ISSN: 2652-5518
By Emily Callander, Lucas Calais-Ferreira, Haylee Fox, Sue Malta, John Hopper
Summary
Medical advancements and increasing age at first pregnancy have contributed to a worldwide increase in multiple births. However, there is very little information available about the costs associated with having more than one baby at a time, and the potential financial impact this might have long-term. We therefore investigated differences in pregnancy and birth-related events, health service utilisation and out-of-pocket healthcare fees between women who have twins and those who have single births. This paper thus contributes to a better understanding of the inequalities in the experience of mothers of twins compared with mothers who have a baby from a singleton pregnancy.
Main points
- Medical advancements and increasing age at first pregnancy have contributed to a worldwide increase in multiple births
- The costs associated with having more than one baby at a time have not been fully understood until now
- The results shows that there is a substantial difference in pregnancy and birth-related events, health service utilisation and out-of-pocket healthcare fees between women who have twins and women who have single births
- The impact of having two babies at once is not recognised at a local, National or policy level in Australia
- Further research is needed to better understand the implications of these findings
Authors
Emily Callander, School of Public Health and Preventive Medicine, Monash University
Lucas Calais-Ferreria, Sue Malta, John Hopper, Centre for Epidemiology and Biostatistics, Melbourne School of Population and Global Health, the University of Melbourne
Haylee Fox, School of Public Health, James Cook University
Contact: emily.callander@monash.edu
Acknowledgements: We would like to thank the Statistical Service Branch at Queensland Health and the Australian institute of Health and Welfare for their assistance with the data linkage for this study. Emily Callander received salary support through an NHMRC Career Development Fellowship (APP1159536).
Introduction
Globally, the rate of twin births varies between countries from 8 to 17 per 1,000 births 1. Over the last 30 years, the overall rate of twins has increased by more than 70%. The steep increase in twin birth rates is predominantly driven by the increasing rates of dizygotic (nonidentical) twins1. Monozygotic (identical) twin pairs have been reported to occur at a somewhat constant rate of 3.5-4 per 1,000 births2-4. The changing rate of dizygotic twin pairs is likely influenced by environmental factors such as increasing maternal age of first birth5 and the use of assisted reproductive technologies such as ovulation induction and in vitro fertilisation6-8.
Maternal morbidity and mortality are higher for twins at each gestational age compared with singleton pregnancies9,10; and twin pregnancies are associated with an increased risk of obstetric complications during labour and birth. Complications related to twin pregnancies include preterm birth, Intra Uterine Growth Restriction, pre-eclampsia, gestational diabetes, congenital anomalies, malpresentation, cord prolapse, Twin to Twin Transfusion Syndrome, postpartum haemorrhage, stillbirth and neonatal death. Babies from twin pregnancies are more likely to be born with a lower birthweight, have an Apgar score less than 7 at the 5th minute and be admitted to a Special Care Nursery (SCN) or a Neonatal Intensive Care Unit (NICU)10,11. Other conditions that are more likely to be experienced in twin pregnancies are miscarriage and anaemia12. Medical intervention is also higher in twin pregnancies, with drastic recent increases in caesarean sections for women who have twin pregnancies13.
This paper begins with a description of the methods used to investigate differences in pregnancy and birth-related events, as well as health service utilisation and out-of-pocket healthcare fees, between women who have twins and those who have single births using data from a whole-of-population linked dataset called ‘Maternity1000’14.
Methods
Data
Maternity1000 utilises the Queensland Perinatal Data Collection (PDC) to identify women who gave birth in Queensland between 1 July 2012 and 30 June 2015 (n=186,789, plus their babies, n= 189,909). All women were identified by Queensland Health’s Statistical Services Branch from the Queensland Perinatal Data Collection and Queensland Birth Registry. The records were linked to Queensland Hospital Admitted Patient Data Collection (QHAPDC), Deaths Registry, Emergency Department Information System (EDIS) and Hospital and Health Service (HHS) Funding and Costing Unit records between 1 July 2012 and 30 June 2015. The records were then linked by the Australian Institute of Health and Welfare (AIHW) to their corresponding Medicare Benefits Schedule (MBS) and Pharmaceutical Benefits Scheme (PBS) claims records14.
In order to produce national estimates, the Maternity1000 dataset was weighted to the Australian population by mother’s age, Indigenous identification, and area of residence. Australian benchmarks for these variables were taken from the Australian Institute of Health and Welfare’s Mothers and Babies reports relating to the years 2012 – 201515, to align with each of the years covered by Maternity1000. GREGWT, an algorithm developed by the Australian Bureau of Statistics to weight their population-based surveys, was used to produce the weights16.
Twin identification
The PDC identified the plurality of all births. The analysis was limited to those babies that were from a singleton pregnancy or twins. Babies from pregnancies that resulted in more than twins were excluded. This resulted in 186,727 babies being retained for analysis.
Outcomes
In order to identify the pregnancy and birth experiences of women, the number of antenatal visits attended by women (categorised as less than 8; or 8 or more*), number of antenatal ultrasounds accessed, womens gestational age (categorised as <28 weeks; 28 – 31 weeks; 32 – 36 weeks; 37 weeks, 38 weeks, 39 weeks, 40 weeks, 41 weeks, 42 weeks or more), onset of labour (categorized as spontaneous; induced; no labour, caesarean section), mode of birth (categorised as vaginal birth, non-instrumental; vaginal birth, with forceps; vaginal birth with vacuum; and caesarean section), whether or not the baby was admitted to SCN or NICU and the length of stay in SCN or NICU, and the woman’s length of hospital stay were also identified from the PDC.
To identify the number of services accessed by each woman for her health service use in pregnancy, and the baby’s health service use from birth through to the time they were one year of age, the number of inpatient events and emergency department episodes were identified from the QHAPDC and EDIS, respectively. The number of Medicare services accessed by women for their health service use in pregnancy, and their babies from birth through to the time they were one year of age, were identified from MBS claims records. Services were categorised according to the Medicare Broad Type of Service (BTOS)17. The number of each service used by each twin was summed to create the number of services accessed by the women and her babies.
The out-of-pocket fees charged for each service were identified from MBS claims records. Fees were summed to give a total for each woman’s health service use in pregnancy, and for each baby’s health service use from birth through to the time they were one year of age. The out-of-pocket fee charged for each service used by each twin were summed to identify the out-of-pocket fees charged to the ‘family’.
Statistical analysis
Descriptive analysis was initially undertaken to identify the demographic and clinical characteristics of women with singleton and twin pregnancies. The differences in pregnancy and birth-related events, health service use and out-of-pocket costs were then compared between women with a singleton pregnancy and a twin pregnancy.
Multiple logistic regression was utilised to estimate the odds ratio of having eight or more antenatal visits; the baby being born at less than 28 completed weeks gestation, 28 – 31 completed weeks gestation, and 32 – 36 completed weeks gestation; having a spontaneous onset of labour, being induced, and having no labour; having a vaginal birth without instruments, having a vaginal birth with forceps, having a vaginal birth with vacuum, and having a caesarean section birth; being admitted to the SCN, and being admitted to the NICU for twins compared with singletons.
A generalised linear model of the number of ultrasounds scans; length of stay in the SCN; length of stay in the NICU; length of stay in the hospital at time of birth; inpatient events; emergency department presentations; general practitioner consultations; obstetrician services and consultations; other specialist consultations; pathology collection items and tests; diagnostic imaging; and other MBS services was constructed for twins compared with singletons.
Two generalised linear models of the out-of-pocket fees charged for the woman’s health service use in pregnancy, and the baby’s health service use from birth through to the time they were one year of age were then constructed. All models were adjusted for a woman’s age, if she identified as Indigenous, smoking status before 20 weeks’ gestation, rurality, socioeconomic status, and if the woman gave birth in a private hospital. The adjusted odds ratio of each outcome measure is presented. The adjusted mean out-of-pocket fees for twins and singletons produced by the model are presented, along with the cost ratio.
Ethics approvals
Ethics approval for the use of the Maternity1000 dataset was obtained from the Townsville Hospital and Health Service Human Research Ethics Committee (HREC) (HREC/16/QTHS/ 223), James Cook University HREC (H7246) and the Australian Institute of Health and Welfare HREC (EO2017-1-338). We also received Public Health Act Approval (RD007377) to waive consent for data collection.
Results
Who has twins?
Women with a twin pregnancy and women with a singleton pregnancy were similar in terms of their demographic characteristics (see Table 1). The majority of women from both groups were in the middle to high socioeconomic quintiles, were non-indigenous and lived in urban areas. Women who had twin pregnancies were more likely to be non-smokers (89.2%) and to have had assisted conception (27.5%) compared with women who had a singleton pregnancy (87.3% and 4.8%, respectively). Women who had a twin pregnancy were also more likely to have given birth in a private facility (37.8%) compared with women who had a singleton pregnancy (27.4%).
Table 1. Demographic and maternal characteristics of women with a twin pregnancy compared with women with a singleton pregnancy, weighted to the Australian population, 2012 – 2015.
| Characteristics | Women with a twin pregnancy | Women with a singleton pregnancy |
| Age, Mean (SD) | 31.7 (14.0) | 30.1 (14.1) |
| BMI, Mean (SD) | 25.3 (15.4) | 25.7 (15.3) |
| Socioeconomic Status | ||
| Quintile 1 (Most disadvatanged), N (%) | 1,089 (5.9%) | 72,009 (6.1) |
| Quintile 2, N (%) | 589 (3.3%) | 38,163 (3.3%) |
| Quintile 3, N (%) | 3,448 (19%) | 242,063 (20.7%) |
| Quintile 4, N (%) | 7,641 (42.1%) | 496,196 (42.3%) |
| Quintile 5 (Least disadvantaged), N (%) | 5,410 (29.8%) | 323,634 (27.6%) |
| Indigenous identification | ||
| Yes, N (%) | 605 (3.3) | 49,270 (4.2) |
| No, N (%) | 17,916 (96.7) | 1,134,935 (95.8) |
| Location of residence | ||
| Major city, N (%) | 9,709 (53.4%) | 599,621 (51.1%) |
| Inner Regional, N (%) | 3,601 (19.8%) | 219,981 (18.8%) |
| Outer Regional, N (%) | 3,231 (17.8%) | 235,990 (20.1%) |
| Remote and Very Remote, N (%) | 1,627 (9%) | 116,474 (9.9%) |
| Tobacco smoking at 20 weeks’ gestation | ||
| Yes, N (%) | 1,981 (10.8%) | 149,347 (12.7%) |
| No, N (%) | 16,416 (89.2%) | 1,029,859 (87.3%) |
| Assisted conception | ||
| Yes, N (%) | 5,095 (27.5%) | 56,247 (4.8%) |
| No, N (%) | 13,426 (72.5%) | 1,127,972 (95.2%) |
| Private hospital birth | ||
| Yes, N (%) | 6,688 (37.8%) | 330,982 (28.4%) |
| No, N (%) | 11,020 (62.2%) | 833,112 (71.6%) |
| Gestational age | ||
| <28 weeks, N (%) | 864 (4.7%) | 8,690 (0.7%) |
| 28 – 31 weeks, N (%) | 1,350 (7.3%) | 7,401 (0.6%) |
| 32 – 36 weeks, N (%) | 9,984 (53.9%) | 68,123 (5.8%) |
| 37 weeks, N (%) | 4,720 (25.5%) | 85,953 (7.3%) |
| 38 weeks, N (%) | 1,414 (7.6%) | 246,930 (20.9%) |
| 39 weeks, N (%) | 119 (0.7%) | 326,467 (27.6%) |
| 40 weeks, N (%) | 48 (0.3%) | 287,343 (24.3%) |
| 41 weeks, N (%) | 20 (0.1%) | 147,505 (12.5%) |
| 42+ weeks, N (%) | – | 5,890 (0.5%) |
What events do women with twin pregnancies experience?
Table 2 showsthat women who had a twin pregnancy were more likely to have attended more than 8 antenatal visits compared with women with a singleton pregnancy (aOR 1.41, 95% CI: 1.25 – 1.60). Notably, more than 91% of women who had a twin pregnancy gave birth at 36 completed weeks of gestation or less compared with only 14% of mothers who had a singleton pregnancy. After adjusting for confounding factors, women who had twins were more likely to have given birth at less than 28 completed weeks gestation (aOR 6.81, 95%CI : 5.4 – 8.48), or between 28 and 31 completed weeks gestation (aOR 13.27, 95%CI : 11.06 – 15.93), or between 32 and 36 completed weeks gestation (aOR 20-08, 95%CI : 18.45 – 21.86) (Table 2).
The predominant mode of birth for babies from a twin pregnancy was caesarean section (70.3%), which was more than double that of those from a singleton pregnancy (31.6%)(Table 2). After adjusting for confounding factors, babies from a twin pregnancy were 4.96 time more likely to be born by caesarean section than those from a singleton pregnancy (95% CI: 4.64 – 5.30).
Table 2. Pregnancy and birth events experienced by women with a twin pregnancy compared with those with a singleton pregnancy, weighted to the Australian population, 2012 – 2015.
| Women with a twin pregnancy | Women with a singleton pregnancy | aOR twins compared with singletons (95% CI) | |
| Antenatal visits | |||
| Eight or more antenatal visits, N (%) | 2,606 (85.7%) | 945,447 (79.8%) | 1.41 (1.25 – 1.60) |
| Preterm birth | |||
| <28 weeks, N (%) | 864 (4.7%) | 8,690 (0.7%) | 6.81 (5.4 – 8.48) |
| 28 – 31 weeks, N (%) | 1,350 (7.3%) | 7,401 (0.6%) | 13.27 (11.06 – 15.93) |
| 32 – 36 weeks, N (%) | 9,984 (53.9%) | 68,123 (5.8%) | 20.08 (18.45 – 21.86) |
| Onset of labour | |||
| Spontaneous, N (%) | 5,329 (28.9%) | 650,701 (54.9%) | 0.35 (0.32 – 0.38) |
| Induced, N (%) | 3,645 (19.7%) | 297,234 (24.7%) | 0.75 (0.68 – 0.83) |
| No labour, caesarean section, N (%) | 9,527 (51.4%) | 236,371 (19.7%) | 3.95 (3.62 – 4.32) |
| Mode of birth* | |||
| Vaginal, non-instrumental, N (%) | 8,841 (23.9%) | 684,819 (57.8%) | 4.96 (4.64 – 5.30) |
| Vaginal, forceps, N (%) | 732 (2.0%) | 31,600 (2.7%) | 0.74 (0.59 – 0.92) |
| Vaginal, vacuum, N (%) | 1,154 (3.1%) | 93,295 (7.9%) | 0.39 (0.33 – 0.46) |
| Caesarean section, N (%) | 26,304 (70.3%) | 374,599 (31.6%) | 0.24 (0.23 – 0.26) |
| Admission to Special Care Nursery (SCN)* | |||
| Yes, N (%) | 23,930 (64.6%) | 180,875 (15.3%) | 10.78 (10.11 – 11.48) |
| Admission to Neonatal Intensive Care Unit (NICU)* | |||
| Yes, N (%) | 6,298 (17.1%) | 26,316 (2.2%) | 9.49 (8.68 – 10.39) |
*Denominator is the number of individual children, not women.
Following birth, twin babies had 10.78 (95% CI: 10.11 – 11.48) times the odds of being admitted to a SCN (Table 2) compared with singleton babies; and of those admitted, spent 131% longer in the SCN (95% CI: 2.23 – 2.40)(Table 3). The average length of stay in the SCN was 13.5 days for twin babies (Table 3). Twin babies also had 9.49 times the odds of being admitted to NICU than singleton babies (95% CI: 8.68 – 10.39)(Table 2), and their stays were 25% longer (95% CI: 1.13 – 1.39) (Table 3).
Table 3: Health services use accessed by women in pregnancy and by babies in the first year post birth, women with twins and singletons weighted to the Australian population, 2012 – 2015.
| Number of services | Adjusted ratio+, twins compared with singletons (95% CI) | ||
| Women with a twin pregnancy, mean (SD) | Women with a singleton pregnancy mean (SD) | ||
| Women’s health service use in pregnancy | |||
| Number of ultrasound scans in pragnancy | 5.8 (11.2) | 3.0 (5.9) | 1.89 (1.86 – 1.91) |
| Average length of stay at birth | 5.3 (9.8) | 3.0 (5.1) | 1.54 (1.52 – 1.55) |
| Inpatient events | 2.6 (8.1) | 2.0 (6.7) | 1.34 (1.30 – 1.38) |
| Emergency Department presentations | 0.8 (5.8) | 0.8 (5.5) | 1.15 (1.07 – 1.23) |
| General Practitioner Consultations | 13.0 (12.2) | 12.4 (13.1) | 1.06 (1.02 – 1.10) |
| Obstetrician Consultations and Services | 2.6 (4.5) | 2.1 (3.9) | 0.97 (0.91 – 1.03) |
| Other Specialist Consultations and Services | 3.1 (8.3) | 1.8 (5.3) | 1.35 (1.26 – 1.44) |
| Pathology Collection | 9.9 (10.9) | 8.6 (9.8) | 1.08 (1.04 – 1.13) |
| Pathology Tests | 12.1 (17.0) | 9.8 (12.8) | 1.15 (1.10 – 1.21) |
| Diagnostic imaging | 3.1 (3.9) | 2.4 (3.3) | 1.24 (1.18 – 1.31) |
| Other MBS services | 5.5 (11.8) | 5.5 (12.6) | 1.18 (1.09 – 1.28) |
| Baby’s health service use in first year | |||
| Length of stay from those admitted to SCN | 13.5 (30.2) | 5.8 (21.0) | 2.31 (2.23 – 2.40) |
| Length of stay from those admitted to NICU | 13.4 (55.6) | 10.3 (49.4) | 1.25 (1.13 – 1.39) |
| Inpatient events | 3.2 (5.7) | 1.3 (2.4) | 2.40 (2.34 – 2.45) |
| Emergency Department presentations | 2.2 (4.3) | 1.1 (3.0) | 1.98 (1.92 – 2.04) |
| General Practitioner Consultations | 15.1 (10.6) | 9.0 (6.2) | 1.69 (1.64 – 1.74) |
| Specialist Consultations and Services | 13.1 (22.9) | 1.7 (3.9) | 5.13 (4.87 – 5.40) |
| Pathology Collection | 4.8 (14.1) | 1.1 (2.8) | 3.25 (3.02 – 3.49) |
| Pathology Tests | 6.3 (19.5) | 1.2 (5.3) | 3.77 (3.51 – 4.08) |
| Diagnostic imaging | 1.5 (3.7) | 0.3 (1.4) | 4.01 (3.69 – 4.34) |
| Other MBS services | 18.1 (26.2) | 8.7 (7.6) | 2.12 (2.05 – 2.18) |
+ Adjusted for mother’s age, if the mother identified as Indigenous, mother’s smoking status before 20 weeks’ gestation, rurality, socioeconomic status, and if the mother gave birth in a private hospital
What health services do women with twin pregnancies use?
During pregnancy, women with a twin pregnancy had 89% more antenatal scans (95% CI: 1.86 – 1.91) compared with mothers of a singleton pregnancy. Mothers of twins also had 34% more inpatient events (95% CI: 1.30 – 1.38), 15% more emergency department presentations (95% CI: 1.07 – 1.23), 25% more consultations with specialists other than obstetricians (95% CI: 1.26 – 1.44) and 24% more diagnostic imaging services (95% CI: 1.18 – 1.31) than mothers of singletons after adjusting for confounding factors (Table 3). In the 12 months postpartum, women with twins accessed 140% more inpatient events (95% CI: 2.34 – 2.45), 413% more specialist consultations (95% CI: 4.87 – 5.40), 277% more pathology tests (95% CI: 3.51 – 4.08), 301% more diagnostic imaging services (95% CI: 3.69 – 4.34) and 112% more other MBS services (95% CI: 2.05 – 2.18) for the babies’ healthcare.
What does health care services cost women with twin pregnancies?
Women with twins incurred more out of pocket fees during pregnancy ($1,247.71) and their babies’ health service use in the time from birth to their first birthday ($700.69) compared with mothers ($1,077.32) and babies ($210.10) from a singleton pregnancy (Table 4). Women with twins were charged 1.16 times the amount of out of pocket fees for the health services accessed in pregnancy and 3.34 times more for the health services of her babies, relative to families with a singleton pregnancy.
Table 4. Adjusted* mean out of pocket fees for women who had a twin pregnancy compared with women that had a singleton pregnancy.
| Women with a twin pregnancy | Women with a singleton pregnancy | Cost Ratio Twins Compared with Singletons (95% CI) | |
| Mean out-of-pocket fees for women’s health service use in pregnancy | $1,247.71 | $1,077.32 | 1.16 (1.11 – 1.21) |
| Mean out-of-pocket fees for health service use of babies | $700.69 | $210.10 | 3.34 (3.23 – 3.45) |
*Adjusted for mother’s age, Indigenous identification, smoking status at 20 weeks, rurality and area-based socioeconomic status.
Discussion
Overall, women who have twins have a more intensive medical experience from the antenatal period through to 12 months postpartum than women who have a singleton child. This includes a greater number of appointments and scans during the antenatal period, higher rates of caesarean section birth, more admissions to the SCN and NICU, and an extended hospital stay after birth in comparison with women with a singleton birth.
In the postpartum period, women with twins accessed health services for the babies more frequently, with an average of 6 appointments per month compared with 2 appointments per month for women with singleton pregnancies. This is more than double the number of services as would be expected with having two babies. The greater health service use is reflected in the higher out-of-pocket fees for women with twin pregnancy compared with women with a singleton pregnancy, particularly for the babies’ health service use. Overall, women who have twins incur out-of-pocket fees for their babies’ health service use 3.34 times more often than similar aged women with a baby from a singleton pregnancy.
Conclusion
Overall the implications of these differences in pregnancy and birth-related events, health service utilisation and out of pocket healthcare fees between women who have twins and those who have single births is substantial. What these results cannot show, however, are the ongoing implications of factors such as the time (and associated costs of transport, etc.) it takes women and their partners/families to access the additional health services that their twins require, including the additional implications this has for working lives, as well as the social and mental health implications of the additional burdens of caring for two babies rather than just one18.
Recent collaborative initiatives such as that highlighted by Twins Research Australia’s Multiple Perspectives Discussion Paper19 and the joint Australian Multiple Births Association/Per Capita Multiples Matter report20 have documented these additional burdens and made recommendations for research, education, policy and practice initiatives to address them. However, in Australia, at least, there has been little or no recognition at a local, National or policy level of the issues and challenges faced during pregnancy, birth and the early life of twins and mulitples and the ongoing, long-term effect these factors impose on women and their (extended) families21.
Further research is needed to consider the implications of the findings contained in this Conversation Series article, to help address the inequities inherent in the current system and to provide additional support for the urgent development of policy solutions.
References
- Smits J, Monden C. Twinning across the developing world. PLoS One. 2011;6(9):e25239.
- Astolfi P, Ulizzi L, Zonta L. Changes in twinning rate: Italy 1950–1996. Human Reproduction. 2003;18(1):207-211.
- Westergaard T, Wohlfahrt J, Aaby P, Melbye M. Population based study of rates of multiple pregnancies in Denmark, 1980-94. BMJ. 1997;314(7083):775.
- Hall JG. Twinning. The Lancet. 2003;362(9385):735-743.
- Hoekstra C, Zhao ZZ, Lambalk CB, et al. Dizygotic twinning. Human reproduction update. 2007;14(1):37-47.
- Bortolus R, Parazzini F, Chatenoud L, Benzi G, Bianchi MM, Marini A. The epidemiology of multiple births. Human reproduction update. 1999;5(2):179-187.
- Young BC, Wylie BJ. Effects of twin gestation on maternal morbidity. Paper presented at: Seminars in perinatology2012.
- Walker MC, Murphy KE, Pan S, Yang Q, Wen SW. Adverse maternal outcomes in multifetal pregnancies. BJOG: An International Journal of Obstetrics & Gynaecology. 2004;111(11):1294-1296.
- Dodd JM, Robinson JS, Crowther CA, Chan A. Stillbirth and neonatal outcomes in South Australia, 1991-2000. American journal of obstetrics and gynecology. 2003;189(6):1731-1736.
- Santana DS, Silveira C, Costa ML, et al. Perinatal outcomes in twin pregnancies complicated by maternal morbidity: evidence from the WHO Multicountry Survey on Maternal and Newborn Health. BMC pregnancy and childbirth. 2018;18(1):449.
- Australian Institute of Health and Welfare. National Core Maternity Indicators. Canberra, ACT: AIHW;2019.
- South Australia Health. Twin pregnancy. Adelaide, South Australia: SA Health;2018.
- Liu Y, Davey M, Lee R, Palmer K, Wallace E. Changes in the modes of twin birth in Victoria, 1983-2015. The Medical Journal of Australia. 2020 Feb;212(2):82-88. doi: 10.5694.
- Callander EJ, Fox H. What are the costs associated with child and maternal healthcare within Australia? A study protocol for the use of data linkage to identify health service use, and health system and patient costs. BMJ Open. 2018;8(2):e017816.
- Australian Institute of Health and Welfare. Australia’s mothers and babies 2015. Canberra: AIHW;2017.
- Bell P. GREGWT and TABLE macros-Users guide. In. Canberra: ABS; 2000.
- Australian Institute of Health and Welfare. Record—Department of Health Broad Type of Service (BTOS) hierarchy, code NNNN. AIHW. Published 2019. Accessed 27 Feb, 2019.
- Calais Ferreira L, Kane S, Malta S. Twins can face unique risks before and just after birth. Pursuit [The University of Melbourne]. 2022 June; available at: https://pursuit.unimelb. edu.au/articles/twins-can-face-unique-risks-before-and-just-after-birth
- Twins Research Australia (TRA). Multiple Perspectives. Better health and wellbeing for families of twins and multiples: A discussion paper. TRA, The University of Melbourne, 2019.
- D’Rosario M. Multiple Matters. Twice the joy, twice the support: Investigating the financial and emotional needs of multiple birth families. Australian Multiple Births Association (AMBA) & Per Capita, 2023.
- Australian Institute of Health and Welfare. Australia’s mothers and babies 2023. Canberra: AIHW; 2023.
* Based on data categorisation on the PDC.
