When a newborn is very small, doctors have to decide how much intensive treatment to give. In Australian hospitals, a birthweight of 1,500 grams often serves as a marker: babies below it tend to receive more monitoring and intensive care than those just above it. A baby weighing a little under 1,500 grams is, on average, very similar to one weighing a little over it, so the extra care given at this mark is close to a natural experiment. Comparing babies born just below and just above the 1,500-gram threshold provides a credible way to estimate the local effect of additional neonatal care, provided that infants cannot be precisely sorted around the cutoff and that predetermined characteristics vary smoothly through the threshold. The project will formally test these assumptions and document whether treatment intensity changes at the cutoff. The project uses linked administrative records covering births in Western Australia. The Midwives Notification System records birthweight to the gram, which is what the comparison at the threshold requires, and the linkage follows each child into hospital, emergency department, death, and school records. This gives the treatment received, its cost, and the later outcomes for a complete population of births rather than a sample. The project examines a broad range of observed consequences, including survival, health service use, costs, later health, schooling outcomes, and potential family spillovers. It first estimates the effect on the child: survival in the first year, the amount and cost of hospital care, later health, and school outcomes. It then asks whether the benefits extend to the wider family, through the mother’s health and the school results of siblings. This wider question is what distinguishes the project, since neonatal services are usually valued only by what happens to the baby, and any gains to the family go uncounted
Aim
The project has two aims.
The project estimates the local effect of the additional neonatal care associated with birthweight just below 1,500 grams, including differences in hospital care, length of stay, costs, readmissions, survival, and later health and schooling outcomes. The second aim is to find out whether the benefits reach the wider family. The project will also examine whether the effects extend beyond the newborn, including maternal health after the birth and siblings’ school outcomes, where linkage quality and sample size permit
Objectives
Using linked population records, the student will construct an analysis-ready dataset from linked population records covering births in Western Australia, subsequent hospital, emergency department, death, and school outcomes, and, where linkage permits, records for mothers and siblings. They will estimate the effect of the extra care given at the 1,500 gram mark using modern applied econometric methods, and test that the approach holds up. They will then estimate the effects on the child’s later health and schooling, and, where the numbers allow, on the health of mothers and the school results of siblings.
Significance
Caring for very small newborns is among the most resource-intensive services the health system provides, and how much that care improves outcomes for babies weighing around 1.5 kilograms remains clinically and economically important. The difficulty for evidence is that babies who receive more care are sicker to begin with, so simple comparisons confound the effect of the care with the condition that prompted it. This project breaks that confound by comparing babies on either side of the 1,500 gram mark, where treatment changes but the babies themselves do not. It estimates the effect of the care on the child’s survival, later health, and schooling, and then asks whether the benefits extend to the mother’s health and to siblings’ school outcomes. Where such family effects exist they can be substantial, which means an assessment confined to the treated child understates the value of newborn care. Establishing the wider return matters for how neonatal services are valued and planned, and it bears on a broader question in health economics, namely how far the benefits of medical care reach beyond the patient who receives it. The project also gives the student rigorous training in causal analysis with linked population data.
Ideal Candidate
We are looking for a motivated PhD candidate with strong quantitative skills and a background in economics, econometrics, biostatistics, epidemiology, or a related field. Experience with modern applied econometric methods for causal analysis would be an advantage, as would experience working with large administrative datasets in Stata or R. The candidate should be careful with data, have sound statistical judgment, and be interested in health policy. Candidates must be eligible to enrol in a PhD program at Curtin University.
This project is open to International and Domestic applicants.
Internship
We anticipate that this project will include an internship, which we expect to arrange by drawing on the research group’s industry and health system connections. Potential hosts include the Western Australia Country Health Service and Fiona Stanley Hospital, among other partners. The aim is to place the student where they can see how newborn and maternity services are planned, funded, and delivered, and where they can share the project’s findings with the people who make those decisions.
Scholarship
If you are identified as the preferred candidate for this project, you may be considered for an RTP scholarship.
Enquires and How to Apply
For enquires about this opportunity contact Dr Marshall Makat at Marshall.Makate@curtin.edu.au
To formally apply submit an Expression of Interest to Dr Marshall Makat during the Central Scholarship round (July 1st – July 31st 2026)