Implementing caseload midwifery: Exploring the views of maternity managers in Australia – A national cross-sectional survey

G Model

WOMBI-484; No. of Pages 9 Women and Birth xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Women and Birth journal homepage: www.elsevier.com/locate/wombi

ORIGINAL RESEARCH – QUANTITATIVE

Implementing caseload midwifery: Exploring the views of maternity managers in Australia – A national cross-sectional survey Kate Dawson a,b,*, Helen McLachlan a,b, Michelle Newton a,b, Della Forster b,c a

School of Nursing and Midwifery La Trobe University, Bundoora 3086, Australia Judith Lumley Centre, La Trobe University, 215 Franklin St, Melbourne 3000, Australia c The Royal Women’s Hospital, 20 Flemington Road, Parkville 3052, Australia b

A R T I C L E I N F O

Article history: Received 24 August 2015 Received in revised form 22 October 2015 Accepted 22 October 2015 Keywords: Caseload midwifery Sustainability Maternity workforce Continuity of care model Australian maternity service

A B S T R A C T

Background: The benefits of caseload midwifery care are clearly documented, and many policy documents in Australia support its expansion. Despite this, little is known about the availability of caseload across Australia, nor about what proportion of women have access to a caseload model. This paper describes caseload midwifery in the public maternity system in Australia; its prevalence, and factors associated with implementation and sustainability. Methods: A cross-sectional online survey of maternity managers of public hospitals that provide birthing services throughout Australia. Findings: Sixty-three percent (149/235) of eligible participants responded. Respondents were from all states and territories, metropolitan, regional and remote areas, and from hospitals with very small to very large birth numbers. Only 31% reported that their hospital offers caseload midwifery, and an estimated eight percent of women received caseload care at the time of the survey, most of whom were considered to be of ‘low obstetric risk’. Many respondents were planning to implement or expand caseload. Key factors associated with the implementation of caseload were funding to establish the model, the interest and availability of staff to work in the model, organisational support and perceived consumer demand. Conclusion: This is the first study to explore caseload implementation at a national level. Although the number of services offering caseload midwifery care has increased nationally, access remains relatively limited. Women who live in metropolitan areas and who are considered at ‘low obstetric risk’ are most likely to be able to access this model. Funding and support for establishing new models are the main barriers to implementation. ß 2015 Australian College of Midwives. Published by Elsevier Australia (a division of Reed International Books Australia Pty Ltd). All rights reserved.

Summary of relevance:

What this paper adds

Problem Little is known about the availability of caseload across Australia and the enablers and barriers to its further introduction and expansion.

This paper provides a snapshot of caseload midwifery in the public maternity system in Australia, including its prevalence, factors associated with its implementation and sustainability, and potential factors that enable or hinder implementation of the caseload model.

What is already known The many benefits of providing caseload midwifery care are clearly documented, and many policy documents in Australia support its expansion. * Corresponding author at: Judith Lumley Centre, La Trobe University, 215 Franklin St, Melbourne 3000, Australia. Tel.: +61 03 9479 2203. E-mail address: [email protected] (K. Dawson).

1. Background Maternity care policies in Australia recommend providing women with continuity of care for pregnancy and birth.1–5 These policies reflect the large body of evidence that has demonstrated

http://dx.doi.org/10.1016/j.wombi.2015.10.010

1871-5192/ß 2015 Australian College of Midwives. Published by Elsevier Australia (a division of Reed International Books Australia Pty Ltd). All rights reserved.

Please cite this article in press as: Dawson K, et al. Implementing caseload midwifery: Exploring the views of maternity managers in Australia – A national cross-sectional survey. Women Birth (2015), http://dx.doi.org/10.1016/j.wombi.2015.10.010

G Model

WOMBI-484; No. of Pages 9 2

K. Dawson et al. / Women and Birth xxx (2015) xxx–xxx

that women who receive continuity of care have fewer childbirth interventions (e.g. caesarean section), increased satisfaction with care,6,7 and in the context of caseload midwifery,7,8 no evidence of adverse outcomes associated with providing caseload care to women, even among women of mixed obstetric risk.8 Continuity of care incorporates models such as team midwifery and caseload midwifery. Caseload midwifery care (also known as Midwifery group practice (MGP), Know your midwife (KYM) and one to one midwifery) aims to provide women with care from a known midwife throughout pregnancy, labour, birth and into the postnatal period. Given the evidence of improved outcomes for women, it is also important to consider staff views of this model of care. The literature discusses issues associated with caseload work including, burnout9,10 and work life balance.9,11–15 Conversely a recent Australian study found that midwives working in continuity of care models may benefit from caseload, with increased professional satisfaction and lower burnout scores when compared to their non-caseload colleagues.16 Despite the evidence of the benefits of continuity of care, access to this model of care is still limited.5 Potential benefits of the caseload model are not limited only to improved clinical outcomes; it has also been suggested that caseload midwifery could assist in keeping smaller maternity services open in the rural and regional areas and possibly enable some maternity services that have closed to re-open.17,18 However, there is limited information on the availability of caseload midwifery across Australia, and no studies were identified that have explored, at a national level, issues related to sustainability and potential expansion of the model. A number of Australian reports describe factors that have contributed to the successful introduction of caseload programs.19–21 Factors identified in relation to the sustainability of caseload include, engagement and support at all levels within the hospital/service; strong support from the community; key stakeholder engagement and support, including ‘champions’ who will drive implementation; a belief in woman-centred midwife-led care; support for midwives to be able to sustain autonomy and flexibility including occupational and personal support; clear boundaries within the model; adequate cover for extended leave; adequate remuneration; and a clear role for managers within this new way of working.19–21 Given the lack of national data, we conducted a study (ECO – Exploring CaselOad midwifery in Australia) which explored the introduction, expansion and sustainability of caseload in Australia. The views of maternity managers, midwives, and midwifery students have been sought, and aspects explored included enablers and barriers to the implementation, expansion and sustainability of caseload midwifery, as well as an exploration of existing caseload models across the country, and how the models are configured. This paper presents data from one component of the ECO study; the survey of maternity managers. It aims to provide a snapshot of caseload midwifery in the public maternity system in Australia (where two thirds of maternity care is provided22), including its prevalence, factors associated with its implementation and sustainability, and potential factors that enable or hinder implementation of the caseload model. 2. Methods This study used a cross-sectional survey design.

which provides information on public hospital services throughout Australia, and lists the number of admissions for childbirth at each hospital. ‘My Hospitals’ was searched in early March 2012 to determine which hospitals had births in 2011. Public hospitals which provided ‘planned’ birthing care to women were included, thus, hospitals with a maternity service but without a birthing service were excluded from the study, with the exception of one hospital that was reintroducing a birthing service through introduction of a caseload model. For hospitals with a low number of births (less than 50 as per the ‘My Hospitals’ site), phone contact was made to determine if there was actually a birthing service at the hospital, as very small birth numbers could have been the result of births occurring enroute to another hospital where the birth had been ‘planned’. All eligible hospitals were contacted by phone to obtain the email details of the maternity manager, in order to invite them to participate. 2.2. Data collection tools An online survey tool was developed specifically for the study, informed by the data tools and findings of a previous study of midwives’ experiences of caseload.24 The survey explored the characteristics of the hospital, existing models of midwifery-led care, respondents’ views and intentions regarding caseload, and where caseload models already existed, the operation, structure and functioning of the models. The survey contained open- and closed-ended questions as well as Likert-type scales where respondents were required to select from a five-point response scale. Response options used were: ‘Strongly agree’, ‘Agree’, ‘Neither agree nor disagree’, ‘Disagree’, ‘Strongly disagree’. The survey was designed to have embedded skips, enabling the completion of the survey to be responsive to certain questions about caseload, thus enabling appropriate questions to hospitals with or without that model. Four rounds of piloting of the survey were undertaken with researchers, midwifery academics and midwifery managers within the research team’s professional network. Changes were made following each round of piloting, then the survey finalised. The survey was distributed by an email with an embedded link to Survey Monkey.25 The email was sent to maternity managers of the eligible public maternity hospitals between February and April 2013. Reminders were sent by email at two and four weeks following the initial invitation to participate. Return of the survey was considered consent to participate in the study. 2.3. Data management and analysis Data were downloaded from Survey Monkey25 into an Excel spreadsheet26 and then transferred into STATA version 11.27 Data cleaning included range and logic checks, and where possible inaccuracies that were identified were corrected. Descriptive analysis was undertaken and frequencies and proportions presented. Open-ended questions were analysed using content analysis.28 The responses were coded then collapsed into categories and then into themes. Agreement on coding, categories and themes were undertaken by two of the authors. Ethics approval was granted by the institutional ethics committees, in September 2012.

2.1. Participants

3. Results

Public maternity hospitals in Australia were identified using the ‘My Hospitals’ website,23 an Australian Government website

An overview of results is presented first, then the findings discussed in three sections; hospitals not intending to set up a

Please cite this article in press as: Dawson K, et al. Implementing caseload midwifery: Exploring the views of maternity managers in Australia – A national cross-sectional survey. Women Birth (2015), http://dx.doi.org/10.1016/j.wombi.2015.10.010

G Model

WOMBI-484; No. of Pages 9 K. Dawson et al. / Women and Birth xxx (2015) xxx–xxx

caseload model, hospitals planning to set up a caseload model in the future and hospitals with a caseload model at the time of the survey.

Table 1 Characteristics of responding hospitals (n = 149).

3.1. Availability of caseload models in Australia Of the 331 public hospitals in Australia that were identified as having had births in 2011, 79 hospitals did not have a birthing service, leaving 252 hospitals eligible to participate. Following distribution of the invitation to participate in the study a further 17 hospitals were excluded; 13 because they no longer had a birthing service, and four were campuses of larger hospitals. This left a total of 235 eligible hospitals invited to participate (Fig. 1). The response rate was 63% (149/235). Respondent hospitals were from all states and territories (including approximately half of all eligible hospitals in each state and territory invited to participate), every geographical region, and represented a variety of sizes throughout Australia (as represented by birth numbers). Over half were located in regional areas (55%), 29% in remote or very remote areas and 17% in major cities. Hospitals with less than 100 births represented 20% of the sample, 38% were hospitals with 101–500 births, 14% with 501–1000 births, 11% with 1001–2000 births, and 17% with annual births greater than 2000 (Table 1). We asked all respondents about the midwife-led models of care they offered within their hospital. Over half provided midwife-led antenatal care, and caseload midwifery was the next most common midwife-led model reported, with approximately one third (31%) of the responding hospitals offering this model of care (Table 1). In addition to hospital-based care services, antenatal care was provided in a range of settings outside the hospital (n = 149). These included general practitioner (GP) clinics (81%), obstetricians’ rooms (36%), Aboriginal health services (35%), community midwife clinics (including remote and outreach clinics) (31%), other hospitals (15%), and in women’s homes (14%). Forty-three (31%) of the responding hospitals had a caseload model at the time of the survey (Fig. 2). The proportion with a caseload model differed by hospital location and size; 63% of

3

Responding hospitals

Hospitals with a caseload model

n

%

n

%

State/Territory (n = 135) ACT NSW NT QLD SA TAS VIC WA

1 41 3 22 17 2 36 13

0.7 30.4 2.2 16.3 12.6 1.5 26.7 9.6

0/1 9/41 2/3 9/22 6/17 1/2 15/36 1/13

0 22.0 66.7 40.9 35.3 50.0 41.7 7.7

Hospital location (n = 144) Major city Inner regional Outer regional Remote Very remote

23 31 48 39 3

16.6 20.1 34.5 26.6 2.2

14/23 10/31 12/48 7/39 1/3

60.9 32.3 25.0 17.9 33.3

Births in 2012 (n = 142) 1–100 101–500 501–1000 1001–2000 2001–9000

28 54 20 16 24

19.7 38.0 14.1 11.3 16.9

8/28 9/54 5/20 7/16 15/24

28.6 16.7 25.0 43.8 62.5

Existing midwifery led maternity models of care (n = 143) Midwifery-led antenatal care 79 55.2 – Caseload 44 30.8 – Team midwifery 41 28.7 – Birth centre (within a hospital) 11 7.7 – Public home birth 5.6 – 8 (within a caseload model) 3.5 – Public home birth 5 (not within caseload) 2 1.4 – Free standing birth centre

– – – – – – –

hospitals with large birth numbers (2001–9000) and 61% of hospitals in a major city had a caseload model, compared with just under one third (28%) of hospitals with low birth numbers, and 18%-33% of hospitals in areas other than a major city (Table 1). Twenty-three percent of those without a caseload model were in the process of implementing the model (85% of these within the next 12 months) and 36% of those without a model were considering it for the future (Fig. 2). 3.2. Hospitals NOT planning to implement a caseload model Of those hospitals without a caseload model, over one third of respondents (36/96, 38%) indicated that they were not considering implementing a caseload model. Respondents were asked the reasons for their response and 32 responses were received (89%). The vast majority (31/32) of those not planning on implementing caseload were from a regional or remote area, and 78% (25/32) had less than 500 births in 2011. The main theme related to staffing issues, including a lack of medical and midwifery staff support, insufficient midwifery staff availability and absence of funding for the model. Half of the respondents (16/32, 50%) indicated that staffing the model was an issue. ‘‘. . . we have a chronic shortage of midwives’’ (id 50, outer regional) ‘‘Midwifery resources currently stretched. . .’’ (id 124, outer regional)

Fig. 1. Recruitment of ‘participants’.

‘‘Lack of midwives’’ (id 26, remote)

Please cite this article in press as: Dawson K, et al. Implementing caseload midwifery: Exploring the views of maternity managers in Australia – A national cross-sectional survey. Women Birth (2015), http://dx.doi.org/10.1016/j.wombi.2015.10.010

G Model

WOMBI-484; No. of Pages 9 4

K. Dawson et al. / Women and Birth xxx (2015) xxx–xxx

Fig. 2. Caseload availability (n = 139).

Others raised the issue that the midwifery workforce was needed to staff other areas of the hospital (particularly noted in the remote setting, with 10/15 hospitals from a remote area commenting on this issue). ‘‘Midwives work in combined medical, surgical and midwifery ward so difficult to work around this component’’ (id 113, unknown area) ‘‘The current midwives also participate in the general roster. . .’’ (id 117, outer regional) ‘‘Midwife is required to take a general patient load as well as her maternity load’’ (id 86, remote) Almost half (15/32, 47%) indicated that midwives were not interested in working in this way. ‘‘No interest from midwives; midwives here prefer shift work and are not prepared to have their work life balance interrupted. . .’’ (id 22, outer regional) ‘‘Midwives working here support caseload models however in general are not prepared to work in a caseload model due to the call requirements’’ (id 82, outer regional)

3.3. Hospitals CONSIDERING implementing a caseload model in the future At the time of the survey, 35 organisations (36% of those without caseload at the time of the survey) were considering implementing caseload in the future. We asked respondents to comment on the reasons influencing their decision to implement caseload, and 33 responded. Similar to the themes from organisations not intending to implement a caseload model, support for the model, staffing issues, and funding to support implementation were all identified as factors influencing their decision. Resources and funding for the model were considerations for 45% (15/33) of respondents. ‘‘Funding is the biggest issue, if funding was available caseload could be implemented’’ (id 79, major city) ‘‘Financial support’’ (id 149, outer regional) ‘‘Financial support for a project person’’ (id 78, outer regional) A number of respondents (n = 15/33, 45%) also discussed issues relating to staffing a caseload model. ‘‘Retain[ing] and recruitment of rural midwives to ensure a viable service delivery. . .’’ (id 144, inner regional)

‘‘Staff were not interested, as they are all part time staff’’ (id 60, remote)

‘‘Midwives interested in working in that model’’ (id 73, outer regional)

A further one quarter (8/32, 25%) cited lack of medical staff support or availability, with a medical model of care currently in operation at the hospital.

‘‘. . ..enough midwives to ensure adequate time off’’ (id 12, outer regional)

‘‘The hospital is run on a VMO model and all women are cared for by GP obstetrician within the community’’ (id 52, major city) ‘‘Most antenatal care given by GP in private practice or obstetrician in private practice, there would be objection for caseload midwifery as this would affect their care of the patient’’ (id 124 outer regional) Nearly one quarter (22%) cited cost and funding as an obstacle to introduction of the model. ‘‘Cost factor to staff both caseload and on ward staff is prohibitive’’ (id 64, outer regional) ‘‘We do not have the resources to support [this model]. . .. Resourcing would be required to implement a caseload model’’ (id 117, outer regional)

A further 24% (8/33) reflected on the medical support required to implement the model. ‘‘VMO’s would have to be supportive’’ (id 29, outer regional) ‘‘This model would require strong support by the local GP Obstetrician and Specialist Obstetrician. . .’’ (id 126, outer regional) A small number of respondents also discussed the support needed from hospital executive (n = 3/33), while 15% (5/33) cited consumer support as another consideration. 3.4. Barriers and facilitators to setting up caseload midwifery In order to explore similarities and differences between the three groups that did not have a caseload model at the time of the survey, a series of statements regarding perceptions about

Please cite this article in press as: Dawson K, et al. Implementing caseload midwifery: Exploring the views of maternity managers in Australia – A national cross-sectional survey. Women Birth (2015), http://dx.doi.org/10.1016/j.wombi.2015.10.010

G Model

WOMBI-484; No. of Pages 9 K. Dawson et al. / Women and Birth xxx (2015) xxx–xxx

5

Do not have a caseload model planned Consider caseload in future Planning to implement caseload 24.3 Medical staff would be supporve of caseload

41.2 81.8 37.8

Midwives in hospital supporve of caseload

81.8 95.5 48.6

Midwives intesterested to work in caseload

88.2 95.5 83.8

Need an inial block of funding to establish a model

94.1 77.3 33.3

Believe there is lile consumer demand for caseload

15.2 4.5

Fig. 3. Requirements for implementation of caseload (percentage ‘Agreed’ or ‘Strongly agreed’).

implementation was presented. While all groups indicated that there is consumer demand for the model, this was stronger in the group planning to implement the model (Fig. 3). Agreement was strong in all groups in regard to needing funding to initiate a caseload model. Midwifery and medical support for caseload was perceived to be strongest in the group planning to implement the model, and least in the group not planning a model; and midwifery interest in working in a caseload model was perceived to be weaker in the hospitals not planning to implement a caseload model. 3.5. Hospitals with a caseload midwifery model Hospitals that already had an established caseload model (n = 44) were asked a series of questions about the model establishment and operation. The first model was established in 1995, and the number of models increased to 44 by 2013 (31% of all respondents), indicating a substantial increase in caseload model availability in Australia over time (Fig. 4). Antenatal care was provided in a range of settings within the caseload model. The vast majority offered antenatal care within the hospital (90%). Other options included community-based clinics (56%); antenatal care at home (more than one visit (44%) or one visit only (12%)); and care within a GP clinic (5%).

3.5.1. Eligibility Respondents were asked to indicate what eligibility criteria they applied to their caseload model (Table 2). The majority of hospitals accepted women considered to be at ‘low obstetric risk’ at the time of booking (81%), and one third offered caseload midwifery care to women in an all risk model. Forty percent accepted women aiming for a vaginal birth after caesarean section. Some caseload models targeted specific groups of women, such as women of Aboriginal and/or Torres Strait Islander background (37%) and young women (30%). By calculating the reported birth numbers and the percentage of women receiving caseload at each of the responding hospitals, we estimated that the overall percentage of women receiving caseload among responding hospitals across Australia at the time of the survey was eight percent. 3.5.2. Midwifery workforce We asked how many midwives were employed in caseload in terms of numbers and FTE (full time equivalent). Respondents

45 40

Number of hospitals

Thirty-six of the 44 hospitals indicated that ‘seed’ funding was received to set up the caseload model within their hospital. This funding came from a variety of sources and special initiatives (such as the ‘Rural Maternity Initiative’ (n = 10), Closing the Gap, Towards Normal Birth, Chronic Disease Prevention, New Directions), one-off grant funding (e.g. Department of Health funding, funds allocated by Regional Board), and in one case from funds transferred from a community health budget.

35

Table 2 Eligibility criteria for women for caseload midwifery models (n = 43).

30

Criteriaa

25

‘‘Risk’’ level accepted for caseload Accept women low risk and follow through even if risk level changes Accept low risk and transfer out if risk level changes Accept women aiming for a VBAC All risk model

20 15 10

n

%

21

49

19 17 14

44 40 33

16 13 6 5 2 2

37 30 14 12 5 5

5 2013

2012

2011

2010

2009

2007

* No new models established between 1998-2004.

2008

2006

2005

*2004

*2003

*2001

*2002

*2000

*1999

*1998

1995

1997

0

Annual total (new models) Cumulave total

Fig. 4. Year caseload models established.

Women targeted for caseload care Aboriginal &/or Torres Strait Islander Young mums Women with alcohol or drug issues Minority ethnic groups Refugee women Women with high body mass index a

Respondents could tick more than one response.

Please cite this article in press as: Dawson K, et al. Implementing caseload midwifery: Exploring the views of maternity managers in Australia – A national cross-sectional survey. Women Birth (2015), http://dx.doi.org/10.1016/j.wombi.2015.10.010

G Model

WOMBI-484; No. of Pages 9 6

K. Dawson et al. / Women and Birth xxx (2015) xxx–xxx

indicated that a total of 311 midwives were working in caseload at the time of the survey, and this was calculated as approximately 276 full time equivalent midwives based on the total number of midwives working in caseload and the FTE of midwives currently working in caseload for each hospital. There was a large variety of employment classifications for midwives working in caseload for each state and territory; a total of 21 different classifications were reported for midwives working in caseload models. This is due in part to the different classifications systems in each jurisdiction, but demonstrates the diversity even within states (Fig. 5). Of the 42 hospitals responding to questions regarding the method of remuneration for midwives working in caseload, 36 (86%) paid caseload midwives an annualised salary (i.e. they were paid a certain percentage ‘loading’ above the base rate for their classification each fortnight in an attempt to capture the varied work patterns of a caseload midwife), and the remaining six respondents paid midwives as per the local industrial award (i.e. they were paid for the hours and type of work performed in accordance with the midwives’ contract and enterprise/industrial agreement in place). Of the 36 hospitals that paid an annualised salary, 30 reported the loading that they paid; 12 (40%) paid a loading of 29% above base rate, seven paid 30%, seven paid 35%, two hospitals 27%, one hospital 28% and one paid 32%. It should be noted that in South Australia and Queensland, a loading is stipulated in the state-wide industrial award for midwives working in caseload, and is set at 35%. In states such as Victoria, a local agreement is made with individual hospitals and no statewide agreement exists. The majority who paid midwives as per the enterprise agreement/industrial award (n = 6) were in a small hospital (<100 births) (4/6) and/or in a remote area (3/6) or outer regional area (2/6). Managers were asked about current staffing vacancies in the caseload model. The majority (71%) indicated that they currently had no vacancies in their model, although a few of these also noted that staff were needed to provide relief in times of extended sick leave, annual leave, and to fill some on call periods. Of the responding hospitals with a caseload model there were 39/311 (13%) reported resignations from midwives working in caseload over the previous 12 months. The majority of those leaving caseload work remained working in midwifery; 12 midwives returned to shift work, two to ‘9 to 50 work, nine to other midwifery work, and six no longer worked in midwifery. When managers were asked if there were ‘usually’ enough suitable applicants for vacant positions, 51% either ‘Agreed’ or ‘Strongly agreed’ with this and 28% ‘Disagreed’ or ‘Strongly disagreed’. 3.5.3. Community demand Respondents were asked if there was enough caseload places to meet community demand, and the majority (28/43 (65%)) noted that more women wanted caseload than there were places available. Three indicated that they had more places available in their caseload models than women wanting the model, 10 indicated that places available in the model met the current demand, and two were unsure. All 43 responding hospitals considered that their caseload model was sustainable into the future; 19 indicated that their caseload models would be staying the same size, while 21 planned on expanding. One hospital was planning to reduce the size of its model. 3.5.4. Advantages and challenges of a caseload model Managers from hospitals with a caseload model were asked in an open-ended question if they had any comments regarding the advantages and challenges of caseload.

Twenty-eight respondents commented on the advantages of caseload. The main themes were improved outcomes and satisfaction for women, and increased job satisfaction for midwives. Benefits to women were the most common theme, and 82% (n = 23) commented to this effect. ‘‘Continuity of care by a known midwife is really important for women and improves birth outcomes’’ (id 5, major city) ‘‘. . ..The birth outcomes are excellent’’ (id 9, outer regional) ‘‘All women should be in a caseload model, our women love it. It’s important to consolidate education and increase [women’s] self-esteem’’ (id 103, major city) ‘‘Increased satisfaction for women and their families’’ (id 65, outer regional) Job satisfaction for midwives also emerged as a theme, with 14/28 (50%) respondents making a comment. ‘‘This is an excellent model for midwives as they work across their scope of practice’’ (id 5, major city) ‘‘Midwives practising autonomously and really fully exploring their roles leads to great job satisfaction’’ (id 74, outer regional) Twenty-nine respondents commented on challenges associated with the caseload model. The overall theme to emerge from the comments was that of ‘acceptance and integration’ of the model, that is, how it was accepted by the whole organisation and how it was integrated into the functioning of the hospital. Comments about working relationships and support were the most common, with 59% (n = 17) of respondents making a comment on this. ‘‘Getting buy-in from executive was tough’’ (id 121, remote) ‘‘Integration with core midwives remain an ‘us and them’ situation with some staff’’ (id 148, major city) ‘‘. . . not handing over, want to stay and wait for the birth’’ (id 10, inner regional) Workforce management reflects issues such as recruitment and retention of staff, burnout, coverage of sick and annual leave, and maintaining a skilled workforce, and was apparent in the comments. These were all seen as challenges in caseload by nearly half of hospitals with a caseload model (48%, n = 14). ‘‘Covering sick leave/annual leave etc. is one of the main challenges’’ (id 42, remote) ‘‘Just starting the model and dedicating time to organising’’ (id 111, major city) ‘‘Recruiting and retaining a skilled workforce. Ensure strategies are in place to reduce the risk of burnout’’ (id 65, outer regional) In summary, although respondents clearly noted benefits to women and to midwives working in caseload care, obstacles remain from an operational perspective in terms of acceptance and integration of the model within the organisations. 4. Discussion This study explored the caseload midwifery model throughout Australia from the maternity managers’ perspective; including its

Please cite this article in press as: Dawson K, et al. Implementing caseload midwifery: Exploring the views of maternity managers in Australia – A national cross-sectional survey. Women Birth (2015), http://dx.doi.org/10.1016/j.wombi.2015.10.010

G Model

WOMBI-484; No. of Pages 9 K. Dawson et al. / Women and Birth xxx (2015) xxx–xxx

7

Fig. 5. Position classifications of midwives working in caseload by state and territory (y-axis = n, x-axis = classification). CMC – Clinical Midwife Consultant; CMS – Clinical Midwife Specialist; AUM – Associate Unit Manager. (This figure aims to highlight the diversity of classifications of the midwives working in caseload, not to be a comparison of actual salaries or levels.)

prevalence, and factors associated with its implementation and sustainability. This is the first study to explore these issues at a national level, and to describe the elements of the caseload models in existence. There is evidence of growth of the model throughout the country, both in terms of hospitals establishing the model for the first time and planned expansion in organisations where caseload already exists. These findings also indicate there is strong consumer demand for the caseload midwifery model, with the majority of hospitals with a caseload model having more women wanting caseload care than there are places available, and community demand evident in sites where caseload was being considered. Key factors regarding the implementation of caseload related to the interest and availability of staff to work in the model, as well as organisational support and consumer demand. These key

themes are congruent with the literature and documents that guide organisations in establishing new caseload models,19–21 where factors such as engagement and support at all levels within the hospital/service, strong support from the community, and a belief in women centred midwife-led care have been identified. While there are many recent state and national policy documents in Australia regarding the promotion of the caseload model,1–5 it is evident that not all services have responded to these policy directions, with 28% of respondents not considering implementing the model. The majority of these was from regional or remote settings and had annual birth numbers less than 500. Barriers to the implementation of caseload in these settings included lack of interest from staff, staff shortages, and hospitals needing midwifery staff to work across both nursing and midwifery practice areas. Many of these sites also indicated that

Please cite this article in press as: Dawson K, et al. Implementing caseload midwifery: Exploring the views of maternity managers in Australia – A national cross-sectional survey. Women Birth (2015), http://dx.doi.org/10.1016/j.wombi.2015.10.010

G Model

WOMBI-484; No. of Pages 9 8

K. Dawson et al. / Women and Birth xxx (2015) xxx–xxx

their maternity care was predominantly medical (i.e. did not have existing midwifery – led care models). A perception of low consumer demand was also discussed by these respondents. Given these factors it would require a significant change to the way maternity care is provided in these settings and significant commitment from management and staff in order to enable a sustainable caseload model to be established. It has been discussed elsewhere that ‘buy-in’ from all levels of the hospital is critical for the sustainability of caseload models, and that each model needs to be adapted to suit the needs of each organisation.17 Given the variety of settings and differences in the organisation of maternity care around Australia, the model may not be sustainable at all sites all the time, however, it has been suggested that caseload could assist in sustaining a rural midwifery service.17,18 Closure of maternity services is an issue that impacts on women and midwives, and it is concerning that over one quarter of all the hospitals identified as having births in 2011 no longer provided a birthing service at the time of this study. It may be that with careful consideration of their identified issues, and with negotiation, caseload could assist in these rural maternity services reestablishing birthing services. Data have also been collected regarding the operational aspects of caseload including work patterns and model logistics, and will be reported elsewhere. In contrast, hospitals that would consider implementing caseload in the future were more focused on elements of support which would enable the model to be established rather than the barriers to implementation. Support came in many forms; the most common was the requirement of financial support that would help in the establishment of the model (e.g. enabling the appointment of a midwife into a project role). These hospitals also identified that both medical and midwifery staff would be supportive of establishing a caseload model, factors that have been reported elsewhere as being key to successful implementation of the model.17,19–21,29 Increased availability of the caseload model is evident in our findings; from 2004 the model has seen significant growth throughout Australia, and this is the case in all states and territories, in urban and regional areas, and in hospitals of various size (NB: there was no respondent from ACT with a caseload model, but we are aware that there is a caseload model in operation). The largest proportion of hospitals with a caseload model were located in metropolitan areas. Further growth of the model is also expected, with 16% of respondents implementing caseload at the time of the survey and 25% considering it in the future. Within hospitals that had an established caseload model, half (49%) were planning on expanding the model within their organisation. It is therefore timely to be investigating factors associated with sustainability, and encouraging to see that all hospitals that currently have a caseload model believed that their model was sustainable into the future. Funding of caseload models was identified by respondents in this study as a challenge, particularly during the establishment phase. Although studies have demonstrated that the cost of running caseload is either equivalent or slightly cheaper than standard care,30,31 the majority of respondents in this study noted that start-up or seed funding would be necessary in order to establish a caseload model. Hospitals with a caseload model reported receiving funding to set up the model from a wide variety of sources; where funds were not sourced externally they were sometimes redirected or realigned from other areas within the organisation, a strategy that was also suggested by Queensland Government20 in their discussion regarding establishment costs of a caseload model. Maternity care funding in Australia is complex, with some components of care being funded by state government and in others federal funding is used, and different arrangements exist even within individual states, in relation to pregnancy care.

This adds to the complexity of introducing a model such as caseload. This is an issue which may require consideration by policy makers to enable further implementation of the caseload model to occur. In Australia women’s access to midwife-led models of care is limited. Only a minority of women are able to access continuity of care models5 and this is particularly so for the most vulnerable women.32 Despite the benefits of caseload and the increasing availability of the model nationwide, only eight percent of women from the responding hospitals had access to caseload care at the time of this survey. Whilst there are models that target socially disadvantaged women, the majority of places in caseload were available to a ‘low risk’ cohort of women. In a qualitative study of caseload in a multi-ethnic community in the United Kingdom, women from minority groups reported improved outcomes such as feeling known by the midwives caring for them; receiving personcentred care; having improved social support; experiencing more flexible care; and feeling more informed and in control when in a caseload model.33 This is also reflected in findings of a study which investigated the introduction of a caseload model for remote dwelling Australian Aboriginal women; women reported a more positive experience of the health system, and staff also seemed to become more culturally aware and sensitive.34 Providing caseload care to women of any risk status has been reported as safe.8 However, a recent Cochrane review suggests that further consideration should be given to the suitability of this model for women with substantial medical or obstetric complications.6 In terms of staffing caseload midwifery, we found that hospitals planning to implement caseload midwifery had midwifery staff available, and there were midwives interested in working in this way. However, in a recent study where midwives’ intentions to work in caseload were explored in hospitals where caseload was already in operation, it was found that although there was support for caseload midwifery, only a small percentage of additional midwives were prepared to work in the model at the time of the study.24 In the current study we found that approximately one quarter of hospitals had vacancies in their caseload models at the time of the survey and nearly one third of these hospitals reported not having enough suitable applicants to fill the vacant positions. However, respondents also noted that they believed working in this way brought midwives job satisfaction. A further longitudinal investigation is warranted to investigate the long term sustainability of working in caseload midwifery in terms of midwives’ outcomes. The advantages and challenges of a caseload model identified by respondents in this study are similar to those reported elsewhere; while improved outcomes for women and increased job satisfaction for midwives are commonly reported,7,16,35 the challenges, such as funding to set up the model, acceptance of and support for the model and buy-in and support from stakeholders (maternity management and medical staff) appears vital, not only for the implementation, but also the sustainability of the model.19–21 Variations on how the model is staffed and operates will add valuable information to this data (to be reported elsewhere). 4.1. Strengths and limitations This is the first study to evaluate the implementation and status of the caseload midwifery model in Australia at a national level. The response rate of 63% is similar to other comparable Australiawide studies of hospital services36 and is higher than the a component of the Australasian Maternity Outcomes Surveillance System (AMOSS) study (53%) which mapped maternity services in Australia.37 In this study, there may have been variable biases and motivations in responding to the survey for those either with or without a caseload model, therefore, although this is a substantial

Please cite this article in press as: Dawson K, et al. Implementing caseload midwifery: Exploring the views of maternity managers in Australia – A national cross-sectional survey. Women Birth (2015), http://dx.doi.org/10.1016/j.wombi.2015.10.010

G Model

WOMBI-484; No. of Pages 9 K. Dawson et al. / Women and Birth xxx (2015) xxx–xxx

national sample, it does not represent all maternity hospitals in Australia. It does however give an overview of the prevalence of caseload in Australia at the time of the survey, and makes a contribution to understanding the barriers and enablers to the future sustainability and expansion of the caseload model in Australia. 5. Conclusion This national cross-sectional survey exploring caseload midwifery found that at the time of the survey, 31% of hospitals offered the caseload care, and an estimated eight percent of women had access to the model. The number of hospitals with caseload midwifery in Australia has increased steadily over the past 10 years. Access is still primarily metropolitan-based and for women considered low risk at the time of booking for maternity care. In view of the benefits caseload provides for women and the strong consumer demand for the model, the potential for increasing the availability of caseload should be further explored, especially in terms of providing equity and access to the model for all women. Although caseload midwifery appears to be well supported, interest and availability of staff to work in the model as well as seed funding were the main barriers reported to its implementation. The issues raised here are important and provide a basis for thinking about these issues at a national level, and the data provide information that can be used at the policy level. Conflict of interest We have no conflicts of interest to declare. Author’s contributions Study design, data collection tools, study implementation, data analysis and manuscript drafting: KD, HMc, MN, DF. Acknowledgments We would like to acknowledge all the maternity managers who made the time to respond to this survey and thanks also to those colleagues that assisted with piloting and survey development. References 1. Department of Health Western Australia. In: Department of Health, editor. Improving maternity services: working together across Western Australia. Western Australia: Department of Health; 2007. 2. NSW Department of Health, editor. Primary Maternity Services in Australia – A framework for implementation. Sydney: Australian Health Ministers’ Advisory Council; 2008. 3. Queensland Government, editor. Midwifery models of care: implementation guide. 2008. 4. Department of Human Services Victoria. Future directions for Victoria’s maternity services. Melbourne: Victorian Department of Human Services; 2004. 5. Bryant R. Improving Maternity Services in Australia. The report of the Maternity Services Review. Canberra: Commonwealth of Australia; 2009. 6. Sandall J, Soltani H, Gates S, Shennan A, Devane D. Midwife-led continuity models versus other models of care for childbearing women. CDS Rev 2013;(8). AN CD004667. 7. McLachlan H, Forster D, Davey MA, Farrell T, Gold L, Biro MA, et al. Effects of continuity of care by a primary midwife (caseload midwifery) on caesarean section rates in women of low obstetric risk: the COSMOS randomised controlled trial. BJOG 2012;119(12):1483–92.

9

8. Tracy S, Hartz D, Tracy M, Allen J, Forti A, Hall B, et al. Caseload midwifery care versus standard maternity care for women of any risk: M@NGO, a randomised controlled trial. Lancet 2013;382(November (23)):1723–32. 9. Stevens T, McCourt C. One-to-one midwifery practice. Part 3: Meaning for midwives. Br J Midwifery 2002;10(2):111–5. 10. Wakelin K, Skinner J. Staying or leaving: a telephone survey of midwives exploring the sustainability of practice as Lead Maternity Careers in one urban region of New Zealand. N Z Coll Midwives 2007;37(October):10–4. 11. McCourt C. Working patterns of caseload midwives: a diary analysis. Br J Midwifery 1998;6(9):580–5. 12. Sandall J, Davies J, Warwick C. Evaluation of the Albany midwifery practice. London: Florence Nightingale School of Nursing and Midwifery, Kings College; 2001. 13. Tracy S, Hartz D. The Quality Review of the Ryde Midwifery Caseload Practice – a primary level maternity service at Ryde Hospital, September 2004–October 2005. Final Report. Northern Sydney and Central Health Coast; 2006. 14. Fereday J, Oster C. Managing a work-life balance: the experiences of midwives working in a group practice setting. Midwifery 2010;26(3):311–8. 15. Collins CT, Fereday J, Pincombe J, Oster C, Turnbull D. An evaluation of the satisfaction of midwives’ working in midwifery group practice. Midwifery 2010;26(4):435–41. 16. Newton M, McLachlan H, Willis K, Forster D. Comparing satisfaction and burnout between caseload and standard care midwives: findings from two cross-sectional surveys conducted in Victoria, Australia. BMC Pregnancy Childbirth 2014;14(426). 17. Brown M, Dietsch E. The feasibility of caseload midwifery in rural Australia: a literature review. Women Birth 2013;26:e1–4. 18. Yates K, Usher K, Kelly J. The dual roles of rural midwives: the potential for role conflict and impact on retention. Collegian 2011;18(3):107–13. 19. NSW Ministry of Health. In: Office TNaM, editor. Midwifery Continuity of Care Model Tool-kit. Sydney, NSW: NSW Ministry of Health; 2012. 20. Queensland Government. In: Office NaM, editor. Delivering continuity of midwifery care to Queensland women. Brisbane: Queensland Health; 2012. 21. Homer C, Brodie P, Leap N. Midwifery continuity of care: a practical guide. Sydney: Elsevier Australia; 2008. 22. Hilder L, Zhichao Z, Parker M, Jahan S, Chambers GM. In: Unit ANPS, editor. Australia’s mothers and babies. Canberra: AIHW; 2012. p. 2014. 23. National Health Performance Authority. My Hospitals. 2012http://www. myhospitals.gov.au/. [accessed March 2012]. 24. Newton M. Implementing caseload midwifery: exploring the views and experiences of midwives and key stakeholders. (Unpublished Thesis) Bundoora, Melbourne: La Trobe University; 2013. 25. SurveyMonkey I. Palo Alto, CA: SurveyMonkey; 2012. 26. Microsoft Corporation Microsoft Office Excel. Microsoft; 2010. 27. Stata Corporation. STATA 11.0 Statistics/Data Analysis. TX, USA: Statacorp LP; 2012. 28. Liamputtong P, editor. Research methods in health. South Melbourne: Oxford University Press; 2010. 29. Scherman S, Smith J, Davidson M. The first year of a midwifery-led model of care in Far North Queensland. Med J Aust 2008;188(2):85. 30. Tracy S, Welsh A, Hall B, Hartz D, Lainchbury A, Bisits A, et al. Caseload midwifery compared with standard or private obstetric care for first time mothers in a public teaching hospital in Australia: a cross sectional study of cost and birth outcomes. BMC Pregnancy Childbirth 2014;14(46). 31. Gao Y, Gold L, Josif C, Bar-Zeev S, Steenkamp M, Barclay L, et al. A costconsequences analysis of a Midwifery Group Practice for Aboriginal mothers and infants in the Top End of the Northern Territory, Australia. Midwifery 2014;30:447–55. 32. Sutherland G, Yelland J, Brown S. Social inequalities in the organization of pregnancy care in a universally funded public health care system. Matern Child Health J 2012;16(2):288–96. 33. Beake S, Acosta L, Cooke P, McCourt C. Caseload midwifery in a multi-ethnic community: the women’s experiences. Midwifery 2013;29(8):996–1002. 34. Josif C, Barclay L, Kruske S, Kildea S. ‘No more strangers’: investigating the experiences of women, midwives and others during the establishment of a new model of maternity care for remote dwelling aboriginal women in northern Australia. Midwifery 2014;30:317–23. 35. McLachlan H, Forster D, Davey MA, Farrell T, Gold LC, Nicolle Oats J. A randomised controlled trial of caseload midwifery for women at low risk of medical complications (COSMOS): women’s satisfaction with care. J Paediatr 2012;48(Suppl. 1):41. 36. Rumbold AR, Crowther CA. Guideline use for gestational diabetes mellitus and current screening: diagnostic and management practices in Australian hospitals. Aust N Z J Obstet Gynaecol 2001;41(1):86–90. 37. Homer CSE, Biggs J, Vaughan G, Sullivan E. Mapping maternity services in Australia: location, classification and services. Aust Health Rev 2011;35:222–9.

Please cite this article in press as: Dawson K, et al. Implementing caseload midwifery: Exploring the views of maternity managers in Australia – A national cross-sectional survey. Women Birth (2015), http://dx.doi.org/10.1016/j.wombi.2015.10.010