Introduction

The Australian healthcare system and its maternity services operate through a two-tier system of both public and private providers [1, 2]. The national publicly funded universal healthcare program, Medicare, provides hospital and community-based care to Australian and New Zealand citizens [2, 3]. Consumers with Medicare accessing public hospital care predominantly experience no out-of-pocket costs, while community-based care is made more affordable through the provision of government-funded rebates [2, 3]. Conversely, private hospitals and services are accessible with private health insurance and a co-payment or alternatively, incur a full fee for service [2, 3].

Within this system, almost all pregnant women will make initial contact with a general practitioner (GP) when commencing their maternity care journey [4, 5]. GPs provide healthcare across the lifespan and play a key role in advising and directing prospective parents to a maternity model of care [2]. There are many factors that determine birth location and care provider for women. Beyond maternal choice, factors such as geographic location, visa status, medical conditions and socioeconomic identity may impact the models of care available [1, 6]. In addition, GPs may consider the service delivery context, the patient’s health insurance status and the available expertise prior to making a referral [7–9]. They have a gatekeeping function in the way that access to certain models of care is only granted through the GP referral [2]. For example, in Australia, access to a private midwifery service requires a ’collaborative arrangement’ between the participating midwife and medical practitioner in order for the services to be subsidized by Medicare [10]. Similarly, a GP referral is required when accessing private obstetric models of care in order for women to receive a Medicare rebate [2, 11]. It is also important to note that GPs working in the GP obstetrician model of care can be both referring provider and the maternity care provider [2, 12]. Although other referral pathways exist such as through other clinicians or via self-referral, the GP referral pathway is the most common way women access models of care [2, 13].

In Australia, there are 11 models of maternity care according to the Maternity Care Classification System (MaCCs), each has their own specific features, personnel, and fee structure [2, 12]. These can be grouped into five major model of care categories which include standard public care, shared care, public midwifery continuity care, private obstetric care and private midwifery care [2, 12]. Standard public care is provided by hospital doctors and midwives and is government funded [2, 12]. Shared care involves antenatal care delivered by a community doctor and or midwife, with several key antenatal visits and intrapartum care occurring in the public hospital [2, 12]. Hospital-based care within a shared care model is government funded though women may incur out of pocket costs for GP-led visits [2, 12]. Public midwifery continuity of care models are government funded and provided by a hospital midwife or small team of midwives with birth largely occurring in a birth centre or public hospital [2, 12]. Within the public midwifery continuity of care category is midwifery group practice caseload care. Midwifery group practice caseload care involves antenatal care provided by a known midwife or group of midwives in the hospital, community, home, or birth centre setting with intrapartum care taking place in the hospital, birth centre or home [2, 12]. Private obstetric care is provided by an obstetrician with birth occurring in a private or public hospital and is largely accessible via women’s private health insurance though some Medicare rebates may be available [2, 11, 12]. Subsumed within the private obstetric care major model category is general practitioner obstetrician care which involves community or hospital clinic antenatal care and public or private hospital intrapartum care provided by a GP obstetrician [2, 12]. Finally, private midwifery care involves care by a privately practising midwife who supports birth in the home or hospital settings and can be partly government funded for antenatal, postnatal and hospital intrapartum care and partly consumer funded [2, 12].

It is known that 97% of Australian women intend to and will give birth in a hospital setting but little is known about how women distribute across the models of care at a national level as this is not yet routinely reported [2, 14]. The first national model of maternity care report was published in November 2021, has been re-released annually and focusses on the maternity model of care characteristics [1, 14, 15]. As such, there remains limited data comparing and evaluating the models of care in Australia [2, 8, 16, 17]. Despite this, model of care report data and birthplace data can be used to infer how women distribute across the models of care. In 2023, approximately 1000 model of care services were active in Australia and most of these identified as public hospital maternity care (41%), 14% were midwifery group caseload care models, 15% identified as shared care models and 11% were private obstetrician specialist care models [14]. In 2019, Australian birthplace data revealed that 75% of women gave birth in a public hospital, 22% in a private hospital, 2.3% in a birth centre, 0.6% of women constituted births before the arrival to a health service and 0.3% of births occurred in the home setting [18].

Methods

A qualitative description design was used given the recognised utility to derive rich descriptions from participants and its suitability for understanding novel phenomena/on [38, 39]. The research aim prompted the use of a critical realist lens which is ontologically realist asserting that an authentic reality exists separately to human knowing while also being epistemologically subjectivist asserting that multiple versions of reality can be produced [40]. The critical realist lens informed the use of semi-structured interviews for data collection so as to appreciate the multiple versions of reality detailed by GPs. Moreover, its alignment with the reflexive thematic analysis process allowed for the development of key themes from the data thus fulfilling the assumption that an authentic reality exists separately to human knowing.

Findings

The demographic profile of the participants is outlined in Table 1. Of the twelve participants, eleven identified as female. Participants were between 30 and 59 years of age with most participants aged between 40 and 49 years. Most participants were GP Fellows of the Royal Australian and New Zealand College of General Practitioners. Five participants had less than 10 years of experience working in general practice, five participants had between 11 and 20 years of experience and two participants had more than 20 years of experience working in general practice. Most participants were currently living and working in Western Australia. Most of the participating GPs were working in metropolitan areas however five participants were currently working in a rural or regional area.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Demographic profile of participants.

https://doi.org/10.1371/journal.pone.0296537.t001

Two main themes were generated from the data analysis process. The first theme, entitled “GP factors” incorporates three subthemes including “1) Gaps in GP knowledge”, “2) GPs previous model of care experience” and “3) GPs perception of the models of care”. The second theme or the “Woman’s factors” encapsulates two subthemes including “4) Woman’s preferences” and “5) Access to models” (Fig 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Final thematic map of the factors that influence GP referral to maternity models of care demonstrating two themes and five subthemes.

https://doi.org/10.1371/journal.pone.0296537.g001

Discussion

To our knowledge, this is the first exploratory study to describe factors that influence GP referral to maternity models of care. Gaps in knowledge identified in this study may impact GPs ability to offer a comprehensive range of models to women. GPs descriptions of reduced knowledge of midwifery-led continuity of care models may mean that the benefits associated with these models such as reduced intervention and increased spontaneous vaginal birth [45], go unrealized. Given the important referring function of the GP it is likely that these referral practices may be impacting maternal and neonatal outcomes at a population health level.

Resources supplied by Primary Health Networks (PHNs) via the HealthPathways portal was described as being used by two GP participants during their interviews. PHNs can be an important source of information for GP knowledge of the maternity models of care. PHNs assist with health service planning so as to ensure that primary care meets the community’s needs. PHN-developed resources, such as the HealthPathways portal is used to provide local clinical management pathways and referral advice [46]. However, the portal is limited by its completeness and relevance of information across Australia. For example, the Western Australian HealthPathways portal does not include locally relevant model of care referral information and instead has adapted its ‘Antenatal Consult–First’ pathway from New Zealand [47]. Embedding a decision tool into the HealthPathways portal with the up to date and locally relevant model of care service information could provide GPs with support during model of care discussions and provide women with a broader range of model of care options.

Of the interviewed GPs, most participants described personal and work experience in the medical-led models of care and reported a heightened familiarity with those models. With most medical education and GP obstetric training occurring in medical -led- models, this medical focus appeared to impact the exposure and familiarity of GPs with non-medical led models. Moreover, GPs described referring to providers they had previously worked with, and when predominantly occurring in the hospital setting, this reinforced referral to medical providers. Whilst there are no studies that have previously explored medical referring within the maternity context, a cross-sectional study of American primary care physicians (n = 1252) demonstrated that together with other factors, past experience with a specialist was highly important to their referral practice [48]. Further, a prospective cohort study also conducted in the United States found that personal knowledge of a specialist was the most important consideration for primary care doctors when referring patients [49]. Given that most of the interviewed participants had hospital-based model of care experience, early indications are that GPs may preferentially choose obstetricians working in those models of care as opposed to community-based GP obstetricians or midwives.

GPs disclosed referring to providers they perceive positively. A study by Choudhry et al. found that broadly, referral recommendations made by physicians have little or no objective basis. Rather, that the physicians’ perception of the specialist’s reputation, knowledge base and feedback from previously referred patients instead are highly valued when making referral recommendations [50]. Addressing this practice is made difficult as one needs exposure to a model to develop a positive perception of that model. GP training that includes exposure to a variety of maternity models of care, including midwifery-led models of care, may broaden GPs’ perceptions of a range of models of care.

The woman’s choice and preferences were reported by GPs as central to the referral decision, giving insight into what they consider women view as important including birth location, mode of birth and continuity of carer. At face value, these findings may indicate GP alignment with the key intentions of the national maternity Strategy, particularly regarding choice and informed-decision making [20]. The reluctance demonstrated by some participants to refer to privately practising midwives in this research has been explored in previous studies [8, 51, 52]. In a national survey of n = 1657 participants, narratives from women describing challenges of having to ‘fight’ with their GP to obtain a referral to a private midwife or to receive midwife only care [52]. GP reluctance or refusal to refer to midwifery models of care requires further exploration at a national policy level. Changes in the National Law requiring private midwives to have a collaborative arrangement with a GP are underway, it is anticipated that this will improve women’s access to private midwifery and remove the requirement for referral from a GP [10].

Finally, GPs described access to the models of care as a key consideration during interviews. GPs reported how rural and remote women have limited options for maternity care. This finding is congruent with current literature that outlines how women residing in remote geographical areas experience fewer or minimal options for maternity care access compared with their metropolitan counterparts [53]. Improving maternity care access for rural and remote women is a key objective within the national maternity Strategy. Strategies such as optimising the of the maternity workforce, service innovations such as telehealth and outreach care, as well as national pathways, are critical to achieving improved access for rural and remote women [20]. A national review of health professionals’ scope of practice is currently being undertaken by the Commonwealth government which aims to address barriers in equity and access for all Australians including women seeking rural maternity services [54]. The interviewed GPs also described how characteristics like maternal fetal risk profile, the financial status of the woman, and convenience, may mean that current model of care options do not meet the needs of women. Despite current risk stratification practices, there is compelling evidence regarding the benefits of continuity of midwifery care for women of all risk profiles alongside a multidisciplinary team [25, 55]. Although publicly funded models of care represent those most commonly accessed by Australian women, rising out-of-pocket costs serve as a barrier to accessing care particularly for those living in rural areas with limited primary maternity services offering Medicare rebates that cover the full cost of care [56]. Strategies such as funding models that support continuity of care for all women, shifting GP practice funding to support patients in need, bundled funding models, and expanding access to expert primary care through clinically activated PHNs have been discussed [20, 57–59]. The interviewed GPs described the woman’s proximity to the model of care as similarly important to other doctors when undertaking a referral [48, 50, 60]. The benefit of convenient, proximal maternity models of care include reduced work and family disruption, reduced travel related safety issues, and the ability to facilitate birthing on country, a priority for many Aboriginal women [20].

Maintaining currency of knowledge of the maternity models available within a local referring area presents a challenge for GPs that could be met by providing online resources to support broader conversations with women and referral practices. Around the country, individual jurisdictions have begun to address the gap in knowledge, examples include the Pregnancy, Birth and Baby website and the MyBaby WA App which targets new or expectant mothers but may also be used by health professionals to support a range of conversations including model of care choice [61, 62]. Education, beginning early on in medical schools, and in GPs training programs on the breadth of model of care options available is important to increase awareness and enable GPs to refer from a broader knowledge base. In their public health education role, GPs could partner with women to enable a shared understanding of their options so all stakeholders can experience the benefits of consumer-demand and led access to a wider variety of models.

Within GP education and training, exposure to non-hospital-based models of care may provide the opportunity for GPs and training doctors to gain a sense of familiarity and enhance their knowledge of those models. This and expanding the role of midwives in the primary health sector including in primary health network (PHN) stakeholder groups would foster networks between GPs and other providers improving interdisciplinary care and health service integration. Despite no current evidence evaluating the benefit of midwifery integration into primary care clinics or PHN stakeholder groups in Australia, a predominantly European study that surveyed GPs and patients across 34 countries evaluated GP outcomes and patient experiences of GPs co-located with other health professionals including midwives [63]. Of the 7183 GPs surveyed, most were co-located with nursing staff while only one in three described themselves as co-located with midwives, specialists, physiotherapists, dentists or pharmacists [63]. The analysis demonstrated that co-location with the aforementioned allied health staff or midwives resulted in improved coordination with secondary care [63]. As such, the integration of midwives into GP practices or other primary settings such as community centres may confer improved coordination with secondary care. More research into the benefits of midwifery integration into the primary health sector and in PHN stakeholder groups is needed.

Finally, insights gained around women’s ability to access models of care is helpful for health system planning with a view to improve health outcomes, choice for women, and the range of models of care available to them. Strengthening the rural workforce by training GP obstetricians, enabling scope fulfilment for midwives who have recognized expertise in primary maternity care, using telehealth and outreach services and national pathway development have been highlighted as vital to improving healthcare delivery to rural women [20, 64]. Moreover, the development of all-risk caseload midwifery models of care [25, 55], funding that supports continuity of care options [20, 57, 58] and model of care design that improves the convenience elements of maternity models [20] have the potential to better meet the needs of new and expectant parents in Australia.

Strengths and weaknesses

The strengths of this study lie in the qualitative descriptive approach to this newly investigated topic. Selecting this methodology enabled an open exploration of the phenomena which provides novel evidence and delivers recommendations for further related areas of enquiry. Many of the participant GPs were from one Australian jurisdiction which may limit the generalisability of this research although this not a goal of qualitative research. The detailed demographic data provided enables the reader to determine the transferability of the study findings to their individual setting.

Conclusion

Study findings have delivered novel evidence regarding the factors that influence GP referral to maternity models of care in Australia. There are known perinatal health, economic, societal benefits conferred at a population level for women who access selected models of care. The findings of this study provide important signposts about why and how models of care referral practices could be broadened and improved to access these conferred benefits for all Australian families. Further research, education and policy reform is recommended to improve women’s access to models of care via more informed GP referral practices.

Supporting information

Acknowledgments

We would like to acknowledge the general practitioners who dedicated their time to this research while still working among a global pandemic to share their experiences providing us with a deeper understanding of the factors influencing referral to maternity models of care.