SEARCH

SEARCH BY CITATION

Keywords:

  • maternal health;
  • performance indicators;
  • severe maternal morbidity

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Severe maternal morbidity
  5. Discussion
  6. Conclusion
  7. References

Maternal mortality has traditionally been the key element in the monitoring of maternal health and adequacy of obstetric services in Australia and around the world. In developed countries, the ability of maternal mortality to serve this purpose is reduced because of the rarity of maternal mortality, reflected in very low maternal mortality ratios. Internationally, there has been increasing interest in severe maternal morbidity as an indicator to monitor maternal health and maternity services. The aim of this paper is to critically examine the capacity to measure and monitor maternal morbidity in Australia. There is a paucity of reliable maternal morbidity data in Australia; Australia is lagging behind peer countries that are endeavouring to monitor severe maternal morbidity. Dedicated efforts and adequate resources are needed in order to monitor severe maternal morbidity in Australia.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Severe maternal morbidity
  5. Discussion
  6. Conclusion
  7. References

The organisation of maternity care in Australia has undergone restructure with a number of models of antenatal care now available to pregnant women. The implementation of multiple models of midwife-led care including midwives clinics, team midwifery and midwife supported homebirths has broadened the options of women in major Australian cities, from existing models of care delivered through standard public antenatal services and private obstetrician care.1 Shared care, where most antenatal care is provided by general practitioners who have little or no involvement in the birth admission, is also on the rise.2 In comparison, both access to and options for maternity care for pregnant women living in rural and remote areas have been adversely affected by the closure of many rural birthing facilities.3 Change is being driven by the constraints of the obstetric workforce, government funding and institutional reorganisation as well as consumer and practitioner advocacy over choice in birthing.

In addition to the organisational changes in maternity care, there have been major changes in the demographic and biological characteristics of women having babies. Women in Australia as in other developed countries are choosing to birth later and are having fewer children.4 Data in Table 1 show some of the major reproductive and obstetric trends in Australia evident over recent years. The increasing age of birthing women, increasing incidence of morbid obesity, increasing use of assisted reproductive technology (ART), increased multiple births and increased interventions during birth all pose an increased risk to maternal health. For example, women who conceive by ART are more likely to experience severe maternal morbidity, such as ovarian hyperstimulation syndrome,8 ectopic pregnancy9 and pre-eclampsia.10

Table 1.  Summary of changes in reproductive and obstetric characteristics
Descriptor1994520046
  • 1994 and 2003.7

  • ART, assisted reproduction technology.

Mean age of women having their first child (years)26.328.0
Mean age of all women birthing (years)28.329.7
% of all birthing women who are ≥ 35 years12.619.6
% of primiparas ≥ 35 years6.712.5
Induction rate21.5%25.3%
Spontaneous vaginal birth rate68.0%59.2%
Caesarean birth rate19.4%29.4%
Multiple pregnancy rate1.4%1.7%
Number of ART cycles in Australia~19 00036 040
Live neonates born as a result of ART 2 318 6 026

Maternal deaths are so rare in Australia that the sensitivity of the summary measure, the maternal mortality ratio (MMR) as the only index of maternal wellness, is now inadequate. This is evident by a stable MMR over the past 20 years ranging from eight to 12 per 100 000 confinements, despite major changes in reproductive characteristics and in the organisation and delivery of maternity services. Consequently, low and static MMRs in developed countries demonstrate the requirement to measure severe maternal morbidity as a main indicator of maternal health and as a quality indicator for obstetric services.11

Severe maternal morbidity

  1. Top of page
  2. Abstract
  3. Introduction
  4. Severe maternal morbidity
  5. Discussion
  6. Conclusion
  7. References

While there is widespread international support for the monitoring of severe maternal morbidity in developed countries, there is no uniform definition of severe maternal morbidity. Severe maternal morbidity has been variously reported as near-miss maternal mortality12 or severe acute maternal morbidity (SAMM)13 and there are multiple definitions in use.14,15 A general understanding of severe maternal morbidity includes potential life-threatening events during pregnancy and the post-partum, including all serious pregnancy complications and diseases and the outcomes of such events. ‘Maternal outcome’ may refer to a management outcome such as red blood cell transfusion or admission to an intensive care unit (ICU) or relate to organ failure for example acute renal failure.16 Diagnosis-based criteria, for example severe pre-eclampsia, have also been used.17

A number of countries have recognised the need to monitor severe maternal morbidity and invested dedicated resources towards ascertainment and reporting. Canada, Scotland, UK, South Africa, Europe and the USA have each implemented a program to examine aspects of severe maternal morbidity. While the programs differ between these countries with limitations as to the methods used, Australia has not undertaken any systematic approach to monitoring severe maternal morbidity. Table 2 demonstrates both the approach by other countries to measure severe maternal morbidity and the magnitude of severe maternal morbidity itself.

Table 2.  Summary of international programs to monitor severe maternal morbidity
Country or regionMethodDefinitionSummary of resultsStatusFunding body
  • *

    Third or fourth degree tear, post-partum haemorrhage – data did not report on ‘severe morbidity’ as a category.

Canada18Administrative data analysisICD and procedure codesSMMR 4.8 per 1000 deliveriesOngoingHealth Canada
USA19Administrative data analysisICD and procedure codes28.6% of women who gave birth had at least one obstetric complication*One-off studyUS Centres for Disease Control and Prevention
Pretoria (South Africa)20Clinical auditManagement-based and organ failure-basedSMMR 10.2 per 1000 birthsOngoingSouth Africa Medical Research Council
Scotland14Clinical auditManagement-based and organ failure-basedSMMR 3.8 per 1000 deliveriesOngoingNHS Quality Improvement Scotland
Europe (including UK)15Clinical auditDiagnosis-basedSMMR 9.5 per 1000 deliveriesOne-off studyEuropean Union
UK Obstetric Surveillance System21Clinical auditDiagnosis-basedCases reported following one year of data collectionPeripartum hysterectomy (n = 315)Eclampsia (n = 209)Antenatal pulmonary embolus (n = 94)Ongoing for specific conditions for pre-specified timeframesNational Perinatal Epidemiology Unit Researchers pay to use system which contributes to funds
ICD, International Classification of Diseases; SMMR, severe maternal morbidity ratio – each jurisdiction uses different definitions.

Severe maternal morbidity mainly occurs in young and previously healthy women.18 It occurs relatively frequently and the incidence is increasing. A UK prospective study found that 1.2% of all pregnant women (≥ 24 weeks gestation) experienced at least one of the following four serious complications: severe haemorrhage, severe pre-eclampsia, sepsis or uterine rupture.17 A Victorian record-linkage study found that in 2002, hysterectomy associated with post-partum haemorrhage occurred nearly three times more frequently than in 1999.22 Whether this rise reflected a change in morbidity or was a one-off increase is not known as the linkage has not been repeated. Likewise, the numbers of women who have eclampsia, require admission to ICU, receive a blood transfusion to treat haemorrhage along with all other examples of severe maternal morbidity remain unknown in Australia. Given the urgent need to monitor severe maternal morbidity, it is timely to review the potential of existing data sources that may be used in order to measure severe maternal morbidity in Australia.

Potential data sources in Australia

There are three prominent data sources in Australia that may serve to identify cases of severe maternal morbidity. The National Perinatal Data Collection (NPDC), the National Hospital Morbidity Dataset (NHMD) and the Australian and New Zealand Intensive Care Society (ANZICS) adult patient database were each critically reviewed to establish the capacity to measure severe maternal morbidity. The strengths and limitations of each data source are summarised in Table 3. Additionally, clinical performance indicators are an evolving option to monitor severe maternal morbidity and an appraisal of maternity performance indicators in Australia is presented.

Table 3.  Potential data sources to measure severe maternal morbidity nationally
Existing data sourceStrengthsLimitations
  1. ANZICS, Australian and New Zealand Intensive Care Society; ICU, intensive care unit.

National Perinatal Data Collection23• Agreed national perinatal minimum dataset• ≥ 20 weeks gestation or > 400 g birthweight only
• Reliable data on labour and neonatal outcomes• Lack of clinical and data definitions for maternal morbidity
• All births represented once• Maternal and neonatal data are recorded together for each pregnancy• Low sensitivity and specificity for direct obstetric complications
• Only severe maternal morbidity events that occur during the hospital admission for birth are captured
 • No severity of illness stratification
National Hospital Morbidity Dataset24• Agreed national minimum hospital morbidity dataset• Not developed for perinatal surveillance purposes
• Inexpensive and readily available• Data are generated for each hospital separation, with individuals potentially represented multiple times during the one pregnancy
 • Low sensitivity and specificity for severe maternal morbidity events
 • Poor discrimination of severity of disease
 • Pregnancy status for women with a non-obstetric cause of admission dependent on the relevant Z code being recorded
 • Maternal and neonatal data are recorded individually with no link
ANZICS adult patient database25• Extensive data available about the acute event; including severity of illness scoring• ICU admissions only
• Pregnant and postnatal women identifiable in the database by a pregnancy status data item• Voluntary and hence incomplete source of population-based data
 • No obstetric data recorded, eg, gestation
 • No neonatal data recorded

National Perinatal Data Collection

All states and territories in Australia maintain perinatal or midwives collections. These databases collect epidemiological data about pregnancy, labour and to a limited extent the puerperium. Data are collected for every birth in Australia that occurs ≥ 20 weeks gestation or at least 400 g birthweight. An extract of these data, the minimum perinatal dataset about each birth, is provided to the Australian Institute of Health and Welfare (AIHW) National Perinatal Statistics Unit (NPSU) to form the NPDC. These data are annually reported by the NPSU in the ‘Australia Mothers and Babies’ report.4 These reports have been useful in observing the changes to some aspects of the child-bearing population in Australia over recent years, demonstrating increasing maternal age, increasing multiple births and increasing caesarean section rates. Monitoring of postdelivery maternal health is limited to length of stay and discharge status.

The current purpose of the perinatal collections is to collect data about women who give birth ≥ 20 weeks gestation, therefore any conditions or interventions that result in abortion, ectopic pregnancy, post-abortion complications or other severe maternal morbidity such as malignancy or pulmonary hypertension prior to 20 weeks gestation are not included. Furthermore, because the data are collected at the hospital at the time of birth, there is also the potential to miss other cases of interest, that present to either the same hospital or to a different hospital following discharge after the birth such as peripartum cardiomyopathy, secondary post-partum haemorrhage (PPH) or deep-vein thrombosis. Consequently, the perinatal datasets are a useful source of data for obstetric and neonatal outcome, but offer an incomplete resource for maternal morbidity ascertainment.

Additionally, there is a lack of nationally agreed clinical and data definitions for particular morbid conditions to monitor maternal morbidity through the NPDC. This is demonstrated by a number of data validation studies that have shown poor reliability on items that warrant inclusion in the monitoring of severe maternal morbidity. In particular, obstetric conditions like eclampsia, PPH and placental abruption are all significantly underreported in the data.26–28

National Hospital Morbidity Database

All states and territories in Australia maintain minimum hospital morbidity data for each hospital separation which are then provided to the AIHW to form the National Hospital Morbidity Database (NHMD). This is a routine data collection of hospital separations that uses the International Classification of Diseases Australian Modified version 10 (ICD-AM-10) to code diagnoses and procedures.24 These data are routinely collected and may be an inexpensive resource for monitoring maternal morbidity.27 There are a number of limitations as the NHMD is not an epidemiological dataset designed for use in the study of pregnancy-related conditions.29 There are four major concerns with using the NMHD as a source to extract data on severe maternal morbidity at the national level: poor sensitivity and specificity for conditions associated with maternal morbidity, each dataset represents a hospital separation rather than an individual, non-obstetric conditions are not reliably identified in the dataset and there is poor discrimination of the severity of disease.

The quality of data in the state and territory hospital morbidity collections varies. Victorian studies in the 1990s found that coding errors were common especially in more complex cases30 and that less than half of the cases coded as eclampsia, were so confirmed in the medical notes.26 More recent studies in NSW comparing the sensitivity and specificity of the ICD-AM codes in administrative datasets with individual medical histories have found that only 33% of women with hypertensive disorders of pregnancy were coded correctly,28 and low sensitivity for ICD-AM codes for specific obstetric diagnoses such as pre-eclampsia (50%) and PPH (59%), respectively.27 While ICD-AM-10 coding data are readily available, the quality of the data is a valid concern.

The second major limitation of the ICD-10 coding system is that retrievable data represent an episode of care rather than an episode of illness. Each hospital discharge (separation) generates data that are lodged with the local health department. Women with multiple admissions during a pregnancy for hypertension or antepartum haemorrhage and women who are transferred between hospitals during the same episode of illness are represented in the dataset multiple times. The current practice to overcome this limitation is restricting analysis to cases that coincide with a birth.18,31 Until a woman's single episode of illness can be represented or identified in the data as one event even over multiple hospital sites and admissions, the ICD-AM-10 codes and the NMHD will be limited in its capacity to measure and monitor maternal morbidity. Cases that do not occur in conjunction with, and in the same hospital as birth are presently excluded, for example, transfer following birth in one hospital to another to access an ICU bed.

The ICD-AM-10 Z codes, which identify pregnancy as a comorbidity, are not uniformly coded in cases where women require a hospital admission for a non-obstetric condition. Similarly, this limits the examination of severe maternal morbidity to cases where the hospital admission coincided with a birth and potentially excludes cases. For example, pregnant women admitted to hospital with pneumonia and discharged pregnant are not reliably identifiable in the NHMD which limits the ability to examine the interaction of acute illness (pneumonia) in previously healthy pregnant women.

The fourth major limitation of the ICD-AM-10 coding system is the poor discrimination of severity of disease in the coding categories. For example there is a single ICD-AM-10 code for PPH. Women with a 600-mL haemorrhage are allocated the same code as those who have a haemorrhage of 3500 mL. Revision of the ICD-AM-10 coding system would be necessary to enable identification of cases with severe maternal morbidity.

ANZICS adult patient database

The ANZICS maintains an adult database for ICU admissions. This is a voluntary database with approximately 65% of adult ICU admissions in Australia and New Zealand entered into the database.32 Because of its voluntary nature, its use is limited as a means to obtain population-based data on the obstetric utilisation of ICU services. A welcome addition to this database in 2007 is a pregnancy status data item. It is anticipated that the new data item will overcome the previous failure to identify pregnant and postnatal women in the dataset. Identification of parturients in the dataset offers an opportunity to examine extensive details about ICU utilisation by pregnant and postnatal women in Australia. However, admission to ICU as a measure of severe maternal morbidity would only capture about one third of cases of interest, with most cases of severe maternal morbidity not involving an ICU admission.14 A further limitation is that obstetric data like gestation and neonatal outcome are not recorded in this dataset.

Performance and clinical indicators

Performance indicators are used increasingly to monitor the provision of health services and health outcomes. Performance indicators could prove useful to monitor severe maternal morbidity, but to date there has been limited development in this area. For example, Victoria has developed a number of maternity service performance indicators as a mechanism to improve the monitoring and quality of maternity services in the state.33 Maternal morbidity was deemed by the committee to be too complex and was not endorsed as a performance indicator to be developed. Additionally, the current version of the Australian Council on Healthcare Standards obstetric indicators (version 5) does not contain any indicators relating to severe maternal morbidity.34 The AIHW introduced a ‘maternal morbidity rate’ as part of its reproductive health indicator set, which they defined as the ‘number of women experiencing severe maternal morbidity per 1000 confinements’29 (p. 117). The ‘maternal morbidity rate’ was unable to be reported due to the lack of uniform definitions of maternal conditions, and the incomplete and unreliable state of the available data; the NHMD and ICD-AM-10 codes and their associated limitations.

Women's Hospitals Australasia, a not-for-profit association, advocates for the health-care needs of women and babies in Australia and New Zealand. This association has also been interested in the application of clinical indicators to monitor maternal health. They have developed clinical indicators including post-partum haemorrhage, blood transfusion, admission to ICU and peripartum hysterectomy. While these indicators are highly clinically relevant and offer potential for improved understanding of maternal morbidity, their use to monitor severe maternal morbidity is limited by the small number of participating hospitals, with 21 Australian hospitals listed as members.35 In addition, the clinical indicators are used for benchmarking by member hospitals, with the benchmarking reports only accessible by member hospitals and staff. Furthermore, no verification procedures on the dataset have been published.

A major problem exists with performance indicators and their application to monitor severe maternal morbidity, as the source of data for many indicators is the ICD-AM-10 codes and procedures recorded in administrative datasets. Thus, many population-based performance and clinical indicators suffer the same limitations as the administrative datasets discussed earlier.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Severe maternal morbidity
  5. Discussion
  6. Conclusion
  7. References

It is evident that the surveillance of severe maternal morbidity is now internationally accepted as a core component in programs of maternal health.36,37 Many of the international programs suffer from the same limitations in data quality as seen in Australia; nevertheless, these countries have assigned resources to monitor severe maternal morbidity. Australia has made little progress in this area to date.

There is no current capacity in routinely available data to reliably capture events of severe maternal morbidity, the incidence of which is increasing.18,22,38 Furthermore, Wen et al. assert that service factors and not demographics may be more responsible for the observed increase in the incidence of severe maternal morbidity in Canada, as controlling for age, multiple pregnancy and prior caesarean section did not impact on the incidence of severe maternal morbidity.18 Likewise, the presence of chronic health conditions only slightly increased the risk of severe maternal morbidity. The work in the USA by Geller et al.39 is further evidence that health provider factors are associated with the occurrence of severe maternal morbidity and that much severe maternal morbidity is preventable by changes in patient management. Consequently, a capacity to monitor severe maternal morbidity in Australia is required as a quality of care indicator. The factors that have a negative impact on maternal health are likely to be responsive to public policy intervention.

There is a poor understanding of the interaction of acute illness and pregnancy, and the resultant impact on the pregnancy because of illness. Expansion of maternal mortality reviews to include ‘incidental’ causes of death has enabled identification of conditions/circumstances that were previously deemed ‘incidental’ but are now considered to be influenced by or to influence the outcome of pregnancy; for example domestic violence, psychiatric causes and some malignancies.40 It is likely that similar ‘discoveries’ would occur if it were possible to reliably identify all cases of severe illness during pregnancy. Without the capacity to monitor specific diseases and complications of pregnancy, there is a void in data that could be used to plan, provide and coordinate appropriate health services and to guide clinical decision-making.

The economic impact of severe maternal morbidity needs to be considered. There is insufficient information to calculate the costs of the acute inpatient episode, let alone the long-term burden on society that results from severe maternal morbidity.

There have been attempts to focus on severe maternal morbidity in Australia by isolated groups of researchers.41,42 Severe maternal morbidity is a complex issue, not least because of the difficulty in obtaining accurate and complete ascertainment of cases as highlighted in this paper. A national coordinated response is necessary to identify what approach to severe maternal monitoring should occur in Australia, along with what resources are required to achieve it. There is a pressing need to develop capacity to monitor severe maternal morbidity in Australia, given the mounting evidence that the incidence of severe maternal morbidity is increasing in developed countries, including Australia.18,20,43

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Severe maternal morbidity
  5. Discussion
  6. Conclusion
  7. References

There is Australian and international consensus that the review of maternal health must be extended to include severe maternal morbidity.40,44 Yet accurate information on acute illness during pregnancy and childbirth is not routinely available in population-based data collections, and there has been a lack of investment in the development of capacity to develop measures to monitor severe maternal morbidity in Australia. ‘Sound information is the prerequisite for health action: without data on the dimensions, impact and significance of a health problem it is neither possible to create an advocacy case nor to establish strong programmes for addressing it’45(p. 1). The need to monitor severe maternal morbidity has not yet gained the prominence it deserves in Australia. Only with the establishment of a dedicated severe maternal morbidity monitoring system will we be able to evaluate the impact of changing reproductive trends and maternity service models on women's health. A significant commitment of resources by our health organisations and research and public policy funding agencies is necessary before the measurement and monitoring of severe maternal morbidity become a reality in Australia.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Severe maternal morbidity
  5. Discussion
  6. Conclusion
  7. References
  • 1
    Department of Human Services. Future Directions for Victoria's Maternity Services. Melbourne, Vic.: Programs Branch, Metropolitan Health and Aged Care Services, Victorian Government Department of Human Services, 2004.
  • 2
    Brown S, Darcy M-A, Bruinsma F. Having a baby in Victoria 1989–2000: Continuity and change in the decade following the Victorian Ministerial Review of Birthing Services. Aust NZ J Public Health 2002; 26: 242250.
  • 3
    Rural Workforce Agency Victoria. RWAV Statewide Forum. Preserving Choice for Rural Women: Restoring Birthing Services in Country Victoria, Melbourne; 2005. Available at http://www.rwav.com.au/resources/publications/2005ObstetricsForum.pdf.
  • 4
    Laws PJ, Sullivan EA. Australia's Mothers and Babies 2003. (No. AIHW cat. no. PER 29.) Sydney, NSW: AIHW National Perinatal Statistics Unit (Perinatal Statistics Series No. 16), 2005.
  • 5
    Day P, Lancaster P, Huang J. Australia's Mothers and Babies 1994 (No. AIHW cat. no. PER 4). Sydney, NSW: AIHW National Perinatal Statistics Unit, 1997.
  • 6
    Laws PJ, Grayson N, Sullivan E. Australia Mothers and Babies 2004 (No. AIHW cat. no. PER 34). Sydney, NSW: AIHW National Perinatal Statistics Unit (Perinatal Statistics Series Number 18), 2006.
  • 7
    Waters A-M, Dean D, Sullivan E. Assisted Reproduction Technology in Australia and New Zealand 2003. (No. AIHW cat. no. PER 31.) Sydney, NSW: AIHW National Perinatal Statistics Unit (Assisted Reproduction Technology Series No. 9), 2006.
  • 8
    Orvieto R. Can we eliminate severe ovarian hyperstimulation syndrome? Human Reprod 2005; 20: 320322.
  • 9
    Karaer A, Avsar FA, Batioglu S. Risk factors for ectopic pregnancy: A case-control study. Aust N Z J Obstet Gynaecol 2006; 46: 521527.
  • 10
    Kallen B, Finnstrom O, Nygren KG, Olausson PO, Wennerholm UB. In vitro fertilisation in Sweden: Obstetric characteristics, maternal morbidity and mortality. Br J Obstet Gynaecol 2005; 112: 15291535.
  • 11
    Geller SE, Cox SM, Callaghan WM, Berg CJ. Morbidity and mortality in pregnancy: Laying the groundwork for safe motherhood. Womens Health Issues 2006; 16: 176188.
  • 12
    Fitzpatrick C, Halligan A, McKenna P, Coughlan BM, Darling MRN, Phelan D. Near miss maternal mortality (NMM). Irish Med J 1992; 85: 37.
  • 13
    Say L, Pattinson RC, Gulmezoglu AM. WHO systematic review of maternal morbidity and mortality: The prevalence of severe acute maternal morbidity (near miss). Reproduct Health2004; 1: 3. Available at http://www.reproductive-health-journal.com/.
  • 14
    Brace V, Penney G, Hall M. Quantifying severe maternal morbidity: A Scottish population study. Br J Obstet Gynaecol 2004; 111: 481484.
  • 15
    Zhang WH, Alexander S, Bouvier-Colle MH, Macfarlane A. Incidence of severe pre-eclampsia, postpartum haemorrhage and sepsis as a surrogate marker for severe maternal morbidity in a European population-based study: The MOMS-B survey. Br J Obstet Gynaecol 2005; 112: 8996.
  • 16
    Mantel GD, Buchmann E, Rees H, Pattinson RC. Severe acute maternal morbidity: A pilot study of a definition for a near-miss. Br J Obstet Gynaecol 1998; 105: 985990.
  • 17
    Waterstone M, Bewley S, Wolfe C. Incidence and predictors of severe obstetric morbidity: Case-control study. BMJ 2001; 322: 10891094.
  • 18
    Wen SW, Huang L, Liston R et al . Severe maternal morbidity in Canada, 1991–2001. CMAJ 2005; 173: 759764.
  • 19
    Danel I, Berg C, Johnson CH, Atrash H. Magnitude of maternal morbidity during labor and delivery: United States, 1993–97. Am J Public Health 2003; 93: 631634.
  • 20
    Pattinson RC, Macdonald AP, Backer F, Kleynhans M. Effect of audit on critically ill pregnant women. Clin Govern Int J 2006; 11: 278288.
  • 21
    Knight M, Kurinczuk J, Spark P, Brocklehurst P. United Kingdom Obstetric Surveillance System (UKOSS) Annual Report 2007. Oxford, UK: National Perinatal Epidemiology Unit, 2007.
  • 22
    Haynes K, Stone C, King J. Major conditions associated with childbirth in Australia: Obstetric haemorrhage and associated hysterectomy. Melbourne, Vic.: Department of Human Services, 2004.
  • 23
    Australian Institute of Health and Welfare (2007, 5 June 2007). National Perinatal Data Collection. [Retrieved 22 June 2007.] Available from http://www.npsu.unsw.edu.au/NPSUweb.nsf/page/NPDC.
  • 24
    Australian Institute of Health and Welfare (2007, 27 July 2006). National Hospital Morbidity Data Collection. [Retrieved 16 May 2007.] Available from http://www.aihw.gov.au/hospitals/nhm_database.cfm.
  • 25
  • 26
    Riley M, Halliday J. The accuracy of eclampsia cases reported to the Victorian Inpatient Minimum Database and the Perinatal Data Collection Unit. Health Inf Manag 1998; 28: 1315.
  • 27
    Taylor LK, Travis S, Pym M, Olive E, Henderson-Smart DJ. How useful are hospital morbidity data for monitoring conditions occurring in the perinatal period. Aust N Z J Obstet Gynaecol 2005; 45: 3641.
  • 28
    Thornton C, Makris A, Ogle R, Hennesy A. Generic obstetric database systems are unreliable for reporting the hypertensive disorders of pregnancy. Aust N Z J Obstet Gynaecol 2004; 44: 505509.
  • 29
    Ford J, Nassar N, Sullivan EA, Chambers G, Lancaster P. Reproductive Health Indicators Australia 2002. (No. AIHW cat. no. PER 20.) Sydney, NSW: AIHW National Perinatal Statistics Unit, 2003.
  • 30
    McIntyre CR, Ackland MJ, Chandraraj EJ, Pilla JE. Accuracy of ICD-9-CM codes in hospital data, Victoria: Implications for public health research. Aust N Z J Public Health 1997; 21: 477482.
  • 31
    Panchal S, Arria A, Harris A. Intensive care utilization during hospital admission for delivery: Prevalence, risk factors, and outcomes in a statewide population. Anesthesiology 2000; 92: 15371544.
  • 32
    Stow PJ, Hart GK, Higlett T et al . Development and implementation of a high-quality clinical database: The Australian and New Zealand intensive care society adult patient database. J Crit Care 2006; 21: 133141.
  • 33
    Kryriakopoulos V. Victorian Maternity Services Performance Indicators. Complete Set for 2003–04. Melbourne, Vic.: Department of Human Services, 2005.
  • 34
    The Australian Council on Healthcare Standards. ACHS Clinical Indicator Program Information 2007. Sydney, NSW: The Australian Council on Healthcare Standards, 2006.
  • 35
    Women's Hospitals Australasia. Annual Report 2003–04. Canberra, ACT: Women's Hospitals Australasia, 2005.
  • 36
    Health Canada. Special report on maternal mortality and severe morbidity in Canada – Enhanced surveillance: The Path to Prevention. Ottawa, ON: Minister of Public Works and Government Services Canada, 2004.
  • 37
    Penney G, Brace V. ‘Near misses’ and severe maternal morbidity: The Scottish experience. In: LewisG, ed. Why Mothers Die 2000–2002. Report on confidential enquiries into maternal deaths in the United Kingdom. London: Department of Health, 2004.
  • 38
    Vandecruys HIB, Pattinson RC, Macdonald AP, Mantel GD. Severe acute maternal morbidity and mortality in the Pretoria Academic Complex: Changing patterns over 4 years. Eur J Obstet Gynecol Reprod Biol 2002; 102: 610.
  • 39
    Geller S, Rosenberg D, Cox S et al . The continuum of maternal morbidity and mortality: Factors associated with severity. Am J Obstet Gynecol 2004; 191: 939944.
  • 40
    CEMACH. Why Mothers Die 2000–2002. Report on confidential enquiries into maternal deaths in the United Kingdom. London: Confidential Enquiries into Maternal and Child Health and the RCOG, 2004.
  • 41
    Heard AR, Dekker GA, Chan A, Jacobs DJ, Vreeburg SA, Priest KR. Hypertension during pregnancy in South Australia, Part 1: Pregnancy outcomes. Aust N Z J Obstet Gynaecol 2004; 44: 404409.
  • 42
    Roberts CL, Algert CS, Morris JM, Ford JB, Henderson-Smart DJ. Hypertensive disorders in pregnancy: A population-based study. MJA 2005; 182: 332335.
  • 43
    Cameron CA, Roberts CL, Olive EC, Ford JB, Fischer WE. Trends in postpartum haemorrhage. Aust N Z J Public Health 2006; 30: 151156.
  • 44
    Ford J, Sullivan E, Walters W, Beischer N, King J. Report on maternal deaths in Australia, 1994–1996. NHMRC cat. no. 0145246, AIHW cat. no. PER 13. Sydney, NSW: Australian Institute of Health and Welfare, National Perinatal Statistics Unit, National Health and Medical Research Council, 2001.
  • 45
    Abouzahr C. Global burden of maternal death and disability. Br Med Bull 2003; 67: 111.