SUMMARY

In 1997, an innovative model of community-based midwifery care was implemented in Australia as a response to numerous state and federal government reports. The St George Outreach Maternity Project (STOMP) provides continuity of care through the antenatal, intrapartum and postpartum period. Hospital salaried midwives and obstetricians provide antenatal care from community-based settings. Community-based antenatal care was unusual in Australian settings at that time. Postnatal care is provided in hospital and in the community.

A randomised controlled trial was conducted to test the efficacy of the new model of care. Women who attended the community-based antenatal clinics perceived that they had a higher ‘quality’ of antenatal care compared with control group women. Women allocated to STOMP also had a significantly lower caesarean section rate. The model was instituted within the current hospital budget using existing personnel. The mean cost of providing care per woman was lower in the STOMP group compared with the control group.

This endeavour used research as a means to introduce change in an Australian maternity service. The STOMP model was cost-effective, liked by women and had positive clinical outcomes. It is now fully integrated into the hospital system and offered as an option for all women.

INTRODUCTION

Community-based maternity services, particularly those providing antenatal care with midwives, are uncommon in the Australian public health system. This chapter will present the experience of introducing a community-based model of continuity of midwifery care in an Australian public health system. This model is known as the St George Outreach Maternity Program (STOMP). STOMP was implemented in 1997 and was evaluated using a randomised controlled trial. The model continues to operate in much the same manner as it was first designed. The results of the STOMP study have been widely published (Homer 2000, Homer et al 2000a, Homer et al 2001a, Homer et al 2001b, Homer 2002, Homer et al 2002a, Homer et al 2002b). Only a summary of the results is presented in this chapter. The chapter will use STOMP to illustrate some of the issues currently facing midwifery and the health system in Australia.

FACTORS INFLUENCING THE DEVELOPMENT OF THE STOMP MODEL

A number of factors influenced the development of the STOMP model. These include recommendations from local, state, national and international policy documents, research evidence and the commitment to improved services that was present at St George Hospital where the study was conducted. A primary health care approach was used to design STOMP. This framework (encompassing equity, access, the provision of services based on need, community participation, collaboration and community-based care) was used as a means to bring about change in a public health system that was generally dedicated to the provision of an acute care model.

Other important determinants that guided the development of the STOMP model included financial considerations, the consultation process within the organisation and the experience of maternity units in the UK, where team midwifery schemes have been discontinued. The evaluation was influenced by the characteristics of the population, the need to address issues of disappointment or measurement bias and the importance of a rigorous appraisal.

POLICY STATEMENTS ADVOCATING CHANGE

A number of state and national government reports in Australia have recommended major changes to the provision of maternity services. Recommendations include providing opportunities for continuity of care, increasing collaboration between midwives, obstetricians and general practitioners (GPs) and moving antenatal care to the community (NSW Health Department 1989, Victorian Department of Health 1990, NHMRC 1996, NSW Health Department 1996, Maternity Services Advisory Committee 1999, Senate Community Affairs References Committee 1999). Essentially these reports recommend moving maternity care from an acute care framework to one which has primary health care as its focus.

Two of the Australian reports were most instrumental in the development of the STOMP model in late 1996. In New South Wales (NSW), the landmark review of maternity services known as the Shearman Report (NSW Health Department 1989) emphasised certain principles in its recommendations. These included: equitable access to quality care; recognition of the needs of women from non-English speaking backgrounds (NESB); maximising each woman’s participation in decision-making during pregnancy, childbirth and the postpartum period; and promoting co-operation and collaboration among doctors, midwives and other health professionals. The report recommended that options should be explored to expand and to redefine the role of hospital-employed or salaried midwives, and suggested that these midwives could be located in community health centres to provide care during pregnancy and childbirth for low-risk women. There were also a number of strategies to meet the needs of women from NESB. These included: increased funding for interpreter services; development of new models of care, including midwives’ clinics and shared care with bilingual GPs; and the establishment of ethnic obstetric liaison midwives to provide continuity of care and education.

Widespread change in the provision of maternity services and the development of new models of care did not occur in NSW public maternity services as a result of the Shearman Report (NSW Health Department 1989). This dearth of change was one of the driving forces behind the development of STOMP. The Shearman Report provided a valuable framework of recommendations on which to guide the design of the model.

The peak health body in Australia, the National Health and Medical Research Council (NHMRC) released Options for Effective Care in Childbirth in 1996 (NHMRC 1996). This report also guided the development of the STOMP model and the evaluation. Recommendations in this report included facilitating continuity of care and carer in the antenatal period and encouraging the development of small teams of midwives and general practitioner obstetricians. The report stated, ‘we suggest [a model of joint practice run by midwives and obstetricians providing continuity of care] deserves more attention and appropriate evaluation by both professional and health planners’ (1996:26).

Continuity of care and community-based care were important components both in this report and in others conducted in Australia at a similar time (Department of Health Western Australia 1990, Victorian Department of Health 1990). The Australian National Non-English Speaking Background Women’s Health Strategy (Alcorso & Schofield 1992) also made recommendations which assisted the development of STOMP. For example, the strategy suggested that outreach midwifery schemes offering continuity of care to women from NESB should be introduced in Australian public hospital systems to ensure that care is provided within local communities.

Two reviews of maternity care in the United Kingdom (UK) were also influential in the development of the STOMP model. The Winterton Report (House of Commons 1992) highlighted the need for women to have choice, continuity and control in the birth of their babies. Changing Childbirth, also known as the Cumberledge Report (Department of Health Expert Maternity Group 1993), was the English government’s response to the Winterton Report. Changing Childbirth focused on the provision of maternity services and set specific targets and indicators for the providers of maternity care. The recommendations from the report (1993:18) were based on three fundamental principles of care, which were most relevant in the development of STOMP:

Despite all these reports and recommendations over a decade, by 1996 it seemed that few public hospital maternity services in Australia had managed to achieve the widespread change necessary to introduce the components of continuity of care and community-based care into the provision of maternity care. The STOMP model was an endeavour to achieve change within a public sector metropolitan hospital in Sydney.

RESEARCH SUGGESTING CHANGE

Previous research into models of care that provide continuity of midwifery care suggested that there were positive benefits for women and health systems. Continuity of midwifery care has been shown to reduce interventions in labour, particularly augmentation of labour, analgesic use and electronic fetal monitoring (Flint et al 1989, Kenny et al 1994, Rowley et al 1995, Waldenström & Nilsson 1993, Waldenström & Nilsson 1997). A small Canadian trial in 200 women demonstrated a significant reduction in caesarean section rate (Harvey et al 1996) and one of the Australian trials reported a trend towards a reduced elective caesarean section rate in high-risk women (Rowley et al 1995). A retrospective cohort study in California has also shown that supportive nurse-midwifery care in labour was associated with a reduced caesarean section rate (Butler et al 1993).

Continuity of midwifery care has been shown to improve women’s experiences with care during pregnancy and childbirth (Flint et al 1989, MacVicar et al 1993, Waldenström & Nilsson 1993, Kenny et al 1994, Rowley et al 1995). In particular, women who have received continuity of care report greater preparedness for birth and early parenting (Flint et al 1989, McCourt et al 1998), increased satisfaction with psychological aspects of care (Waldenström & Nilsson 1993) and higher participation in decision making (Turnbull et al 1996) than women who received standard care.

Continuity of midwifery care has been associated with reduced costs to the health system in two Australian studies (Kenny et al 1994, Rowley et al 1995), although there were deficiencies in both cost analyses, demonstrating the need for more research. Results from these studies were compelling and influential in the development and design of the STOMP model.

LOCAL COMMITMENT TO CHANGE

Another important factor in the development of the STOMP model was the extent of the hospital’s commitment to change. The maternity unit at St George Hospital in Sydney had been committed to improving their service over a number of years. This was evident from a series of innovations that had already occurred in the maternity unit. For example, a birth centre was established in 1990 as a result of the Shearman Report (NSW Health Department 1989). The birth centre was one of only three in Sydney at the time. Despite initial difficulties, with opposition from obstetricians and midwives, the birth centre remains a well-established option for women and excellent clinical outcomes have been reported (Homer et al 2000b).

The establishment of a midwives’ clinic in 1995 was another example of the maternity unit’s commitment to an improved service. The midwives’ clinic enables women of low obstetric or medical risk to have continuity of midwife ‘carer’ throughout the antenatal period. This clinic was established partly as a result of the Shearman Report (NSW HealthDepartment 1989) but also in response to a customer survey conducted in 1994 (Everitt et al 1995).

This customer survey, known as the Maternity Services Customer Satisfaction Research Project (Everitt et al 1995), was another important factor in the development of STOMP. The survey used a combination of qualitative and quantitative methods to establish customer satisfaction levels and identify problem areas in the service provided by the hospital. A sample of women from English and NESB who were current, recent or potential users of the service were included in the survey. Problems identified included: the lack of continuity of care and carer in the antenatal and postnatal periods; insufficient respect for individual opinions and beliefs; and conflicting advice regarding breastfeeding. Difficulties accessing antenatal care at the hospital (because of a lack of car parking facilities) were also reported. Women from NESB reported difficulties in obtaining culturally appropriate care and accessing adequate information. The survey made 26 recommendations, including the establishment of new models of care that provide continuity of care and carer, and the consideration of community-based antenatal clinics. The STOMP model was developed to target specifically these two recommendations.

THE CONSULTATION PROCESS

The process of implementing the new model of care began during the latter half of 1996 with a series of formal and informal discussions between midwives, obstetricians and managers in the maternity unit. The purpose of these early meetings was to discuss the principles of continuity of care and community-based care and to canvass opinions about the proposed shift to a model of team midwifery. The researcher and others wrote a paper describing the issues around continuity of care and the change to the organisation that would result from the introduction of midwifery teams. This paper was distributed to all midwives, managers and obstetricians. Numerous in-service sessions and frequent informal interactions were conducted with staff members. External consultation also took place. This included discussions with experts in models of midwifery care in Australia (Kenny et al 1994, Rowley et al 1995) and in the United Kingdom through a study tour. A working party, which included midwives, obstetricians, managers and researchers, was established to develop the model initially, and subsequently to guide the implementation and evaluation.

During the consultation and development phase, it was decided that two STOMP teams would be established. Each team would consist of six midwives and provide antenatal, intrapartum and postnatal care for 300 women per year. Establishing two teams of midwives was unusual in Australia. Both previous projects had been based on only one team of midwives (Kenny et al 1994, Rowley et al 1995) which meant that access to the model was limited to fewer than 300 women annually. Other research in the UK and Sweden also involved only one team of midwives (Flint et al 1989, Turnbull et al 1996, Waldenström et al 1997).

FINANCIAL CONSIDERATIONS

The maternity unit at St George Hospital did not have any additional funds to establish new models of care. Therefore, the STOMP model was designed with the understanding that no additional funding would be available for the implementation. The model was aimed at women without private health insurance who were attending a public hospital for maternity care. Charges were not levied on the women receiving care. The STOMP model was implemented by reorganising the current maternity service’s existing resources and staff. Internal restructuring provided the midwifery staff for the two STOMP teams by shifting 12 midwives from their existing wards or units, for example antenatal, labour and delivery and postnatal wards, to create the teams.

In many ways, the lack of additional funding was an advantage, rather than a disadvantage. Implementation within an existing budget meant the model was embedded in the organisational structure from the outset. We anticipated that full integration would make the new model less vulnerable to discontinuation in times of budgetary constraint.

Integration of innovation in maternity care has been uncommon in Australia. Pilot programmes have usually been established with the assistance of additional funding either from federal or state government bodies (Kenny et al 1994, Hambly 1997, Thiele & Thorogood 1997). This can mean that programmes are vulnerable to discontinuation when the support ends. It also can mean that the programmes are not seen or managed as a part of the existing or ‘mainstream’ service.

EXPERIENCE IN THE UK

The 1993 report from the UK, Mapping Team Midwifery (Wraight et al 1993), was another important determinant in the development of STOMP. Mapping Team Midwifery was a review of team midwifery schemes that had been implemented as a result of Changing Childbirth recommendations (Department of Health Expert Maternity Group 1993). Of concern was the finding that more than one quarter of schemes established in 1990 were discontinued by 1991. Discontinuation occurred because of inadequate staffing levels, problems with deployment onto the teams, lack of commitment from midwives and obstetricians, lack of consultation, discontent among midwives, failure to increase continuity of care, and personality clashes within teams. These factors were important in the development of the STOMP model.

SITUATING THE SERVICE

The STOMP model responded to the state and federal government reports that recommended antenatal care be based in the community (NSW Health Department 1989, Victorian Department of Health 1990, NHMRC 1996). Women in Brisbane, Queensland, have reported choosing community-based care because of availability with appointment times and decreased travel and waiting time (Del Mar et al 1991, Ramsay 1996). Previous models of continuity of midwifery care in Australia have provided antenatal care from hospital-based clinics (Kenny et al 1994, Rowley et al 1995). Community-based antenatal care in Australia is generally only available to women who attend private medical practitioners, either specialist obstetricians or GPs. Midwives have a limited, if any, role in these models.

Community-based antenatal services provided by GPs and midwives have been evaluated in the UK and found to be feasible, satisfactory for the majority of women (Williams et al 1989, Fleissig et al 1996) and offer greater flexibility and choice (Perkins & Unell 1997). The ‘One-to-One’ midwifery project is an example of a model of continuity of carer in a community setting. ‘One-to-One’ midwifery (McCourt & Page 1996) was established as a demonstration project in the UK as a result of Changing Childbirth (Department of Health Expert Maternity Group 1993). The project provided care for all women regardless of risk group, in both hospital and community settings. The results indicated that women had a strong preference for community-based antenatal care.

Community-based maternity services, other than those provided in the private sector, are uncommon in the general Australian public health system. A review of the literature failed to uncover any reports of a ‘mainstream’ community-based antenatal program in Australia. A number of small pilot projects have provided community-based antenatal care by midwives (Ramsay 1996, Hambly 1997, Thiele & Thorogood 1997). These projects were small and available to a limited number of predominantly ‘low-risk’ women. A small number of special community-based antenatal services that have catered for specific disadvantaged groups, for example, adolescent (Brodie 1994) or indigenous women (Bartlett et al 1998) have been reported. These services are unavailable to most women because they only cater for minority groups.

In Australia, the move to community-based care has been interpreted by public health systems as a cost-saving measure. By virtue of the manner in which health services are funded in Australia (Leeder 1999), costs of providing care can be shifted from the state-funded public hospitals to the federally-funded GP in the community. It does not always follow that women receive better care, or indeed more cost-effective care. Problems have been reported in research from Melbourne (Brown et al 1999) including: fragmentation of services and provider; an increase in the number of antenatal visits and costs to women; duplication in investigations; variability in the quality of care; and a lack of co-ordination of care.

Difficulties have been experienced in Australia when midwives have attempted to work in small teams in the community. One of the reasons for this is the system of health care funding in Australia mentioned above. State governments fund public hospitals and their associated services, while the federal government funds GPs through the Medicare system (Leeder 1999). Therefore, there are currently no mechanisms to allow midwives to be contracted to provide antenatal care in GP practices. General practitioners are thus reluctant to employ midwives as they are unable to attract a Medicare rebate for midwifery care. It is unlikely that this current funding system is going to change, at least in the next decade (Senate Community Affairs References Committee 2000).

Stay updated, free articles. Join our Telegram channel

Join