The ships of the First Fleet, which landed in Sydney in 1788, carried several women who were free settlers as well as a number of convict women. The ships’ logs record that during the long voyage to Australia several women gave birth, allowing others to gain midwifery experience (Adcock et al 1984). The military and ships’ surgeons accompanying the colonising forces probably had little or no midwifery expertise, and there were no midwives listed among either the free settlers or the convict women. Therefore it appears that midwifery in Australia began with women helping each other as best they could, accessing medical help where it was available and when it was required. Some of the women who found themselves in the role of midwife continued to assist women in childbirth and became well known, loved and respected for their abilities.

Female convicts were transported to Australia for the next 50 years in an attempt to empty English prisons of ‘hardened cases’ but covertly also to provide sexual services for men (who outnumbered the women six to one) and ultimately to stabilise the economy of the new colony (Rowley 1998). Early census records reveal that by 1806, 2% of women were in skilled trades, which included two women who listed their occupation as midwife, although with every woman under the age of 42 producing a baby each year, there were clearly more than two ‘midwives’ in the colony (Adcock et al 1984). No ‘learned’ midwives were recorded among those early settlers until 40 years later, when Mrs McTavish, identified as the first ‘trained’ midwife to settle in Australia, advertised her services in a Hobart newspaper (Barclay 1993, cited in Rowley 1998). Therefore it is reasonable to propose that the midwifery traditions of Australia were established by community-based, ‘lay’ midwives, without access to any theoretical knowledge or teaching other than what they had gleaned from observation and experience. These were the ‘accidental’ or empirical midwives of the convict era assisting women to give birth in whatever place constituted ‘home’.

The first maternity hospitals in Australia

In 1820, a midwife was appointed to the Female Factory at Parramatta, which was built to house female convicts. Some midwives chose to work in the Female Factory for short periods to gain experience before moving out into the community and private practice. Convicts in domestic service who became pregnant were sent to Parramatta for punishment and confinement. Once delivered, the women returned to their employer, often leaving the baby behind, where their infants were wet-nursed by 30 convict ‘nurses’ who resided there. The Female Factory became the first maternity hospital, as the pregnancy rate among convicts was high. Soon, poor and destitute women also sought to be confined there, since the authorities were reluctant to build hospitals for the general populace. The Female Factory was eventually closed due to an epidemic of puerperal sepsis. Following the closure, convict women continued to give birth at home, as did the free settlers, attended by relatives, neighbours, or a midwife if they could afford one or find one, but rarely by a doctor.

The transportation of convicts ended in 1848 just before the discovery of gold near Bathurst in 1851. The new colony prospered. The government of the day encouraged the immigration of young single women to redress the imbalance in the sexes and to populate the country. As settlers moved out into rural areas, even neighbours were sparse, and there are numerous accounts of women being attended by Aboriginal women during childbirth (Willis 1989). Learned midwives who had received midwifery training in England and Scotland were also among the new immigrants. As the settlements grew into towns, some midwives began taking women into their own homes for ‘confinement’, and thus began the first private maternity homes, which were eventually to become community or primary hospitals (Shephard 1989, 1991). Later, concern over the deplorable conditions under which poor and destitute women were ‘confined’ in their homes led to the establishment of initially charitable and ultimately State-funded women’s hospitals in the cities of Melbourne (1886) and Sydney (1893) just prior to the turn of the 20th century (Forster 1965).

The origins of midwifery in New Zealand

Childbirth for Pākehā ¹ women in New Zealand prior to the 1904 Midwives Act was described by Donley (1986) as a neighbourhood affair conducted in homes. Women were attended by either (English or Scottish) trained or lay midwives, who took charge of domestic responsibilities as well as supporting the woman in labour, delivering the baby and getting breastfeeding established. As in Australia, these early midwives were loved and respected for their competence and care, and there are several accounts of the good records of the pioneering midwives in terms of maternal and perinatal mortality (Donley 1986). As the towns grew and cities evolved, many midwives set up their own small, private maternity homes. It is estimated that by the turn of the 20th century, there were over 200 one- or two-bed maternity homes run by midwives or by doctors, located throughout the towns and cities of New Zealand (Mein Smith 1986).

The first maternity hospitals in New Zealand

Several events coalesced just after the turn of the 20th century that initiated major changes in the way childbirth was managed, and changed it forever from a relatively private family affair into a concern of the State. Reports to Parliament had for some time recorded the fluctuating maternal and infant mortality rates in the new colony and, in particular, the rate of maternal deaths from puerperal sepsis. In 1903, a peak in the maternal mortality

St Helens Hospital, Dunedin, New Zealand c. 1923. Reproduced with the permission of Alexander Turnbull Library, Wellington, New Zealand. Photo from the SC Smith Collection

rate caused alarm in government circles. At the same time a Royal Commission set up in New South Wales in 1904 to investigate falling birth rates in both Australia and New Zealand found that the decline was highest among the ‘better classes’ and that ‘while the “unfit” were having many children … [the better classes] had a higher rate of infant mortality’ (Donley 1986, p 32). The Premier, Richard Seddon, demanded action and a champion emerged in the person of Grace Neill,² who was able to persuade him that the way to increase the birth rate and improve the appalling rates of maternal and infant mortality was to register all midwives and establish State-subsidised hospitals, where the wives of working-class men (the deserving poor) could give birth in comfort and safety. This saw the setting-up of St Helens Hospitals in the major centres of New Zealand, with the first established in a rented cottage in Rintoul Street, Wellington in 1905 (Donley 1986). The hospitals provided midwifery training for both nurses and women without a nursing qualification, and offered either hospital or domiciliary care (without prejudice).

Why did women move from home to hospital?

Over the next 20 years, simultaneously in both Australia and New Zealand and in many other parts of the world, women started to move from home to hospital birthing in increasing numbers. Why this happened is an intriguing question. A simple answer would be to find that mothers and babies died in large numbers at home in the care of midwives, and that women chose to move into hospital, where medically managed birth was safe. This is not the case, however. Different authors quote a variety of maternal and infant mortality rates, all purporting to provide evidence of either a dramatic improvement in, or worsening of, mortality as a consequence of the move (Ehrenreich & English 1973; Shorter 1983). Gaining a clear picture

Trainee midwives,St Helens, Wellington 1927. Reproduced with the permission of Archives New Zealand

of what was happening at the time is difficult, and this allows different interpretations of the significant events to emerge. What motivated women to move from birth at home to hospital can never be known for certain, but the parallel movement in Australia and New Zealand suggests that the motivators for the change may have been similar, and several issues can be identified which may have played a part.

The first is the issue of falling birth rates at the turn of the 20th century and the intervention of both colonial governments in childbirth, with the aim of increasing the size of their respective populations and ensuring their health and vigour (Donley 1986). The health departments in both countries promoted the hospitalisation of birth in order to decrease the rate of maternal deaths particularly from puerperal sepsis, and to ensure women accessed antenatal care that would lead to the birth of a healthy baby. Both governments were disturbed by the parlous physical state of many of the men recruited into the armed forces during the first World War, and saw the birth of a healthy baby as essential to the health of the nation in the event of another war (Mein Smith 1986).

Other themes relate to the views of women themselves and what they may have been seeking. Some may have sought increased material comfort around the time of the birth, because many homes in both New Zealand and Australia were described as lacking in all but the barest of necessities (Mein Smith 1986; Rowley 1998; Tew 1995). Other women may have found the promise of a temporary release from domestic burdens attractive (Tew 1995). Still others may have been seeking the support and company of other women, which had been dislocated by the Industrial Revolution (Wilson 1995), or greater access to doctors and their forceps (Loudon 1992; Rowley 1998; Tew 1995), or greater access to midwives since the lay midwife had largely disappeared from the community following the setting-up of registration and hospital-based training. Added to these issues were the promises of a pain-free labour (Loudon 1992) and increased safety for themselves and their infants, largely and falsely promoted by the medical profession (Tew 1995). In New Zealand, the medical profession actively encouraged women’s groups in political activity to persuade the government to build more maternity hospitals, which then became a focus for the growing power of the emerging medical specialty of obstetrics (Mein Smith 1986). All these issues have been debated in the literature cited, and the student of history is encouraged to pursue particular lines of inquiry using the references and further reading lists at the end of this chapter as a guide. It is interesting to note, for instance, that both forceps and the pain relief offered by twilight sleep were liberally administered by doctors attending women in childbirth at home, so these two reasons alone do not seem to be convincing arguments for the move to hospital (Forster 1965; Mein Smith 1986).

Far from increasing safety as promised, deaths from puerperal sepsis increased with hospital birth, in all but the St Helens Hospitals (Wood & Foureur 2005; Mein Smith 1986), but this appeared to go unnoticed by women as they started to move into hospitals in increasing numbers. Mein Smith (1986) asserts that by 1920, most New Zealand women continued to give birth at home, while approximately 35% of deliveries occurred in hospitals. In Australia, ‘births in public institutions … increased from 3% in 1907 to 7% in 1920 but then leapt ahead to 55% in 1929’ (Tew 1995, p 65), with one account quoting a rate as high as 67% for hospital deliveries in 1925 in the state of Victoria (Loudon 1992).

Reflective exercise

Review the literature about puerperal sepsis and contributors to this fatal infection. Reflect on the effects of maternal mortality to the families during this period of time. Review the reasons why the rates of puerperal sepsis were higher in some maternity homes as compared with the St Helens Hospitals.

By 1935, deaths from puerperal sepsis were rarely seen, due to the advent of the drugs Prontosil and, later, sulfanilamide (Tew 1990). Perinatal mortality also began to decline between the world wars, due to dramatic improvements in the general health of women and raised living standards (Johanson et al 2002). However, in the minds of many, moving to hospital for birth had improved safety for women and babies, and it was not until 1990 that a critical history of maternity care undertaken by Marjorie Tew (1990) was able to convincingly demonstrate the fallacy of this belief.

Today, the vast majority of women in either country will give birth in hospital, be it a birth centre, primary birth unit or secondary/tertiary hospital. The most recent figures indicate that home birth occurs in less than 1% of the population of childbearing women in Australia (601 women in 2005; AIHW 2007). The estimated rate is much higher in New Zealand, at up to 6% nationally (3535 women in 2005; NZHIS 2008). The National Minimum Data Set in New Zealand includes only hospital births, so rates of home birth can only be estimated by subtracting the number of hospital births from Birth Registration data. Whatever the true figures, birth at home is the choice of few women. Or, paradoxically, is it that there is no choice?

WHAT WAS LOST IN THE MOVE?

Several important things were lost in the move from home to hospital. The first was the opportunity to labour in a familiar environment. The second was the close personal and trusting relationship between the woman and her midwife and the continuous support in labour that the midwife provided. The third was the belief in the concept of birth as a social event; a normal, physiological life event. These were, and are still, universal aspects of home-birth provision, and the whole package of care provides clear benefits for women (Walsh 2004). Let us explore these ideas a little more.

The concept of ‘environment’ is multifaceted and encompasses much more than the geographical location or physical bricks and mortar of the location for birth. It is important to consider that ‘environment’ also includes the spiritual and emotional space and place in the mind and heart of the woman (Fahy et al 2008; Simkin & Ancheta 2001). We must also acknowledge that the environment too exerts a powerful influence on the midwife and that, in the future, new areas of research in what some have termed ‘neuroarchitecture’ (or ‘psychogeography’) will improve our understanding of this concept (Foureur 2002; Lepori 1994; Lepori et al 2008; Newburn 2003; Page 2002; Walsh et al 2004).

In their calls for more home-like environments for birth, more continuity and more choice and involvement in decision-making, women may have unknowingly articulated their longing to replicate the idealised birth environment of home. Policy-makers have attempted to put back components of the package, and researchers have undertaken numerous studies to explore the safety and impact of differences in location for birth (home, birth centre, primary unit, hospital), type of care provider (medical, midwife, doula), models of care (fragmented versus continuity of care and carer) and philosophies of care (belief in birth as a normal physiological event or only normal in retrospect; risk-embracing or risk-averse). However, if the three components of birth at home are an integrated and inseparable package, it becomes apparent that most studies to date have focused on either one or another part of the package. As a consequence, most studies are limited in what they can contribute to our understanding of this complex event.

Hodnett and colleagues realised the synergistic nature of caregiver and location for birth when they wrote that the environment may favourably influence caregivers’ attitudes towards the care of labouring women, and that therefore it may be the influence of the caregiver more than just the location for birth that leads to good obstetric outcomes (Hodnett et al 2005). We will return to this idea later in the chapter, as it is apparent that midwives who move between homes and hospitals are aware that they practise differently in each location in response to overt and covert cultural messages conveyed by each location.

Birth at home or in hospital: which is safer?

The debate concerning the safety of the home as the place of birth has been in progress for over 100 years and no doubt will continue into the future, unless it becomes possible to conduct an extremely large trial where women are randomly allocated to either a home or a hospital birth. One such study involving only 11 women was identified during the process of systematic review published by the Cochrane Collaboration (Olsen & Jewell 1998). Because of the small size of the study, the reviewers were forced to conclude that there is no strong evidence to favour either planned hospital birth or planned home birth for low-risk women.

Many studies using less-robust designs (such as observational, case-control or cohort studies and audit of maternity services) have been conducted internationally (for example: Chamberlain et al 1997; Rooks et al 1989; Young & Hey 2000). Most studies of this nature suffer from a lack of ‘denominator data’, meaning that the researcher cannot be certain that all women planning to give birth at home in their particular data set have been accounted for, and many also suffer from an inability to exclude unplanned home births. Therefore, studies may under- or over-estimate the risks. However, one of the largest prospective cohort studies of home birth, published in 2005, accurately identified all births and concluded that outcomes for home-birth women were substantially better than for low-risk American women having hospital births (Johnson & Daviss 2005). Among the 5418 women who planned to give birth at home when labour began, there was a 12.1% transfer rate at the beginning of labour. Intervention rates were significantly reduced, with epidural at 4.7%, episiotomy 2.1%, instrumental delivery 1.06% and caesarean section 3.7%. The intrapartum and neonatal mortality was 1.7 deaths per 1000 planned home births when congenital anomalies were excluded.

A more recent population-based study of a cohort of 529,688 women was conducted in the Netherlands, which has a uniquely supportive maternity care system where 30% of women elect to give birth at home (De Jonge et al 2009). This study established that perinatal mortality and morbidity were not increased by planning to birth at home in a society and settings where home birth is well supported and midwives are well integrated into the maternity care system.

A further robust matched-cohort study conducted over a four-year period between 2000 and 2004 in British Columbia compared outcomes for 2889 low-risk women who chose to birth at home with 4752 who chose to birth in hospital with the same cohort of midwives (Janssen et al 2009). A third matched group of 5331 women cared for by physicians and birthing in hospital was also recruited to the study. Perinatal mortality was comparable between the three groups, indicating that home birth in the care of midwives was as safe as hospital birth. In addition, the study revealed lower rates of obstetric intervention (e.g. continuous electronic fetal monitoring, assisted vaginal delivery) and lower rates of adverse maternal and perinatal outcomes for women birthing at home. It is of note that the same cohort of midwives provided care for two of the three groups of women, at home and in hospital, with different outcomes in each location. We need to understand why this is the case.

In a New Zealand retrospective cohort study using Midwifery Maternity Provider Organisation data for 2006 and 2007, Davis and colleagues compared childbirth outcomes for low-risk women in the care of midwifery lead maternity carers (LMCs) planning to birth (at the start of labour) at home, in primary, secondary and tertiary facilities (Davis, personal communication 2009). A cohort of 16,200 low-risk women was identified. Compared with primary birthing facilities, women planning to birth at home had significantly less risk of augmentation of labour, pharmacological pain management and artificial rupture of membranes. There was no difference in mode of birth, admission to neonatal intensive care, or 5-minute APGAR score less than or equal to 5.

Compared with primary birthing facilities, women planning to birth in secondary or tertiary hospitals had increased risks of augmentation of labour, pharmacological pain management and artificial rupture of membranes. Mode of birth was also significantly different for this group, with increased risk of ventouse, forceps and emergency caesarean section. The risk of emergency caesarean section was six times higher for women planning to birth in tertiary hospitals (RR 6.12, 95% CI 4.88–7.68)³ and three times higher for those planning to birth in secondary hospitals (RR 2.99, 95% CI 2.39–3.75) compared with those planning to birth in a primary facility. While there was no difference between groups for 5-minute APGAR score less than or equal to 5, babies of women planning to birth in tertiary hospitals had almost twice the risk of being admitted to a neonatal intensive care unit (RR 1.88, 95% CI 1.39–2.53) (Davis, personal communication 2009).

While an important and robust study, the retrospective nature of the Davis study means that the data may be subject to bias, and therefore further prospective studies of this nature are required. One such Australian-led study has received National Health and Medical Research Council (NHMRC) funding. This study will prospectively recruit women who have chosen to labour in midwifery-led units in Australia (two units in New South Wales) and in New Zealand (two units in Christchurch), and will compare outcomes for these women with matched cohorts of women who would be eligible for care in the midwife-led unit but who have chosen instead to labour in the referral hospital in the same area health service (Tracy, personal communication 2009). We await these findings with interest.

The published studies all strongly support the findings of the comprehensive UK study ‘Where to be born?’ which concluded that, given the current state of knowledge, ‘there is no evidence to support the claim that the safest policy is for all women to give birth in hospital’ (Macfarlane et al 2000). Despite this strong evidence to the contrary, the rhetoric that began over 100 years ago persists, and society still views birth at home (everywhere except the Netherlands) as a poor choice. Indeed, Tew (1995) identified that data had been deliberately misinterpreted in UK studies between 1958 and 1970 to support the claim that ‘the family home is the most dangerous place for birth’ (p 29). Tew stated that an impartial observer could clearly see that the perinatal mortality rate was higher in hospitals, yet this fact was distorted in reports of the time. Obstetricians throughout the world used the false interpretation of these statistics to influence the future development of maternity services.

The false interpretation of current home-birth statistics continues to influence the current development of maternity services, particularly in Australia. A nationwide review of maternity services undertaken by the Australian government in 2009 recommended that midwives be provided with a greater role in maternity care provision throughout the country. This included referral and prescribing rights and access to the federally provided funding arrangements for healthcare provision. This was of course met with much opposition from the medical professions, including obstetricians who were accused of ‘cherry-picking’ evidence to support their case (Community Affairs Legislation Committee 2009). Midwives may also be influenced by the rhetoric and either refuse to provide a home-birth option for women or unconsciously bias the way it is discussed, leading women to ‘choose’ a hospital birth (Walsh 2004).

Reflective exercise

Imagine that you are a midwife who cares for women wherever they want to give birth (indeed, this might be a reality for many midwives). Some women want to be at home, others want to be at the small, local birth unit, and still others want to be in the hospital. Ask yourself the following questions.

1. How do I feel about supporting women in each of these locations?

2. Are the feelings different depending on the location?

3. Where do these feelings come from?

4. How do I react when a woman asks me to support her in planning a home birth?

5. How do others react to her decision for a home birth?

6. Why is this so?

7. How inconvenient/uncomfortable/scary is it for me to have to travel to the woman’s home? Do I have to support her choice?

8. How inconvenient is it for me to have to travel to the birth centre rather than the hospital?

9. Do we (women and midwives) have any real choices of where to birth in our region?

• Is there a need to change anything?

• How can I do that?

• Who will help me?

The birth centre: a halfway house?

Moves to address the loss of the familiar home environment for birth appeared in the late 1970s with the call for more humanised or home-like birth spaces, culminating in the development of the birth centre in many locations throughout Australia and New Zealand (as elsewhere). Birth centres may be freestanding or located within hospitals, either adjacent to or within high-technology and medically staffed labour wards, thus enabling immediate consultation or rapid transfer if the need arises. Primary birth units may be located in urban or rural settings and share many of the attributes of birth centres. Although birth centres may differ in their structure, location, furnishings and staffing, all share a strong philosophical orientation towards assisting women to achieve normal physiological birth (Coyle et al 2001; Kirkham 2003). They are intended only for women classified as ‘low risk’, the very women who would fulfil criteria for birth at home. However, even in this low-risk population, numbers of women are transferred out for medical assistance or pain-relieving drugs before, during or after labour. Transfer rates vary from as low as 12% (Rooks et al 1989) to more commonly around 20% but even up to 63% in some settings (Hodnett et al 2005). In Australia, around 5% of women give birth in primary birth units (Griew 2003). This includes 5379 births in birth centres in 2003, ‘representing 2.1% of all confinements’ (Laws & Sullivan 2005). Unfortunately, Australian Institute for Health and Welfare (AIHW) data for 2005 does not differentiate by birth centre or primary unit but groups these together with hospital births. In New Zealand the combined birth centre/primary birth unit rate in 2001 was 1% (Pairman & Guilliland 2003) and in 2005 was 16.3% (NZHIS 2008).

Numerous randomised controlled trials and observational studies examining the effectiveness and safety of birth centres/primary birth units have been conducted worldwide, and have demonstrated that birth for low-risk women is at least as safe in small low-risk maternity units as it is in hospitals (Kirkham 2003; Walsh & Downe 2004). In addition, consumer satisfaction surveys conducted in New Zealand have revealed that women ‘are more likely to be satisfied with maternity services if they birth at a primary maternity facility’ (Ministry of Health 2002, p 4). Unfortunately the 2007 survey did not repeat this question (HSCR 2008). In a Cochrane Systematic Review, Hodnett et al (2005) included evidence from six trials involving almost 9000 women, and concluded that there appear to be some benefits from home-like settings for birth. However, one study has raised concerns about the safety of out-of-hospital births, and it needs to be considered here (Gottvall et al 2004).

A 10-year retrospective review of the Stockholm Birth Centre undertaken by Gottvall et al (2004) revealed a trend to higher perinatal mortality in primigravid women.⁴ Following scrutiny of each perinatal death by an obstetrician, Gottvall and colleagues (2004) claimed that a potential risk of birth-centre care is the ‘philosophy that emphasises a strong belief in the natural process’ (p 77). Hodnett et al (2005) echoed concerns regarding the emphasis on normal birth and stated that this belief might delay recognition of imminent complications or the ability of the midwife to take averting action. However, concluding comments from Gottvall et al (2004) indicate a potential bias in the study introduced through an underlying concern with intrapartum care that does not include technology and medical assistance. Many of the perinatal deaths in the original Stockholm trial (Waldenstrom et al 1997) occurred after transfer and were associated with clearly documented suboptimal care in the receiving hospital. Gottvall and colleagues did not comment on this.On the other hand, Walsh (2004) refuted any suggestion that birth-centre midwives are over-orientated to normal birth and therefore may delay recognition of complications. Walsh asserts that birth-centre midwives are highly skilled practitioners with an astute awareness of normal labour and that these midwives are particularly diligent in updating their skills in emergency care. Walsh acknowledged that it might very well be the midwife’s belief in physiological labour, especially for primigravid women, that enables such women to achieve normal birth in birth centres. In his most recent ethnographic study, Walsh (2006) asserts that very little attention has been paid to organisational dimensions of childbirth care until recently. He claims that freestanding birth centres ‘subvert’ the processing mentality of modernist organisations such as large maternity hospitals. Birth centres demonstrate greater flexibility in labour care so as to accommodate women’s preferences and enhance the autonomy of midwives and women, such that a self-managing and self-regulating ethos flourishes (Walsh 2006).

Sociologists who have paid particular attention to childbirth also show concern regarding interpretations of safety. Annandale (1988) used both quantitative and qualitative methods to study the structure of birth in a North American birthing centre. Her study included 18 months of observation, repeated focus-group interviews and content analysis of 900 women’s records over a five-year period. Obstetricians did not see women unless a risk factor arose; however, Annandale commented that midwives and obstetricians disagreed about what constituted a risk factor. Midwives tended to disagree with the assertion that a post-term induction was high risk, and that intervention was required after 12 hours of ruptured membranes. Annandale found that birth-centre midwives adopted strategies to maintain the ‘normal’, such as encouraging women to stay at home until active labour was well established. This strategy reduced the likelihood of transfer to a large hospital for perceived prolonged labour.

Reflective exercise

Think about the place of birth in which you work and ask yourself the following questions. (It may be helpful to write down your responses.)