The 2005 review of Queensland Health maternity services revealed low attendance rates of Aboriginal and Torres Strait Islander (hereafter Indigenous) women at antenatal care, higher rates of tobacco and alcohol use during pregnancy, generally poor maternal health, as well as higher rates of low birthweight and premature babies . In Cape York, the northernmost region of Queensland with a large Indigenous population, maternal and child health (MCH) is particularly poor with high rates of maternal and neonatal morbidity and mortality in comparison to the rest of Australia .
In 2005/06, 70 % of pregnant women in Cape York were reported to have smoked at some time during their pregnancy; there were high rates of gestational diabetes and more than double the number of low birthweight babies in comparison to the rest of Queensland . Regular antenatal care reduces the risk of disease or complication via early identification and treatment, as well as providing opportunities for education about healthy parenting behaviours such as good nutrition, alcohol and smoking cessation and the benefits of breast feeding. The World Health Organisation (WHO) recommends a minimum of four visits to health professionals during pregnancy . Mothers who attend antenatal care regularly and early in their pregnancy are more likely to give birth to babies with normal birth weight and at normal gestation. Perinatal mortality rates also improve when women commence antenatal care earlier in their pregnancy . Culturally respectful healthcare contributes to engagement with the health service and important to this engagement is relationship building, consistency of service provider, connection with the service and commitment from community elders .
In order to address the region’s poor maternal and child health, the Baby Basket initiative was developed by Apunipima Cape York Health Council (ACYHC). ACYHC is a regional Aboriginal community controlled health organisation (ACCHO) responsible for delivering culturally appropriate, comprehensive primary health care to 11 Cape York communities . This innovative program was primarily designed to encourage Indigenous women in remote communities to present at health care centres earlier and more frequently in their pregnancy, with the aim of providing improved antenatal care as well as better mother and family education. The information provided by health workers in these interactions has the potential to assist women in achieving better maternal health and providing a better start in life for their babies .
The Apunipima Baby Basket program involves the provision of three baby baskets of MCH goods with associated health education resources to women in remote Cape York communities. The timing of handover corresponds with formative stages in their maternal cycle from early pregnancy to six months post-partum: 1). Antenatal at pregnancy diagnosis in the community clinic; 2). Delivery in Cairns around the time of childbirth; and 3). Postpartum during a home visit when the baby is six months old . Baby basket contents are described in detail elsewhere . Basket handover is a point of engagement between Indigenous women and the health service. Conducting the basket handover in local communities increases the likelihood that other members in the family will benefit from the education provided and have the opportunity to ask questions, emphasising the family centred approach. Each basket also contains a fresh food voucher valued at $40, which can only be used to purchase fresh fruit and vegetables in community stores. Each mother is entitled to a maximum of five fresh food vouchers, so the vouchers also function as an incentive for mothers to visit the clinic between baby basket handovers [2, 8].
Economic evaluation: a cost study
While positive outcomes of Indigenous MCH programs are often reported in the literature, very few of these studies have a strong evidence base or employ a sound methodological approach to evaluation [10–15]. Such limitations can hamper economic evaluations in Indigenous MCH, making it difficult to determine the impact of specific programs on health outcomes. The need for better quality evaluations, the use of relevant indicators, and the collection of good quality longitudinal data to assess the impact of MCH programs on health outcomes for Indigenous women, infants and children are commonly raised in the literature [10, 16–20].
A retrospective evaluation of the Baby Basket program revealed improvements in key indicators in MCH . While it appears that the program has achieved successful outcomes for families, the aim of the economic evaluation is to prepare an economic case for the value of the Baby Basket program, by investigating the resources required to deliver the current program. The cost study reported here represents an initial step in the economic evaluation conducted of the Apunipima Baby Basket program.
A brief review of the literature was conducted to inform the methods used in costing the Baby Basket program and to identify the reported costs of similar programs. CINAHL, EconLit, Informit, Medline (PubMed), Scopus, and Web of Science were the databases consulted. The following search terms were used in either the title, abstract or article: (Aborigin* or Indigen* or Torres Strait Island*) and (wellbeing or health) and (Australia) and (child or maternal or parent* or women* or pregnan* or infan*) and (program* or service*) and (economic or cost*) and (analysis or evaluation or study). In addition, websites and clearinghouses related to Indigenous maternal and child health and economic or cost analyses were searched. Finally, reference lists of articles identified by the electronic database search were hand-searched for relevant, previously unidentified sources.
The literature review highlighted the paucity of cost studies conducted of Indigenous MCH programs. Jan et al.  conducted an economic evaluation of the Daruk Aboriginal community midwifery service in outer western Sydney. At the time, Daruk programs included antenatal check-ups, home visits, hospital visits and delivery. The authors compared net health sector costs for Indigenous women receiving antenatal care in the Daruk midwifery service and Indigenous women receiving antenatal care at nearby services (e.g. Nepean Hospital). Patient data was gathered from medical records and the Midwives Data Collection and direct health sector costs were calculated as Daruk operational costs less the associated savings for nearby midwifery services. Downstream costs comprised use of services by Indigenous women (e.g. length of hospital stay, antenatal visits), calculated as the differences in costs incurred at Daruk service and Nepean Hospital. Cost savings were evident for the Daruk midwifery service. The study demonstrates one approach for estimating costs through comparison with other services. Such an approach relies on good co-operative relations with other services and access to patient medical records.
In more recent literature, Cannon et al.  developed a pregnancy simulation model to construct costs based on epidemiological pregnancy data for their obstetric population. Simulation modelling was used due to the paucity of comprehensive data (as noted above) and the small number of births in rural and remote areas. The authors compared pregnancies receiving adequate and inadequate care and their results show that the costs of programs which aim to increase access to antenatal care are likely to be cost effective. While this study was rigorous in its approach, as a simulation, the findings can only be suggestive. The authors recommend further investigation of the provision of improved antenatal care, claiming that only prospective data collection in a clinical setting could improve on their findings, something that would take considerable time to achieve.
A retrospective and prospective cohort study was conducted by Gao et al.  to provide data for a cost-consequences analysis which compared a baseline cohort with a more recently introduced Midwifery Group Practice (MGP). The program timeline of this cost analysis was similar to that of the Baby Basket program including the first antenatal care visit, birth, post natal care in Darwin, and in community up to six weeks after birth. The study took the perspective of the Northern Territory Department of Health and examined the direct costs of the program to compare maternity care costs pre and post introduction of the MGP. The methods employed to derive the costs were particularly rigorous, with cost assumptions (based on opinions of expert informants well versed in midwifery care in the region) used only to account for missing data . The Darwin MGP was less costly, though not significantly so. However, there was an improvement in clinical outcomes, quality of care and cultural safety, maternal and child health data, Aboriginal employment, greater use of services, reduced catastrophic events and length of stay . While this study utilised a pre and post study design, was more focussed on clinical outcomes, and the sample size was small, the methodological rigour employed in costing and the use of retrospective and prospective data informed the economic evaluation of the Baby Basket program.
The literature revealed very few cost studies or economic evaluations of Indigenous MCH programs, and an absence of research designed for causal inference which relates to general weaknesses in data collection and the methodology employed. Even recent economic evaluations that were otherwise methodologically rigorous suffered from a lack of data. Ideally, to minimise cost, the data used for quantitative evaluations should be based on routinely collected information and employ a suitable study design that enables meaningful comparison with a control group.