Impact of a collaborative shared antenatal care program for urban Indigenous women: a prospective cohort study

Births to women of Aboriginal or Torres Strait Islander descent represent about 3% of confinements in Australia each year, and the state of Queensland has the highest number of Indigenous births. 1,2 The poor health of the Indigenous population in Australia is well documented and is reflected in demoralising Indigenous perinatal statistics, with rates of preterm birth, low birthweight and perinatal mortality remaining more than twice that of the non-Indigenous population over the past decade.1-3

High quality antenatal care is seen as a fundamental right of all women to safeguard their health and that of their infants,4 providing opportunities for risk factor intervention.5 Previous research has shown that late antenatal attendance, maternal malnutrition and high rates of tobacco and alcohol use are associated with poor obstetric outcome.6-9 Consequently, providing perinatal care services that are community-based has been advocated as fundamental to improving obstetric outcome in Indigenous communities.10 There have been a small number of such programs, predominantly in more remote locations, showing some success.11,12

With an Indigenous population of 4369, Townsville had the largest Indigenous population (70% Aboriginal and 30% Torres Strait Islander descent) in Australia (5% of total population) at the last census.13 Most Indigenous births in Townsville occur at Townsville Hospital, the tertiary referral centre for North Queensland. In 1998, data from Townsville Hospital showed that 64% of Indigenous women made fewer than four antenatal visits, and only 35% participated in shared antenatal care with a general practitioner, compared with 70% of non-Indigenous women (personal unpublished data). In response to this, and to poor local perinatal outcomes,3 health service providers in Townsville have worked closely with the Indigenous community to improve antenatal services. This collaboration has produced an integrated model of antenatal shared care — the Mums and Babies program — delivered from the community-controlled Townsville Aboriginal and Islander Health Service (TAIHS), an Aboriginal Medical Service with an elected board and over 100 staff members.

We aimed to evaluate the effect of this new antenatal program on antenatal care and perinatal outcomes among the Aboriginal and Torres Strait Islander community in Townsville.

Box 2 shows how the study groups were derived. All multiple births were excluded. The historical control group included 84 births, the intervention group (MB group) 456 births, and the contemporary controls 540 births.

There were no statisitcally significant differences in demographic characteristics or in use of social drugs between the three groups (Box 3). The parity of historical controls was significantly greater than that of the MB goup (P < 0.01). A third of both the contemporary control group and MB group were non-Townsville residents (predominantly from North and Far North Queensland). The non-Townsville residents among both the MB group and contemporary control group were significantly younger (P < 0.01 and P < 0.05, respectively), and more likely to be Aboriginal than Torres Strait Islander (both P < 0.05); in the MB group the parity of non-Townsville residents was significantly lower than that of the Townsville residents (P < 0.01).

There is little evidence in the Australian Indigenous community to guide antenatal care planning. Reviews of Australian antenatal care protocols show that the number, timing and content of antenatal visits often remain based in tradition rather than evidence.18,19 Recent World Health Organization (WHO) reviews suggest that no adverse outcomes are associated with a reduced schedule of four visits,20,21 focusing on specific screening, therapeutic interventions and antenatal education.22 The Mums and Babies program with a care plan focusing on screening, nutritional supplementation, education and monthly recalls is similar to the WHO strategy.

Evaluating and interpreting small community datasets and quality improvement initiatives at the primary health care level is difficult, and randomised controlled trials are not always feasible. The success of community-based programs is influenced by factors such as the strength of community relationships, the presence of energetic champions and tailoring the interventions to local context. Interventions such as the one described in this study are also long-term projects that are difficult to assess over short periods. Consequently, our study has limitations, and selection bias is a significant factor. Both the historical control and MB groups were self-selected, and a small number of high risk pregnancies were referred directly to hospital clinics. The limited numbers of women using TAIHS for antenatal care before the program suggests that they were a different group of women, despite having similar demographic characteristics to the MB group. Also, little is known of the Indigenous women in Townsville in the contemporary control group who did not use the Mums and Babies program. Their demographic characteristics were also similar to those of the MB group, but they may have differed in other characteristics, such as education, employment and socioeconomic status. The data for both control groups were limited by what was available — educational status was missing in a signifcant proportion of records, and gestation, but not parity, was recorded in Obicare. The paper records and lack of antenatal care planning at TAIHS before computerisation in 2000 resulted in poor recording of clinical information. In spite of these limitations, this study does support some cautious conclusions.

Use of the Mums and Babies program antenatal care service increased significantly over time, with 60% of Townsville-based Indigenous women attending by 2003. The timeliness of the first visit and the number of antenatal care visits improved significantly. Although well short of the traditional antenatal care schedule, the median number of seven antenatal care visits does fall within the bounds of the WHO model, adapted for a developed country with ready access to ultrasound scanning. The use of the program by large numbers of women from outside Townsville, while not an aim, reflects both the mobility of Indigenous families and the ease patients experience in using the Mums and Babies program. This increased access to antenatal care represents an improved opportunity to establish a relationship with pregnant women that may endure through the early childhood years.

Indicators for measuring the quality of antenatal care are lacking for the shared care model. The WHO has recommendations about what constitutes good quality maternal health services, including criteria relating to access, cultural sensitivity, continuity of care, use of protocols and education,4 but gives no specific indicators for evaluating the services. In this study, it appears that care planning, screening and educational activity performance have improved over the 4 years of the project, suggesting that health care workers are taking advantage of the contact these women are having with the health care system.

Ultrasound dates are more accurate than menstrual dates for determining the stage of a pregnancy, and hence for assessing prematurity.23 During the study period, attendance for ultrasound scanning increased. This cohort of Aboriginal and Islander women will thus be one of the few in Australia with accurately dated pregnancies, which is vital to clinical decision making and perinatal research in the Indigenous community.

The significant reduction in preterm births among the MB group (compared with both control groups) is promising, even though the persistence of high mortality rates is disappointing. Encouragingly, the proportion of preterm births in the MB group was not only significantly lower than that for whole-of-population figures for Indigenous women in Queensland, but very similar to figures for non-Indigenous women in Queensland.15 The non-significant trend to reduced low birthweight births in the MB group women may be the result of selection bias, with increased numbers of women with lower risk profiles attending TAIHS for antenatal care, improved infection screening and accurate dating of pregnancies, or differences in socioeconomic status of the Indigenous community in Townsville. This requires further exploration. The demographic profile does suggest that the women in the MB group carry a significant risk load, which may also explain why birthweight and perinatal mortality rate did not improve in tandem with the reduction in preterm births.

Although not a panacea for poor perinatal outcome, it is the right of the Indigenous community to obtain the same standard of antenatal care as the rest of the community. This study shows that integrated services delivered in a “safe” environment increase access to antenatal care in the Indigenous community. Increased access to antenatal care should afford the opportunity to establish programs that minimise risks such as tobacco and alcohol use, possibly leading to reductions in the prevalence of low birthweight and perinatal mortality in future, particularly if the reduction in preterm births can be sustained. This model may be be adaptable to other urban centres with significant Indigenous populations, community-controlled health services and multiple antenatal care service providers.

2 Flow chart of the groups of Indigenous births included in this study

---

*Other: miscarriages (9), termination (8), private care (2), immediate referral to Townsville Hospital (18). †Comprises 294 Townsville-residents and 162 non-Townsville residents. ‡All Townsville residents. §Comprises 361 Townsville residents and 179 non-Townsville residents.

5 Recorded planning activity and screening activity for all Indigenous women attending Townsville Aboriginal and Islander Health Service (TAIHS) for shared antenatal care in the Mums and Babies program between 1 January 2000 and 31 December 2003

P value for trend: *P < 0.001; †P < 0.05.