Over the course of the last century, Australia made significant progress in improving the safety of pregnancy and childbirth. In 1936, there were 600 maternal deaths per 100,000 live births. By 1950, this had dropped to 109 per 100,000 live births and, by 1980, this figure had dropped to below 10 per 100,000 live births.7
Figure 1 shows that on average since 1990 there have been between 10 and 15 deaths directly related to complications in pregnancy in Australia each year, at a time when the number of births is increasing. For its part, the AIHW reported that 29 direct maternal deaths and 36 indirect deaths occurred in the three years 2002 to 2005.8
Figure 1: Number of women who die in childbirth, Australia, 1991–2006
Source: Australian Institute of Health and Welfare (AIHW) National Perinatal Statistics Unit, Australia’s mothers and babies (various), Cat PER 46.
Figure 2 shows the number of fetal/stillborn deaths (from 20 weeks gestation, minimum weight 400 g), neonatal deaths (to 28 days) and infant deaths (to 1 year) in Australia between 1966 and 2006. The significant shift depicted in the figure is for the number of neonatal deaths, which has fallen from 3,364 a year to 864 over the 40-year period.
Figure 2: Number of infant deaths to one year of age, Australia, 1966–2006, 5-yearly
Source: Australian Bureau of Statistics, Year Book Australia (various), Cat 1301.0.
Australia is one of the safest countries in the world in which to give birth or to be born. Data from the OECD shows that over the past decade Australia has had consistently lower maternal 9 and perinatal 10 death rates than the majority of comparable countries. 11
Figure 3: Perinatal mortality, international comparison, 1996–2006
Source: Organisation for Economic Co-operation and Development, 2008, Health Data 2008.
Figure 4: Maternal mortality, international comparison, 1996–2006
Source: Organisation for Economic Co-operation and Development, 2008, Health Data 2008. Note: The OECD maternal mortality for Australia is based on the Australian Bureau of Statistics’ causes of death data, whereas Figure 1 presents mode of separation from hospital discharge data from the AIHW mothers and babies reports.
At the same time, Australia has a high rate of caesarean section (31 per cent of births in 2006,12 compared with the 2004 OECD average of 22 per cent of births).13
There is also some evidence to show variations in caesarean section rates between public and private sectors (see Figure 5) and between states and territories (ranging from 26.9 per cent in Tasmania to 33.2 per cent in Queensland).14 Data relating to the number of caesarean sections performed by particular service providers (public as well as private) is generally not publicly available, and so variations in the rates at which caesarean sections are performed is not available. Some submissions from individuals to the Review requested that specific hospital caesarean section rates be published.
Figure 5: Rates of caesarean section by hospital sector, Australia, 1991–2006
Source: Australian Institute of Health and Welfare (AIHW) National Perinatal Statistics Unit, 2008, Australia’s mothers and babies, 2006, Perinatal statistics series no. 22, Cat. no. PER 46, Sydney.
Figure 5 shows a steady growth in the number of births by caesarean section in both private hospitals (increasing from 22 per cent of all births in 1991 to 41 per cent in 2006) and public hospitals (rising from 16 per cent in 1991 to 28 per cent to 2006). The increasing rate of caesarean sections has been similar for both private and public hospitals over this period.
Figure 6 shows the variation in percentage of caesarean births by hospital in New South Wales in 2005, for hospitals with more than 200 births a year. Private hospitals are shown in light blue. 15
Figure 6: Rates of caesarean section, by hospital, New South Wales, 2005
Source: Centre for Epidemiology and Research, NSW Department of Health, 2007, New South Wales Mothers and Babies 2005, NSW Public Health Bulletin 2007; 18(S-1), pp. 95–6, available http://www.health.nsw.gov.au/pubs/2007/pdf/mdc05.pdf
What the Review Team Heard
- Almost without exception, the priority issue for review participants was that of ensuring Australian women and their babies have access to safe, high-quality maternity services.
- Many contributors to the Review considered that maternal and perinatal rates of mortality were not an adequate measure of the performance and outcomes of maternity services. Severe maternal and perinatal morbidity were identified as important indicators of system performance. There is clearly a group of women, the numbers of whom are unknown, who continue to have short- or long-term sequelae from their pregnancy and delivery. The limited consideration given to stillbirths and understanding of their causes was also highlighted to the Review.
- Intervention rates were identified as another key measure of system performance. While a number of Review respondents saw intervention rates as directly contributing to low maternal and perinatal mortality, this viewpoint was by no means universal. Several submissions referred to the body of scientific evidence, much of which has been published in recent years, to suggest that caesarean section increases the risk to both mother and baby in the index pregnancy, and in subsequent pregnancies.
- Numerous submissions from consumers and some health professionals referred to the ‘cascade of intervention’—the pattern in which interventions in labour are likely, in and of themselves, to increase the need for interventions during the birth and in subsequent pregnancies. At the same time, it was highlighted that observed increases in caesarean section rates could, at least in part, be explained by a range of factors. These include increasing maternal age; increasing co-morbidities such as obesity, diabetes and hypertension; changes in care for preterm deliveries and those involving assisted reproductive technologies (ART); consumer choice and demand; medico-legal risks; and defensive practice.
- Numerous submissions and discussion at the forums advocated improved national data collection, analysis and review, particularly in the areas of maternal and perinatal mortality and morbidity. The need to consider women’s experiences and perceptions of maternity services was also highlighted, along with the need to carefully consider governance arrangements for improved data collection and review.
- Also highlighted to the Review were the disparities in practice between different parts of the system and between individual institutions. Submissions and participants at the forums identified the need for a nationally agreed, consistent and standardised minimum dataset that could provide an evidence-based platform upon which a national benchmarking program for maternity services could be built.
Maternity services need to be regularly audited, hence the importance of adequate data collection and clinical indicators. This is vital particularly with the introduction of any change involving alternative workforce models to assess their effectiveness and establish whether or not there have been improvements.16
- Consumers and some health professionals emphasised the importance of going beyond a largely bio-medical approach to consider a range of other factors in assessing risk, quality and safety of maternity services.
Women need to have access to maternity services that are appropriate to their clinical, cultural and social needs … a strict biomedical approach is unlikely to adequately reflect or accommodate the broader health picture for women.17
Recent Related Initiatives
At the Council of Australian Governments (COAG) meeting on 29 November 2008, it was agreed that the National Healthcare Agreement include performance indicators relevant to maternity care in three areas: proportion of babies with low birth weight; infant and young child mortality rates (including the gap between Indigenous and non-Indigenous); and teenage birth rate.
The Australian Commission on Safety and Quality in Healthcare, through Women’s Hospitals Australasia, initiated a project in 2005 to develop a core set of evidence-based performance indicators for timely comparative analysis of practice and outcomes in maternity care. This project has recently been referred to the Maternity Services Inter-Jurisdictional Committee (MSIJC) subcommittee of the Australian Health Ministers’ Advisory Council (AHMAC), which, at the time of writing, is considering the next steps for this project.
The Review Team acknowledge that Australia’s record of safety and quality in relation to maternity care as measured by mortality alone is an enviable one and that significant improvements in this regard are therefore not likely. The disparity in outcomes experienced by Australian women is discussed in more detail in Chapter 3: Inequality of Outcomes and Access.
However, there are areas where gaps in our knowledge are resulting in discord among the professions and consumers rather than providing useful feedback to inform both policy and practice.
While there is data available on Australia’s maternal and perinatal mortality rates, nationally consistent data and reporting are limited. While most jurisdictions produce annual reports of perinatal statistics, the comprehensiveness and level of detail differ markedly.
As identified above, alongside mortality, severe maternal morbidity18 and stillbirths were other areas identified to the Review where better data collection, analysis and review were needed to inform research, practice and policy. There is currently no national reporting of maternal morbidity, no national dataset and no nationally agreed definition.19 The differences in definitions used for stillbirth across jurisdictions and between the Australian Bureau of Statistics (ABS) and the AIHW, alongside the limited understanding and magnitude of the problem (around 300 stillbirths per year at full term), were highlighted as an area where greater standardisation of data and improved reporting are needed to aid national research and understanding. 20
Most topical, is probably the debate among the professions and consumers about reasons for the current rates of caesarean sections and the impact of caesareans on women, babies and the health system, including the impact on available resources. State perinatal statistics for 2005–06 showed that the proportion of caesarean births that were either ‘elective’, ‘planned’ or did not involve labour varied from 46 per cent in Queensland to 58 per cent in New South Wales. While caesarean section is considered to have a high degree of safety and to contribute to our low levels of mortality, there is also a view that defensive practice is resulting in higher-than-desirable rates of intervention. Caesarean section as a surgical procedure is accompanied by the additional risks associated with any surgery—for example, infection. Furthermore, observational studies have identified the following risks associated with caesarean sections: increased maternal morbidity or mortality, adverse psychological sequelae, negative implications for future fertility, and problems in subsequent pregnancies.21 22
Determining the appropriate approach to caesareans and other interventions in a way that takes account of both the medical safety of mother and baby and the wishes of mothers is a complex issue, where a more informed discussion and debate is required. Women with normal pregnancy also highlighted to the Review the range of interventions they experienced (including repeat ultrasound, electronic fetal heart rate monitoring on admission in labour and induction) that they believed were of no proven benefit. 23
A number of submissions highlighted the need for national guidelines for maternity care (see Chapter 2) and for improved information to assist women in making informed choices (see Chapter 4). Robust, comprehensive data collection, reporting and review—along with targeted research in key priority areas—are necessary in order to monitor and inform more effectively the performance of maternity services. In addition, a number of important areas—including caesarean section, stillbirths, women’s experiences and perceptions of different models of care, and effective models of postnatal care—were identified as lacking evidence and requiring targeted research. It is vital to maintain Australia’s good safety outcomes in the possible future implementation of a range of models of maternity care.
The Review Team concluded that:
- Australia’s strong record of safety in maternity services is an acknowledged strength of our maternity system.
- Changes to maternity services need to be guided by evidence.
- Stable, ongoing arrangements for national maternity data collection, analysis and review must be a priority.
- That the Australian Government, in consultation with states and territories and key stakeholders, agree and implement arrangements for consistent, comprehensive national data collection, monitoring and review, for maternal and perinatal mortality and morbidity.
- That the Australian Government, in consultation with states and territories and key stakeholders, initiate targeted research aimed at improving the quality and safety of maternity services in select key priority areas, such as evidence around interventions, particularly caesarean sections, and maternal experience and outcomes, including from postnatal care.