why you want to birth outside that accepted model . . . you’re battling a mindset. Hospital is perceived as safety, the ‘just in case’ option. They’ve lost the ability to believe in their own bodies haven’t they, this sort of generation coming through don’t actually think even that it’s a natural thing for a woman’s body to do, they just see it as a medicalised model of giving birth.

BMFG2

Changing this perception was seen to require a major cultural shift:

We need to change some of the culture I think. The low-risk women would automatically go there, so it would be seen as part of the norm. But that’s a massive culture shift isn’t it, probably which would take almost a generation, wouldn’t it? So actually a generation of teenagers out there now who are going to have babies in the next 10 years, the message would be, actually if everything is low risk this is the place of birth, this is where you would go and have your baby.

CInt4

The changing perceptions of risk among health professionals was having an impact on access to MUs:

We’ve seen a bit of a reduction [in FMU births], because of the impact of Saving Babies’ Lives [NHS England. Saving Babies’ Lives. A Care Bundle for Reducing Stillbirth. London: NHS England; 2016].

Right, yeah. The policy framework . . .

So we’re seeing more women being scanned for growth, and being induced. More inductions because of the query that the growth’s falling off, or static, or reduced movement. So we are losing births to that.

Services therefore tended to attribute low provision, or take-up, of MUs to what they perceived to be increasing levels of obstetric risk:

Because we have many ladies who are desperate to use the birth centre, but they just don’t fit the criteria. We’re a high-risk population in other words.

DInt5

I think it’s difficult as well, because a lot of our ladies are high risk. We do get the low-risk ladies, but the majority are inductions, we induce a lot of people. It is generally a high-risk population we’re dealing with.

FMFG1

Two of these sites were high performing, indicating that, despite apparently having a higher-risk population, their MUs were still well used. This challenges the view that a changing risk profile in local women is a justification for not having a MU or a rationale for a reduced throughput in an existing MU.

A risk and litigation agenda was clearly evident in all sites but was predominantly constructed around the risks of birthing in MUs:

There’s also the potential clinical risks of people giving birth in those areas [AMUs]. And we had an unfortunate death about 3 years ago, someone who didn’t want a vaginal examination and ended up fully dilated pushing for a considerable period of time and in the end the decision was made, we’ve just got to transfer, and unfortunately the baby died in transfer.

BInt4

There might be a degree of fear that if people started saying that, you can go in there [to the MU], you are constantly like . . . women have to be told the risks . . . because of the litigation now.

FMFG1

We’ve got ladies going through solicitors, it’s going very, very high up within the trust to make sure that these MU women are safe, and make sure the midwives are covered and safe with these ladies. Because they [midwives] are scared.

BMFG2

These litigation concerns expose a paradoxical element because the evidence shows that healthy women who plan birth in OUs are subjecting themselves to greater risk.

Patient needs and resources

The second salient construct of the outer setting was that of patient needs and resources.

Respondents’ awareness about resource scarcity within the NHS was widespread across the case study sites:

You know, the limiting factors are the fact that the country’s in a financial crisis and health care is crippled in terms of finance. So it comes down to the money. I think, you know, there’s a will, there’s an understanding but actually being able to finance these things [MUs] means that you need money from the treasury to do so.

CInt3

I think the whole financial situation within the trust at the moment is a driver . . . Unfortunately all our finance team will only see is the figure at the bottom of the page . . . it is a sort of finance-driven organisation and you’re forever trying to find ways of saving money, cutting costs, etc.

BInt5

I think the finances is something that we all skim around but I guess that’s the crux of what’s gone on, it is the money . . .

DInt1

I just have this undertone that decisions, some of the decisions are being made not for the women, but for money. Not to gain money, but to save money.

EMFG1

This discourse of scarcity was not questioned by anyone we interviewed and there was a sense of resignation to having to compete with other trust services for a limited and apparently diminishing reservoir of money.

These factors emanating from the national context are having real effects on the implementation of MU services locally.

All sites were aware of the imperative to provide the choice of MU provision for service users of a MU, but its application was generally suboptimal, both in the availability of MUs and in the mechanism of how choice was presented. Two case study sites had no AMU and three had no FMU. In sites with MUs, women reported that they either did not know about them or did not know they could choose to birth there:

But I mean, I’ll be honest, MUs, are they a relatively new concept? I didn’t know they existed.

EWFG2

Well it’s just that nobody gave us the information about it. That’s the main thing. I didn’t know nothing about it . . .

DWFG1

I even didn’t know I’d got a choice [where to birth], but then friends said go to the birth centre. I said am I allowed? So then I did my own search.

AWFG1

These women’s experiences raise important questions about how embedded MUs are as a mainstream place of birth option alongside OUs within local maternity services, and to what extent trusts headline their value and purpose.

Under the CFIR, the domain of intervention characteristics and its constructs of ‘cost’ and ‘adaptability’ will be explored next.

Cost

One of the respondents’ key concerns about the introduction of MUs, especially FMUs, was their perceived cost. Economic evaluations of both AMUs and FMUs have been favourable, as discussed in Chapter 1. However, there is a paucity of information about the set-up costs of developing them. All but two of our sites already had MUs, either an AMU or a FMU, and three sites had both. Start-up costs were therefore relevant in only two sites. Virtually all of our findings around cost pertained to FMUs.

Within an environment of financial constraint, the perception that a service is underused quickly translates into it being viewed as inefficient and poor value for money. This especially affected low-volume FMUs, for which, across three case study sites, there were plans to close them:

If you spoke to any of the consultants I am sure they would say it [FMU] should be closed because it’s a waste of money. And it’s an unfair allocation of resources, in a relatively resource-poor environment.

BInt3

Well it [FMU] is small and we do have to understand how viable it [FMU] is because you can’t spread yourself so thin. And I mean, to be fair, we can only provide so much, you can’t spread yourself too thinly and become unsafe. So it [FMU] is difficult to manage because we’re covering so many other areas, and the birth rates numbers are very low.

AInt3

. . . That’s essentially where we’re coming from there, and also our plans look to shut the very small unit that we have . . . because that’s just not viable, the numbers are falling.

DInt5

In addition, the start-up costs for staffing a new FMU were judged to be prohibitive at Ferndale, which had no MUs:

You couldn’t possibly do a standalone unit without having more staff, so it’s a cost thing. It’s staffing levels. And unless the trust are willing to put the money into it, you wouldn’t be able to do it, it wouldn’t be feasible.

FInt2

Adaptability

A second feature of the introduction of MUs was their perceived adaptability and compatibility with existing services. For example, finding an appropriate space was paramount. One approach has been to recommission space within an existing footprint, a less challenging and less costly option than commissioning a new build. This is how AMUs have been principally developed in the case study sites. However, that has also been problematic because of the increased birth rate nationally, so many existing OUs already had capacity issues:

But the problem with [OU/AMU], it was designed for 4500 deliveries and they’re now delivering 6700 through [OU/AMU].

BInt7

Only one site commissioned a new build for a MU. More commonly, consideration of this option did not result in a new MU but years of plans, revisions and more plans:

We had three kind of projects to rebuild in the last 12 years because we are underspaced, undercapacity, each of which have kind of got so far, got the architects plans and then the trust have said, actually we need an A&E [accident and emergency], or we need money for this . . .

BInt1

One potential solution to unequal provision of MUs would be for adjoining local maternity services to share MU facilities (e.g. one service that does not have a FMU could share with a neighbouring service that does have one). Our data showed that this rarely happened and, if it did, it created organisational and logistic problems, as well as access problems for women and their families. This was because of the tariff system of payment, premised on a competitive model:

. . . at the moment the tariff does provide a complete barrier because if we were expecting our two trusts to work together then each trust is going to want to retain all the tariff for the women. So there isn’t an incentive for the women living in the [trust] catchment area to be cared for by the [other trust] because they lost the tariff.

BInt6

Two trusts operated other services from their FMUs, including antenatal clinics, antenatal education and scanning, as a way of attaching women to the service and optimising alternative uses for the building. Respondents believed that this helped justify continued investment in the FMU.

Beliefs about the intervention

Convictions about the benefits of an intervention are known to strongly influence its successful introduction. 110,111 Despite the evidence base of MUs being well known locally, interviews and focus groups revealed substantial ambivalence about value of MUs, especially among clinicians. It was clear that having an AMU or a FMU and working in them positively influenced perceptions, especially among midwives:

. . . it is great and it’s, you know, less medical intervention, it’s cheaper and it’s better for the women. It’s just better all round . . .

AInt6

I think the beauty of midwifery-led units is that it helps midwives to be autonomous practitioners. It allows for choice for the women.

CInt2

Similar comments were made by obstetricians and neonatologists:

I have mainly had positive experiences with midwifery-led units. I worked as a registrar in [place] and there was a co-located one there and from a neonatal point of view I wouldn’t say there are any more troublesome deliveries there than there are elsewhere.

BInt5

Some of the trust staff expressed a preference for an AMU over a FMU:

I think majority of women and all my friends will opt for alongside, because most women do want the option of midwifery led but if anything goes wrong they just want to go down that corridor, through that door.

DInt4

It was notable that, despite the national evidence indicating difference in rates of intervention according to setting, some midwives in strategic posts expressed the view that midwifery care for low-risk women could occur effectively within an OU, with or without dedicated low-risk rooms:

Well there is what we call a midwifery-led unit which is incorporated within the delivery suite in the [OU2] which was set up by [name] which must be about 25 years ago . . . two or three rooms where the midwife-led care was given.

FInt1

. . . it’s about ensuring that you have quite a high percentage of vaginal delivery, and again midwifery-led care being delivered in labour ward and . . . and the protection of it, that’s the important thing . . .

EInt1

. . . you can deliver a woman on midwife-led care wherever you are. We don’t need to have these alongside units. A midwife is a midwife is a midwife.

BInt2

The midwife manager cited above went on to criticise the OU staff for not utilising the three low-risk rooms within the OU:

There are three beautiful, low-risk, midwife-led rooms on the end of the [OU] delivery suite. But for some reason the staff on that site cannot see them in that way, and instead want an AMU.

BInt2

The view that FMUs were unsafe was held by some midwives according to the MFGs:

We’ve had students come back to us and say that the [midwife] mentors have said they’re [FMUs] not safe to birth with in, especially primips, because . . . and they’ve said to the women, you have a greater risk of dying.

DMFG2

Finally, some staff believed that women from particular ethnic groups preferred OUs to MUs. Those expressing this view presented it as something that could not be addressed or changed:

. . . We have women from ethnic minorities that back home, for them, they want the continuous monitoring, they want the doctors to deliver them. So they have that mentality. Yeah, we’ve got couches in our birth centre and they’re like, where’s the bed. On the birth centre, they won’t stay here, there’s no bed.

EMFG2

. . . women coming through of a different culture, and they particularly want it obstetric led.

BInt4

However, two sites with significant populations of women with Asian ethnic backgrounds had AMUs and FMUs that were popular and heavily used by these women. The staff discussed the strategies they used to engage with these communities and change inaccurate perceptions regarding MUs:

We have had others suggest that Asian women are not going to buy into midwifery-led units. I used to speak the husbands, get them to come and visit and talk to them because they’re, oh my husband doesn’t want me . . . now he wants me to go to . . . one lady that came to my clinic and all her family had been to the OU so she was going there . . . I said, just come to the [FMU] and have a look round. And she came and had a fantastic birth. So it’s word of mouth. The Asian girls love it, they absolutely love it.

AMFG1

Although across all case study sites, the provision of MUs was generally viewed favourably, for some maternity stakeholders personal opinions or beliefs about them appeared to trump evidence and this is likely to have affected MU provision and utilisation in some of the case study sites.

Access to knowledge and information

The provision of evidence-based sources and appropriate skills training by local health services is known to facilitate the introduction of optimum care models. 112,113 Working in MUs, although it should be the basic default skill set for every midwife, may require considerable adjustment for midwives more used to working in OUs. Many interviewees suggested that midwives were losing the skills for caring for normal labour and birth:

Because everyone has worked in such a high-risk environment, you become deskilled to an extent, and feel a bit apprehensive about, I don’t know, caring for . . . you know, trusting that women can have babies low risk.

FMFG2

One of the effects that that has had, is that is has decreased a lot of the midwives’ confidence in this unit, of providing low-risk care, because they don’t have the environment, a consistent one, in which to do it properly . . . when you’re on labour ward you become over-reliant on the doctors . . .

EMFG2

Respondents felt that midwives need particular sensory skills and a holistic outlook to work effectively in MUs:

You’re using your eyes and your ear and your hands and you’re more tuned into that woman . . . which is what you need if you go midwifery led.

DInt4

The ability to work autonomously was also seen as important:

It’s for us to show them [students] what working autonomously as a midwife looks like, and how these . . . you know, truly let our women labour uninterrupted under her own steam in a calm and peaceful environment, what it actually looks like. And that’s a precious thing, it really is, and we need to spread the word.

AMFG1

Midwives in the focus groups said there was a need for training in normal birth skills for qualified midwives:

I’d like to make physiological birth and normal birth more accessible to them . . . our sort of mandatory week of training that we do is very obstetric led. So what I would love to do is a session, a day, on physiological birth, and normal birth, and making midwives more confident in doing it.

EMFG1

Others suggested that the default training for student midwives should be in low-risk care. From there, midwives can then learn to recognise risk, rather than the other way around:

I trained at the satellite birth centre and then went on to go to our birth centre. But not every student has the opportunity to actually train on the birth centre, so you are trained in that thought process. You are trained in completely low risk, and then you spot the high risk. Whereas there are a lot of new midwives who are trained solely on high risk because there aren’t enough places on birth centres to send these students.

One case study site had introduced NLS training, especially for midwives working in FMUs:

. . . a few of the midwives in the birth centres are NLS instructors, so they have helped raise the profile of that even higher, so even more than what we thought was the basic requirement, they’ve actually taken it, you know, one step further . . . on every shift there was an NLS-trained midwife. So we’ve actually made that, mandatory really . . .

AInt7

Readiness for implementation

Consulting widely and inclusively, with a particular focus on meaningful and ongoing dialogue with local service users when developing MUs, was highlighted in many interviews:

. . . it’s that desire for genuine and well-informed public consultation so that they know really what is being proposed and what the implications are for it.

EInt4

Armidale was especially effective in this area, particularly during the major reconfiguration, but many participants recognised the need to this to be ongoing:

. . . there was a big consultation with service users, and that was called the Meeting Patient Needs programme . . . I think that was the big driver. It meant massive change for the service in [FMU place] having standalone unit.

AInt2

So I think having service users in at the ground level . . . Having genuine engagement with them rather than tick box; and I suppose also building a network of people that are interested in birth centres to work alongside the midwives, people from the community. But then they’ll be back in their own communities talking to women and that helps filter out information.

AMFG1

Some services struggled to engage with a truly representative sample of service users:

We need to really reconsider how we’re doing engagement with women. We’ve got a couple maternity service liaison committees, but they’re not representative of our service users. So it might be a white middle-class woman who gave birth 8 years ago chairing them, and they’ve got chips on their shoulders about certain aspects of care.

BInt6

I’ve been what’s known as a patient partner, and I think this might be unique, certainly in terminology. And it just means that I go give a patient’s point of view at various board meetings. I’m not . . . I mean I’m not elected, I’m not a representative patient or anything like that, simply a patient.

DInt2

Therefore, multidisciplinary collaboration within the trust, with sign-up across the organisation, was also identified:

In order to get a birth centre off the ground, there has to be the political will and you need all the obstetricians on board.

CInt7

So you know, just good communication as well between the birth centre and labour ward that everybody is singing from the same hymn, and that includes, you know, the multidisciplinary, the obstetricians as well.

EInt9

Culture, networks, communication

Many of the MUs described a poor interface with their linked OUs:

There will always be a ‘them and us’. And I think, because of the medical model of obstetrics, I think there’s always been higher importance put on obstetric care.

AInt7

It seemed that a ‘them and us’ dynamic was exacerbated by individuals:

I think unfortunately the issue was the previous lead. She managed to create that ‘us and them’.

EMFG1

I think, we have certain sort of delivery suite co-ordinators who have no time for the birth centres at all . . . there is a bit of ‘them and us’.

EMFG2

One manifestation of this was the negative reception some FMU midwives received on the OU when transferring a woman:

I went on a transfer, and the reception I got was non-existent.

What do you mean?

There was nobody waiting. They weren’t waiting for me when we go there, there wasn’t a cot in the room, no midwife came, I had to find somebody.

But they were told, they’re always told ahead that you’re coming?

Oh yeah, they know you’re coming. I’ve been greeted with oh, here comes another failure from [FMU].

The MU midwives felt that some OU staff had a perception that workload on MUs was less intense:

I think there is this myth that we low-risk midwives perhaps sit around twiddling our thumbs quite often and we don’t. I think that’s something that’s sort of spread around, the rumour. It’s really sad, it is a very ‘them and us’ situation.

BInt4

It was noticeable that staff in FMUs in two of the three sites had felt detached from and unsupported by their host trusts. In one site, the staff’s attempts at promoting the FMU have been frustrated by lack of engagement from the trust:

And they [trust management] kept putting us back, putting us back . . . If you look at all the leaflets that they do send out from the trust, [FMU]’s this little add-on at the end, in the fine print at the bottom. It’s not promoted, you know.

CMFG3

The staff at another FMU said that the two linked OUs, which seemed ambivalent regarding encouraging women to birth in the FMU, were nevertheless happy to use them for postnatal transfers when they got busy:

And they’re quick to send women back to us with their feeding concerns, because obviously bed shortages over at their units.

Oh I see, you still take postnatal inpatients?

Yeah. From everywhere, yeah. Regularly. And that’s what’s promoted. If they’re busy at their ward, which they always are because they’re rammed, we get phone calls. Can you take them? . . . We’re kind of like a half-way house sometimes, it feels like. But that’s because that suits them. When we talk about being here [for births], they’re – but you’re too far out, nobody can come.

Planning

When establishing a new MU, promotional events and materials, especially regarding FMUs, was crucial. In one site, where a FMU was thriving, promotion initiatives and positive messages from the midwives were fundamental to its success:

I think if you’ve got that promotion, and we’ve got films on our website where we are promoting the birth centre and so I think it’s about promotional material to make sure women know.

AInt5

I think there needs to be a very good website – because I know women shop around and look at websites and we’ve got films on our website where we are promoting the birth centre.

EInt1

Promotional activities need to be ongoing, as another site with an established FMU stopped tours of the facility for women and this appeared to diminish bookings:

One of the things I’ve heard, certainly around the standalone is that they used to have regular tours as part of the introduction for the women, but that was pulled. And I think women will hear about it and they can go and have a look, but it’s that automatic tour that was part of the introduction that they felt affected the numbers.

BInt3

One site actually performance managed their staff on recruiting women to their new FMU:

I guess one of the thoughts was has integration of your community with your birth centre, has that influenced the promotion of birth centres by the community midwives?

Yeah, greatly. When we were really, really trying to embed the unit and promote it, we really needed the midwives to embrace it. So we set targets . . . we had to audit everybody’s referrals. To see how many mums they did, and if they didn’t do any we wanted to know why. We said if you don’t start referring women, we’ll pull you off your clinic.

Finally, writing a strong business case is especially important to help win the financial argument:

It’s a difficult business case to write and in the end the trust accepted that there would never be a financial motivation for this, that it would always be about patient satisfaction, pathway of patient care and it would be all about the patient, so it was not going really going to be about making more money.

CInt6

It really needs the trust, the managers of maternity services, to write a business case, and to get that signed off by their board.

FInt1

Engaging: champions

The second aspect in the process of developing a MU was harnessing the persuasive power of champions to engage and influence other key stakeholders, especially those who remain ambivalent about the change:

If you can get a good obstetrician to support you, in terms of that, I think you need that one person to champion your corner from that clinical leadership side that would be key.

AInt1

Champions are often individuals who have worked within the model previously as they are more likely to be familiar with the arguments against the change:

I think they need a champion. There needs to be someone in the system that drives it who has worked in a midwifery-led unit in the past, they obviously have a very strong view and a very strong voice.

EInt4

Having that presence at a strategic level is also important:

I still think we have a long way to go at getting maternity services discussed at board level, and having an identified board-level champion within the CCG [Clinical Commissioning Group], as well as within the provider, and making sure that that role is more than just a name.

EInt6

Executing

One of the findings from the mapping survey in the first stage of the project was that a key to high utilisation of MUs was the streamlining of the care pathway to access them when women go into labour.

This did not happen in the majority of sites and was especially commented on by the WFGs.

Many women were not aware that they were birthing in a MU, or that they existed within their local maternity service. The actualising of choice was compromised by poor organisational processes that sometimes required a choice of place of birth at the beginning of pregnancy:

I wasn’t aware that you had to decide [where to birth] before you went in for your booking appointment, so I was asked on the spot and I didn’t know. But the midwife said that you have to choose now because they have to book the hospital in advance. Then I called up my mum and she suggested [OU/AMU].

BWFG1

Yeah, because maybe sometimes you feel like well I’ve made my decision [about place of birth] now, so that’s what I need to do. As opposed to knowing that it is actually changeable.

DWFG2

However, on other occasions women were discouraged from booking in a FMU because they might develop complications:

They said ‘don’t make the decision now . . . wait until towards the end and see if you are still OK’ . . .

BInt4

This had a negative effect because the woman wanted the reassurance of having a plan.

Instead, a graduated discussion commencing at the booking appointment and then revisiting it on a later occasion and especially at 36 weeks was recommended:

So I think that orientation into the community needs to include that birthplace discussion and promote birth centres . . . to make sure they offer low-risk women the birth centre.

EInt7

I think when they have their birth plans done at around 36 weeks it’s worth mentioning it [MU] then, because at booking when you get bombarded with so much information, so many leaflets and often they’ll say to us, oh I haven’t read it, or I haven’t had a chance to read it. So giving that a bit later on when they’ve had chance to digest all of that information, might be more beneficial.

DMFG1

Having an opt-out policy was commonly viewed as the best way to maximise access to MUs:

The AMU has a very well-established co-located MU and it works really well and it continues to. So the MU is the default place to go for everybody and that’s where they go when they come in the door and then they’re only sent to the higher-risk unit if there’s a reason, so everybody is in the MU.

DInt9

I think whether you set it up as opt in or opt out makes a big difference. So if you’re low risk you are automatically booked at the birth centre, unless you explicitly opt out. That probably encourages more people to use it because if you present to them this is the appropriate normal place for you to go for your birth because you are low risk, it’s more likely that they’ll agree with it, rather than saying which would you rather have?

FInt2

An opt-out policy would help with solving another recurrent issue, especially for AMUs in the case study sites: the frequency of MU midwives being moved regularly during shifts to OUs to care for women there. Midwives complained about this a lot:

I think it’s dangerous once we start seeing staffing pull from the birth centres to deal with problems in the obstetric unit . . . can be unsettling for the junior midwives coming in to [FMU] work and then being asked to go. And it limits their experience here, so they’ve not got the exposure to the lovely normal births, so they just see the birth centre and somewhere where they get moved from to go and cover the shifts at the OU.

AInt2

There’s been a lot of staffing issues with [AMU], so they were constantly pulling off midwives and shutting it.

BInt12

They don’t like coming out to help on delivery suite, and they probably feel like if they need help they don’t get it as easily as they’re dragged out to help on delivery suite. So that can cause a bit of friction.

DInt2

Only a few sites saw addressing this as a priority for optimising the use of MUs:

I think now it’s been recognised that we’ve worked the wrong way round actually. You need to staff your alongside units, keep the capacity off the labour ward, and let midwives deal with the normal stuff.

BInt4

Having core staff who could not be moved was one way of keeping MUs open:

I would like birth centre to have a proper core of midwives. The staffing are separate from the labour ward, postnatal, so they, obviously they’re still there, but they’re not counted in the numbers. Because birth centre has its own numbers. So midwives can go there to work and not have this fear of being pulled.

EMFG2

Only one site had an integrated staffing pattern with community midwives working shifts in both the FMU and the AMU. This clearly contributed a sense of ownership and pride in the facility and solidarity with the women:

Because midwives like to see the women, you know, not just when they’re coming in in labour, and it helps to give that approach. Although, you know, we can’t do a named midwife as such, but then that helps us so that women fundamentally will only have like the maternity services, a group of four to six women that they know. So in the main the women at the [FMU] birth centre will have met the woman.

AInt5

Reflecting and evaluating

Respondents from one site in particular spoke about the need to audit and reflect on how their MUs were working:

The important thing is to keep auditing their activity making sure that you know, lessons are learnt and people can sort of, you know help their practice get better . . . you know, get improved for the better really.

AInt7

And I think in establishing it I think that’s important, but then also maintaining that. There’s always a risk that things start dwindling if you don’t constantly review and constantly make sure that those underpinning values are still there.

AInt3

This site had the largest FMU in England and one of the highest utilisation rate regarding their MUs.

Keeping midwifery-led units

The importance of keeping MUs open, especially FMUs, by using them for a range of activities and recognising them as community/family resource/early year’s assets was considered helpful. Some of the suggested activities that could be provided in FMUs include antenatal clinics, childbirth education classes, outreach high-risk antenatal clinics, complementary therapies and smoking cessation support.

Frame midwifery-led units as a real option

It was considered important to frame MUs as a real option. Encouraging women to identify and retain MUs as an option in their birth planning, rather than the majority contemporary approach of requiring that women make binding decisions about birthplace early in pregnancy. There was support for presenting MUs as the default option for healthy women without pregnancy complications and for whom a straightforward labour and birth can be anticipated. This was suggested as an approach that would engender confidence in MUs as a place of birth. This would also require that women are provided with information about different options to counter fears about out-of-hospital birth that are prevalent.

Influential individuals

The importance of influential individuals believing in MUs who should be active in their support was identified. Such individuals included obstetricians and senior midwifery managers; the latter should be in positions in which they influence service provision and commissioning decisions (e.g. trust and commissioning boards).

Risk culture

The prevalent risk culture that dominates NHS maternity services was considered a significant negative factor for MUs.

Use of financial information

There were particular concerns around the understanding and use of financial information related to MUs, with perceptions that finance directors required access to better data and improved understanding of the evidence base.

Continuity of midwifery carer

The relative dearth of NHS trusts where care in a MU could be accompanied by access to continuity of a midwifery carer was of particular concern. Other aspects of the organisation of midwifery staffing also perceived as threats related to the default option of supporting OUs by drawing on midwives from AMUs and FMUs.

General comments

General comments included the need for ways to change the discussion around MUs. This encompassed moving from discussion of MUs as a ‘nice opportunity’ to a conversation related to safety.

The provision of information both in a range of ways and at different times in women’s pregnancies was considered important to meet the needs of different families. The need to broaden conversations to consider care provided prior to transfer, rather than simply the birth and increased use of the internet for local maternity systems to demonstrate what their ‘offer’ is to women, were all ideas that should also be considered. The imperative to provide accurate information was considered the responsibility of a wide range of health-care staff, including GPs, receptionists and community midwives. Increased use of MUs again featured, ensuring appropriate senior trust staff, rather than OU labour wards midwives, made decisions on closing units on a short-term basis.

Midwifery leaders were charged with the importance of collecting robust data and utilising it appropriately, including publishing annual reports for their MUs.

Boundary issues featured as current obstacles to women’s ability to access a MU. Ways of sharing access to notes through common care portals and enabling women to access options across trust boundaries were suggested.

Interventions that could be actioned in the short term

An intervention to ensure that women have access to high-quality, evidence-based information to inform their decision-making about place of birth, which may require support by existing or new tools that should be available in different languages or with the provision of language support. Information and discussion of place of birth should not be a single event, but something that is revisited several times during pregnancy. Women should not be required to make binding decisions on place of birth in early pregnancy.

Use of an ‘opt-out’ model for MU use should be considered, when this is not already in place.

Midwifery-led units need leadership and management by midwives, without competing responsibilities. Consultant midwives appear ideally placed to lead and champion MUs, but may need support in achieving this.

Linkage with the Midwifery Unit Network (established in 2016, but probably underutilised) may be helpful for individual organisations to benefit from shared good practice and support with developing their business case.

Midwifery-led units require a marketing campaign, including local and social media, with endorsement of their value by national NHS leaders/local members of parliament/similar.

Maintaining a MU requires collection of robust data, including outcomes relevant to MU care and approaches. Consideration could be given to developing a MU ‘kite mark’ award and for AMU and FMU data to be included in dashboards.

Midwives working in, or rotating to, MUs require access to a continuing education programme that both supports the maintenance of skills related to the care of women with straightforward maternities, but also in maintaining skills if emergencies occur.

Consideration should be given to how continuity of midwifery care can be provided.

Language and terminology should reflect a focus on normality and positive choices and outcomes.

Free-standing midwifery units should be used for a wider range of activity in addition to intrapartum care. This may include consideration of use by other agencies in order that they are identified as a community resource, open continuously or for longer periods of time, and thus perceived as less costly to maintain.

Interventions that could be actioned in the medium term

Several of the interventions suggested for the medium-term action echoed those identified for short-term action. This may have been an artefact caused by the nature of parallel group work or because they were seen as requiring continuing effort. Publicity about MUs is one example and developing educational provision for both qualified midwives and those in training another. Leadership and management, data collection, staffing (continuity and rotational) models and ensuring FMUs were seen as a community hub were all identified as interventions to be applied in the medium term.

Interventions that first appeared in the medium-term grouping comprised those with a wider political perspective and achieving influence (e.g. engaging politicians as allies through the secretary of state/prime minister’s office or engaging influential women’s groups). Engaging local maternity systems as influencers, ensuring senior midwives were appropriately positioned on boards, and involving service users on appointment panels for midwives and obstetricians were all identified. An additional medium-term intervention concerned offering place of birth discussions for groups rather than individual women.

Mentioned for the first time was the need to consider the 32 units where there was currently no MU available to women, although this lacked specific actions.

Interventions that could be actioned in the long term

Similar themes emerged to those reported previously, but with an emphasis on actions that could be embedded to achieve sustained change. It was acknowledged that this would not be without challenge. Here, interventions were linked with the contemporary maternity policy priorities of embedding personalised care planning and integrating MUs fully into maternity transformation plans. Models of midwifery working were again considered important through scaling up continuity approaches and there was emphasis on preparing student midwives during training for work in those models on registration. Cultural and systems issues, again, featured with changes to the risk and blame culture in maternity, advocated together with financial and contracting changes required, including tariff.

Local

• Ensure leadership within trusts, named champions and designated time for those managing MUs. Consideration should be given to the involvement of consultant midwives in leadership of MUs where this system does not currently exist.

• Heads and directors of midwifery should be enabled to operate at a strategic level in which they can inform and champion MUs. Such locations include trust boards, local maternity system boards and commissioning groups.

• Midwifery managers responsible for MUs should be at an equivalent level in their organisation to those responsible for OU birth facilities and free of competing work.

• In each trust, an obstetrician should act as champion for MU.

• MUs require fresh campaigns to raise awareness among various groups: commissioners, health professionals and the local community. This may include a range of local and social media. Campaigns to promote use may benefit from involving those who have used and wish to be ambassadors for these services. Consideration should be given to the involvement of key women’s groups in the local community and in engaging national maternity policy-makers or politicians.

• To enable women to have a realistic choice of accessing MUs, barriers that need to be removed include the absence of information about MUs as an option. This information should be available to all women and include existing and, if required, new tools to present women with information about all of the options for birthplace in language that moves beyond the current emphasis on risk.

• Discussions about birthplace should not be available ‘once only’ in early pregnancy and subsequently binding. They should be revisited throughout pregnancy, especially at 36 weeks, as part of a birth options discussion and again at onset of labour. 114 Encouraging women to retain a number of options should be encouraged, and discussions and language should follow an approach in which normality is the default expectation for the majority of childbearing women.

• Getting potential service users across the threshold was perceived as key in enabling women to start to consider MUs as an option. All of the above should be embedded in a midwife-led care pathway that is standardised across the entire service and addresses criteria for referral from midwifery to obstetric services and the process of referral.

• The current predominant model is that birthing in MUs is available as an opt in rather than opt out. This should be changed so that offering AMUs, where available, should become the default for women with straightforward maternity and should be explored by trusts as the default in settings with FMUs; impacts on utilisation should subsequently be evaluated.

• The ongoing collection of robust data related to MU use, women’s experiences, as well as outcomes, is a priority, as is the regular publication of these. Consideration should be given to framing outcomes in the context of supporting women’s perinatal mental health.

• In the context of MU utilisation, reconsideration of models of midwifery working. There are two main components to this. First, the need to reduce diminishing MU midwifery resources by moving midwives from MUs to OUs in times of high workload, as this may be followed by closure and threaten unit viability. A key to addressing this is the embedding of the MU as equal and necessary as the OU so that women are cared for in the appropriate setting. One suggestion was that in OUs, once per shift, a review is done of the risk status of women to see if any can be transferred to the AMU. Second, the need for women accessing MUs also to benefit from continuity models.

• Reconsideration is required for developing and maintaining skill sets. This includes increasing attention to the skills required to support normality in birthing, including consideration of an advanced competency set that focuses on supporting normal childbirth and thereby increasing midwives’ confidence for normality-focused practice settings. However, maintaining skills for practice when complications develop or when, for example, babies require more assistance at birth is also essential and it may be beneficial to increase multidisciplinary education to support this. Random emergency skill drills in FMUs was highly valued by staff in one case study site and is worthy of consideration by other FMUs.

• Finance managers should be provided with detailed information about the evidence base and wider health economic implications related to utilisation of MU.

• Women’s choice should not be constrained by geographical or organisational boundaries. Systems that support appropriate access to maternity records across geographical areas require exploration. The development of digital records as recommended in Implementing Better Births: A Resource Pack for Local Maternity Systems115 is a priority.

• Within organisations, attention must be given to avoiding divisions between staff groups and encouraging respect for different roles. Maintaining the sense of coherence of a wider service is essential, together with the acknowledgement made by each part of a service. The provision of multidisciplinary education and training may facilitate this.

• Increasing the use of FMUs to include activities beyond intrapartum care has potential to result in FMUs being open for longer periods or all of the time. This may counter both perceptions that FMUs are closed most of the time and some women’s concerns about whether or not the unit will be available to them when their labour starts. Utilisation of FMUs for other activities and as the community hubs, envisaged in Implementing Better Births: A Resource Pack for Local Maternity Systems115 may counter perceptions of elite and limited services and, depending on services using them, has potential to increase income into the maternity directorate.

Commissioning

• Commissioners may benefit from a clear and detailed briefing on the evidence base to MUs and how these can support their efforts to ensure high-quality, accessible services for all women. All Clinical Commissioning Groups (CCGs) should appoint a clinical lead for maternity and, as a minimum, ensure that the focus is specifically detailed in the job description of a CCG clinical lead.

• The option to co-commission services such as FMUs should be explored across geographical and commissioning boundaries, with the development of clear referral pathways. This would facilitate collaboration and provide potential efficiencies.

• The potential for local maternity systems to provide leadership and support for development and utilisation of MUs should be explored.

Midwifery education

• The importance of continuing education for the qualified midwifery workforce was addressed above.

• Preparation of future practitioners needs to include work in continuity models, with consideration of the numbers of births that should be achieve in MU, or similar, settings. Preceptorship following qualification may need review to incorporate support in developing confidence in birth in MU settings.

General

• The Midwifery Unit Network is a useful resource and source of support for individuals and organisations, enabling sharing of good practice and preparation of business cases.

• The use of social marketing campaigns to increase awareness of MUs should be considered.

The subsequent chapter considers all findings in the context of contemporary policy and extant evidence.