Pragmatic indicators for remote Aboriginal maternal and infant health care: why it matters and where to start

Authors


Correspondence to:
Malinda Steenkamp, NHMRC Doctoral Research Scholar, Northern Rivers Department of Rural Health, School of Public Health, PO Box 3074, Lismore, NSW 2480. Fax: (02) 6620 7270
E-mail: malinda.steenkamp@sydney.edu.au

Abstract

Objective: There are challenges in delivering maternal and infant health (MIH) care to remote Northern Territory (NT) communities. These include fragmented care with birthing in regional hospitals resulting in cultural and geographical dislocation for Aboriginal women. Many NT initiatives are aimed at improving care. Indicators for evaluating these for remote Aboriginal mothers and infants need to be clearer. We reviewed existing indicators to inform a set of pragmatic indicators for reporting improvement in remote MIH care.

Methods: Scientific databases and grey literature (organisational websites and Google Scholar) were searched using the terms ‘Aboriginal/maternal/infant/remote health/monitoring performance’. Key stakeholders identified omitted indicators sets. Relevant sets were reviewed and organised by indicator type, stage of patient journey, topic and theme.

Results: Forty-two indicators sets were found. Seven focused on Aboriginal health, 23 on reproductive/maternal health, eight on child/infant health and four on other aspects, e.g. remote health. We identified more than 1,000 individual indicators. Of these, 656 were relevant for our purpose and were subsequently organised into 300 topics and 16 themes for antenatal, birth and postpartum, and infant care by indicator type.

Conclusion: There are many measures for monitoring health care delivery to mothers and infants. Few are framed around remote MIH services, despite poorer health outcomes of remote mothers and infants and the specific challenges with providing care in this setting. Establishing relevant indicators is vital to support relevant data collection and the development of appropriate policy for remote Aboriginal maternal and infant care.

Aboriginal women and their babies have poorer maternal and perinatal outcomes than their non-Aboriginal counterparts.1,2 The remoteness of maternal residence is associated with poorer perinatal outcomes.1,3–5 In the Northern Territory (NT), 66% of Aboriginal births are to remote-dwelling women compared with 5% of non-Aboriginal births.6 Current NT practice is for remote women to be transferred to regional hospitals for birthing around 38 weeks gestation.7 Remote antenatal and infant health services are provided either on-site in community health centres or by outreach services.

A maternity services review commissioned by the NT Department of Health and Families (DHF) in 2007 identified numerous challenges with the current system.7 These include: (a) systemic challenges affecting all maternal services; (b) issues pertaining to Aboriginal women; and (c) challenges particular to remote Aboriginal women (Table 1). The DHF had previously implemented a number of initiatives to improve maternity services (e.g., a birthing centre in Darwin)7 and some recommendations of the 2007 Review are being implemented.8 Concomitantly, other initiatives such as the Commonwealth and NT Governments’ campaigns to achieve Aboriginal health and life expectation equality have occurred, which affect maternal and infant (MIH) services across the NT.9–12 Furthermore, since 2007, a large National Health and Medical Research Council (NHMRC) funded project aimed at advancing the quality and effectiveness of MIH services in remote Aboriginal communities during pregnancy and the year after birth (1+1 A Healthy Start to Life Project) has been undertaken.13 Baseline data collected for this study confirmed the challenges identified by the DHF.14

Table 1.  Issues regarding maternal and infant service delivery.
All NT women and childrenAll NT Aboriginal womenNT Remote Aboriginal women
The NT MIH care model is complex, medicalised and involves multiple organisations and sectors resulting in fragmented care. Fragmented care is associated with a number of adverse outcomes.28–30
Patient record systems use different unlinked computer-based systems with varying degrees of hard and soft copy records exacerbating fragmented care.7
There is an endemic shortage of midwives, together with high staff turnover, insufficient staff orientation, and limited relief for remote staff.7
Remote staff's acute workload is high and7 they often experience priority conflicts.14
Maternal care is described as culturally inappropriate.31–33
Engagement with antenatal services often occurs at a later stage than for non-Aboriginal women.7
There is a shortage of Aboriginal health workers choosing to work in maternity care.7
There is lack of choice regarding birthplace, despite the cultural importance of ‘birthing on country’.34 Current patient travel funding prevents all remote pregnant women from having an escort. The cultural and social dislocation resulting from not birthing on country and without someone familiar in regional centres is considerable.32,34 The separation from other children and family contributes to significant stress for the women and their families,32,34 contributing to health care avoidance for some women.32
Regional centre transfers pose many logistical problems.31,35–37
Pregnant women may experience boredom, problems with money, food and local travel, and lack of security during their stay in regional hostels.31
There are problems with transfer between treatment points in regional centres, especially after hours.7
Transfer policies are not consistent, e.g. one plane service would only transport newborns who were eight days or more.
Distances travelled can be problematic, e.g. women from around Tennant Creek travelling to/from Alice Springs are likely travel for eight hours or more by bus.7

MIH services and service improvements should undergo evaluation to monitor progress and outcome improvements.15 Although many MIH indicators exist in current policy documents and published and unpublished literature,16–22 few are targeted towards remote Aboriginal MIH care and even fewer include measures relevant to less tangible outcomes such as acceptability/cultural safety. For example, the NT Aboriginal Health Key Performance Indicators (NTAHKPI) focus on urban and remote primary care service delivery but only one out of 19 indicators relate to maternity care.18 (Four of the 19 measures include infant health.) Similarly, the ten National Core Maternity Indicators refer to clinical maternity care, but none address continuity or acceptability of care – both issues vital for improving remote maternity care.

This paper reviews existing MIH indicators to inform the establishment of pragmatic indicators relevant to the specific challenges in delivering MIH services to remote Aboriginal mothers and infants.

Methods

From August 2008 – April 2009, health sciences databases [APAIS-Health (Informit); Blackwell Synergy; CINAHL with Full Text (EBSCO); Cochrane Library; PubMed; and Science Direct (Elsevier)] were searched using the terms: ‘Aboriginal/Indigenous health’, ‘maternal /reproductive/ obstetric/women's health’, ‘antenatal/ birth/ delivery/ postpartum/ postnatal’, ‘infant/ child health’, ‘remote/ rural/regional health’, ‘indicators/quality measures/monitoring performance/audit measures’. Most indicators were identified from organisational websites, their reference documents or from Google Scholar searches. Indicators related to mental/social/emotional health, social determinants of health or welfare issues were not included as the focus was MIH care.

An expert group were consulted to identify additional relevant sets that were not identified. The group comprised 14 individuals (mostly from the NT) working in maternal and/or child health and/or with evaluation work or with developing indicators. The group included a child health nurse, two epidemiologists, midwives and NT government employees (one from Aboriginal policy and another involved with the NTAHKPI Project), but no consumers or medical doctors. The obstetrician invited could not participate due to time constraints. Consumers are involved in other aspects of the 1+1 A Healthy Start to Life Project. The review was comprehensive, but not exhaustive.

Individual indicators were examined. Reproductive health indicators other than in relation to pregnancy and indicators related to the health of children older than 1 year were excluded. We distinguished three indicators types: (a) process indicators measuring what is being done in providing healthcare,23 e.g., “Proportion of women offered appropriate interventions in relation to smoking”; (b) outcome indicators providing quantitative data related to the outcomes of health system performance,23 e.g., “Rate of live term infants with an APGAR score of6 at 5 minutes”; and (c) indicators related to determinants of health, e.g. “Prevalence of maternal tobacco smoking during pregnancy”. Indicators were also divided into three periods according to the different stages of pregnant women's and their babies’ progress through the system (the patient journey): (i) the antenatal period, (ii) the birth and postpartum period, and (iii) the first year of life. Individual measures were organised into lists by indicator type and stage of patient journey and then further grouped by topic, e.g., antenatal coverage, delivery method, or birth weight. Finally, the topics were organised into broad themes. These related to the challenges affecting current NT MIH care, e.g., continuity of care, workforce issues, etc. (Table 1).

Results

We identified 67 relevant documents. Seven were excluded as they were not related to MIH care. Forty-two indicator sets were identified: seven related to Aboriginal health; 23 to maternal health; eight to infant health; and four to other aspects such as health and aged care or remote health. More than 1,000 individual measures were identified. (For more details about these measures, please contact the corresponding author.) A total of 656 were reviewed in detail and could be combined into 300 topics (Table 2). There was much overlap. Many indicator sets included measures dealing with similar topics, especially where sets had the same focus, e.g., to monitor hospital clinical care. There were some indicator topics that appeared in nearly all sets: antenatal coverage, maternal tobacco smoking and/or alcohol use during pregnancy, delivery type, labour induction, caesarean section occurrence, perineal trauma, birth weight, perinatal mortality rate, Apgar score at 5 minutes, breastfeeding, and immunisation status. The 300 topics were grouped into 16 themes and organised by indicator type and stage of patient journey (Table 2). For all three stages of the patient journey, the majority of indicator topics dealt with process measures. There were fewer indicator topics where health outcomes and health determinants were concerned. Of the 177 process indicators topics, 41% focused on the clinical care of the mother or infant. Few centred on continuity of care, culturally appropriate care or postnatal care. Of the 80 health outcomes topics, 60% dealt with mothers’ morbidity. About 60% of the 43 topics dealing with health determinants health centred on parental issues.

Table 2.  Indicator topics (n=300) organised by indicator type, theme and stage of patient journey.
 Stage of patient journey
Indicator type and themeAntenatalIntra- and postpartumFirst year of life
Process49 (62%)79 (65%)49 (49%)
Clinical care – Mother1531
Clinical care – Infant/Child522
Culturally appropriate care750
Continuity of care662
Postnatal care6
Service organisation101914
Workforce4311
Health education/promotion74 
Health outcomes15 (19%)36 (30%)29 (29%)
Mortality57
Fertility114
Morbidity – Mother414
Morbidity – Infant/Child1317
Disability5
Health determinants15 (19%)6 (5%)22 (22%)
Maternal/Parental factors1528
Infant/child factors44
Other factors010
Total79 (100%)121 (100%)100 (100%)

Discussion

Current NT initiatives have the potential to mitigate poorer health outcomes and improve services for remote Aboriginal women and infants. There are many existing indicators from a range of policy documents and the literature that are or can be applied to monitor MIH care. However, there is a lack of consistency between indicators across jurisdictions and few existing quality measures are specifically framed around MIH care in remote Aboriginal communities. There are some exceptions, including the NTAHKPI that focus on primary health care services,18 the ABCD Project that aims to improve maternal and child health care at a service level24 and work by Hancock that focuses on all Aboriginal women.25 A significant gap is practical measures that relate to aspects of care around acceptability or cultural safety for Aboriginal people. Although there are proxy measures such as ‘Attendance of antenatal services’, these do not capture the complexities of the issue. This area for further work is a focus of the 1+1 A Healthy Start to Life Project. The challenge is to establish appropriate indicators that have real value in demonstrating improvement for remote Aboriginal women and infants accessing NT MIH services that can be monitored on a regular basis. Such indicators are likely to be useful in other jurisdictions with similar populations. However, performance indicators on their own are not sufficient to improve outcomes.26 There needs to be organisational change and recognition of quality gaps with participatory processes used to implement change. Such a process is in train with the 1+1 A Healthy Start to Life Project. Establishing pragmatic indicators is likely to lead to improvement in what is measured27 and is central to informing future data collection enhancements, as well as the development of policy for Aboriginal MIH care. This paper is the first step towards establishing such indicators; a task that is well advanced and the subject of a subsequent paper.

Acknowledgements

Malinda Steenkamp and Sarah Bar-Zeev are NHMRC Postgraduate Training Award recipients. Malinda Steenkamp received the Beryl Henderson Award from the Australian Federation of University Women Inc. This work has also been supported by a PHCRED bursary awarded by the Discipline of General Practice at the University of Adelaide. The PHCRED (Primary Health Care Evaluation and Development) Program is funded by the Australian Government Department of Health and Ageing. The 1+1 A Healthy Start to Life Project is supported by a NHMRC Healthy Start to Life Grant No. 422503.

Ancillary