Aboriginal and Torres Strait Islander mothers and babies

In addition, babies of Aboriginal and Torres Strait Islander mothers and Aboriginal and Torres Strait Islander babies have higher rates of pre-term birth, low birthweight and being small for gestational age than babies of non-Indigenous mothers and non-Indigenous babies.

These findings were based on a comparison of proportions between groups for the characteristics of interest. However, this analysis does not consider the relationships between multiple characteristics and may not accurately reflect the importance of the characteristic on the outcome of interest.

Many of the topics covered in this report are complex and interrelated, and there are differences between the population of Aboriginal and Torres Strait Islander mothers and the population of non-Indigenous mothers. A multiple regression modelling approach was therefore used to look at the effect of Indigenous status on selected perinatal outcomes.

Multiple regression analyses can control for the effects of multiple characteristics at once and, as a result, can better estimate the influence of these characteristics on the outcome of interest.

The modelling was used to ascertain whether the risk of an outcome was higher, after accounting for the effect of possible confounding variables, for the populations of:

• Aboriginal and Torres Strait Islander females who gave birth compared with non-Indigenous females
• babies of Aboriginal and Torres Strait Islander mothers compared with babies of non-Indigenous mothers
• Aboriginal and Torres Strait Islander babies compared with non-Indigenous babies.

For most outcomes, the estimated relative difference (risk ratio) between Aboriginal and Torres Strait Islander mothers (or babies) and non-Indigenous mothers (or babies) was greatly reduced after adjusting for the confounding effect of the explanatory characteristics included in the models. Please note that there are many other potential confounding factors that are not included in the regression models, as these are not available from the National Perinatal Data Collection.

Some of the difference observed between Aboriginal and Torres Strait Islander and non-Indigenous mothers and babies can therefore be explained by differences in explanatory characteristics, such as maternal demographics and characteristics relating to the antenatal period, labour and birth, and birth outcomes.  Addressing these differences would reduce the difference between the Aboriginal and Torres Strait Islander and non-Indigenous populations.

Table 1 lists the explanatory variables included in the models. It is important to note that not all explanatory variables are relevant to all outcomes (see Table 2), and that explanatory variables were chosen based on their potential biological or social importance to the outcome and their availability in the data collection.

Table 1: Description of explanatory variables used in modelling

Explanatory variable	Description
Maternal Indigenous status	Indigenous status of the mother
Baby Indigenous status	Indigenous status of the baby
Maternal age	Grouped maternal age in completed years at delivery
Parity	Total number of previous pregnancies that resulted in a live or stillbirth
Remoteness area	Remoteness area of mother’s usual residence
Socioeconomic status	SEIFA IRSD quintile of mother's usual residence
Duration of pregnancy at first antenatal visit	Grouped duration of pregnancy at first antenatal care visit
Number of antenatal visits	Grouped number of antenatal visits
Maternal smoking status	Maternal smoking status at any time during pregnancy
Maternal BMI	Grouped maternal body mass index
Maternal diabetes status	Maternal diabetes status of pre-existing diabetes, gestational diabetes or ‘none or not stated diabetes’
Maternal hypertension status	Maternal hypertension status of pre-existing hypertension, gestational hypertension, or no hypertension
Hospital sector	Sector of the establishment where birth occurred
Onset of labour	The manner in which a labour started
Method for birth	How the baby was bon
Gestational age	Duration of pregnancy in completed weeks
Birthweight	The first weight of the baby within an hour of birth
Birthweight adjusted for gestational age	Birthweight percentile for gestational age and sex as determined by national percentiles

Table 2: Outcome of interest and explanatory variables used in modelling

Outcome of interest	Explanatory variables
Maternal risk factors
Smoked at any time during pregnancy	maternal Indigenous status, maternal age, parity, remoteness area and socioeconomic status
Pre-pregnancy obesity	maternal Indigenous status, maternal age, parity, remoteness area and socioeconomic status
Pre-existing diabetes	maternal Indigenous status, maternal age, parity, remoteness area and socioeconomic status
Gestational diabetes (for those don’t have pre-existing diabetes)	maternal Indigenous status, maternal age, parity, remoteness area and socioeconomic status
Pre-existing hypertension	maternal Indigenous status, maternal age, parity, remoteness area and socioeconomic status
Gestational hypertension (for those don’t have pre-existing hypertension)	maternal Indigenous status, maternal age, parity, remoteness area and socioeconomic status
Birth outcomes
Induced labour	maternal Indigenous status, maternal age, parity, remoteness area, socioeconomic status, duration of pregnancy at first antenatal care visit, number of antenatal care visits, maternal smoking status, maternal BMI, maternal diabetes status, maternal hypertension status and hospital sector
Caesarean section birth	maternal Indigenous status, maternal age, parity, remoteness area, socioeconomic status, duration of pregnancy at first antenatal care visit, number of antenatal care visits, maternal smoking status, maternal BMI, maternal diabetes status, maternal hypertension status and hospital sector
No analgesia administered	maternal Indigenous status, maternal age, parity, remoteness area, socioeconomic status, duration of pregnancy at first antenatal care visit, number of antenatal care visits, maternal smoking status, maternal BMI, maternal diabetes status, maternal hypertension status and hospital sector
Had an episiotomy	maternal Indigenous status, maternal age, parity, remoteness area, socioeconomic status, duration of pregnancy at first antenatal care visit, number of antenatal care visits, maternal smoking status, maternal BMI, maternal diabetes status, maternal hypertension status and hospital sector
Maternal postnatal hospital stay <7 days	maternal Indigenous status, maternal age, parity, remoteness area, socioeconomic status, duration of pregnancy at first antenatal care visit, number of antenatal care visits, maternal smoking status, maternal BMI, maternal diabetes status, maternal hypertension status, hospital sector, onset of labour and method of birth
Baby outcomes
Pre-term birth	maternal Indigenous status, baby Indigenous status, maternal age, parity, remoteness area, socioeconomic status, duration of pregnancy at first antenatal care visit, number of antenatal care visits, maternal smoking status, maternal BMI, maternal diabetes status, maternal hypertension status, hospital sector, onset of labour, method of birth and birthweight
Low birthweight	maternal Indigenous status, baby Indigenous status, maternal age, parity, remoteness area, socioeconomic status, duration of pregnancy at first antenatal care visit, number of antenatal care visits, maternal smoking status, maternal BMI, maternal diabetes status, maternal hypertension status, hospital sector, onset of labour, method of birth and gestational age
Small for gestational age	maternal Indigenous status, baby Indigenous status, maternal age, parity, remoteness area, socioeconomic status, duration of pregnancy at first antenatal care visit, number of antenatal care visits, maternal smoking status, maternal BMI, maternal diabetes status, maternal hypertension status, hospital sector, onset of labour and method of birth
Baby hospital stay greater than 7 days	maternal Indigenous status, baby Indigenous status, maternal age, parity, remoteness area, socioeconomic status, duration of pregnancy at first antenatal care visit, number of antenatal care visits, maternal smoking status, maternal BMI, maternal diabetes status, maternal hypertension status, hospital sector, onset of labour, method of birth, birthweight, gestational age and birthweight adjusted for gestational age

Maternal risk factors for Aboriginal and Torres Strait Islander and non-Indigenous mothers

Modelling of maternal risk factors was undertaken to ascertain whether Aboriginal and Torres Strait Islander mothers were more likely than non-Indigenous mothers to have selected risk factors, after controlling for demographic and socioeconomic factors.

After adjusting for the confounding effect of factors included in the model, in comparison to non-Indigenous mothers, Aboriginal and Torres Strait Islander mothers were:

• 2.4 times as likely to smoke at any time during pregnancy
• 1.1 times as likely to be obese
• 2.0 times as likely to have pre-existing diabetes
• 1.2 times as likely to have gestational diabetes
• 1.6 times as likely to have pre-existing hypertension.

Both groups had a similar risk of gestational hypertension (Table 3).

Table 3: Unadjusted (crude) and adjusted risk ratios for select antenatal period outcomes, comparing Aboriginal and Torres Strait Islander females to non-Indigenous females who gave birth, 2017-2020

Outcome	Unadjusted risk ratio	Adjusted risk ratio(a)
Smoked at any time during pregnancy(b)	*5.61 (5.55-5.67)	*2.35 (2.32-2.39)
Pre-pregnancy obesity	*1.53 (1.01-1.51)	*1.14 (1.12-1.16)
Pre-existing diabetes(c)	*2.59 (2.45-2.75)	*2.03 (1.89-2.17)
Gestational diabetes(c) (excludes those with pre-existing diabetes)	*1.02 (0.99-1.04)	*1.16 (1.13-1.19)
Pre-existing hypertension(c)	*1.53 (1.40-1.66)	*1.64 (1.49-1.81)
Gestational hypertension(c) (excludes those with pre-existing hypertension)	0.99 (0.95-1.04)	1.06 (1.01-1.12)

a) Adjusted for explanatory variables listed at Table 2.

b) Mother’s tobacco smoking status during pregnancy is self-reported.

c) Data excludes Vic.

* p = <0.0001

Birth outcomes for Aboriginal and Torres Strait Islander and non-Indigenous mothers

The purpose of modelling birth outcomes was to investigate whether Aboriginal and Torres Strait Islander mothers were more likely than non-Indigenous mothers to have selected birth outcomes, after controlling for demographic, socioeconomic, health care and known risk factors. 

After adjusting for the confounding effect of factors included in the model, in comparison to non-Indigenous mothers, Aboriginal and Torres Strait Islander mothers were:

• 0.9 times as likely to have an episiotomy or
• 0.9 times as likely to have no analgesia administered
• 1.6 times as likely to have a postnatal stay in hospital of 7 days or more.

The risk of induced labour and caesarean section birth was similar for both groups (Table 4).

Table 4: Unadjusted (crude) and adjusted risk ratios for select labour and birth outcomes, comparing Aboriginal and Torres Strait Islander females to non-Indigenous females who gave birth, 2017-2020

Outcome	Unadjusted risk ratio	Adjusted risk ratio(a)
Induced labour(b)	0.98 (0.97-0.99)	0.99 (0.97-1.00)
Caesarean section birth(c)	*0.85 (0.84-0.86)	*1.02(1.01-1.02)
No analgesia administered(d)	*1.04 (1.02-1.06)	*0.89 (0.87-0.91)
Had an episiotomy(e)	*0.49 (0.48-0.51)	*0.87 (0.85-0.90)
Maternal postnatal hospital stay <7 days(f)	*1.78 (1.67-1.89)	*1.56 (1.44-1.70)

a) Adjusted for explanatory variables listed at Table 2.

b) 'Induced' may include cases where induction of labour was attempted but labour did not result.

c) For multiple births, the method of birth of the first-born baby was used.

d) Only women who had a spontaneous or induced labour are included.

e) Only women who gave birth vaginally are included.

f) Only includes women who were discharged home. Excludes women who gave birth in birth centres attached to hospitals. For multiple births, the length of stay after the birth of the first-born baby was used.

* p = <0.0001

Baby outcomes

Modelling was undertaken to explore whether babies of Aboriginal and Torres Strait Islander mothers and Aboriginal and Torres Strait Islander babies were more likely than babies of non-Indigenous mothers and non-Indigenous babies to have an outcome after controlling for demographic, socioeconomic, health care, obstetric and known risk factors. 

Babies of Aboriginal and Torres Strait Islander and non-Indigenous mothers

After adjusting for the confounding effect of factors included in the model, in comparison to babies of non-Indigenous mothers, babies of Aboriginal and Torres Strait Islander mothers were:

• 1.1 times as likely to be low birthweight
• 1.3 times as likely to be small for gestational age
• 1.2 times as likely to have a hospital stay of 7 days or more.

The risk of pre-term birth was similar for both groups (Table 5).

Table 5: Unadjusted (crude) and adjusted risk ratios for select baby outcomes, comparing babies of Aboriginal and Torres Strait Islander mothers to babies of non-Indigenous mothers, 2017-2020

Outcome	Unadjusted risk ratio	Adjusted risk ratio(a)
Pre-term birth	*1.65 (1.62-1.69)	1.04 (1.01-1.06)
Low birthweight (b)	*1.88 (1.83-1.92)	*1.12 (1.01-1.15)
Small for gestational age(c)	*1.47 (1.44-1.51)	*1.25 (1.22-1.28)
Baby hospital stay greater than 7 days(d)	*1.90 (1.84-1.96)	*1.18 (1.13-1.22)

a) Adjusted for explanatory variables listed at Table 2

b) Includes liveborn babies only.

c) Includes liveborn singleton babies only.

d) Includes liveborn babies born in hospital and discharged home.

* = p <0.0001

Aboriginal and Torres Strait Islander and non-Indigenous babies

After adjusting for the confounding effect of factors included in the model, in comparison to non-Indigenous babies, Aboriginal and Torres Strait Islander babies were:

• 1.1 times as likely to have a low birthweight
• 1.2 times as likely as non-Indigenous babies to be small for gestational age
• 1.2 times as likely to have a hospital stay of 7 days or more.

The risk of pre-term birth was similar for both groups (Table 6).

Table 6: Unadjusted (crude) and adjusted risk ratios for select baby outcomes, comparing Aboriginal and Torres Strait Islander babies to non-Indigenous babies, 2017-2020

Outcome	Unadjusted risk ratio	Adjusted risk ratio(a)
Pre-term birth	*1.58 (1.55-1.61)	1.03 (1.01-1.06)
Low birthweight(b)	*1.77 (1.73-1.81)	*1.11 (1.08-1.14)
Small for gestational age(c)	*1.41 (1.38-1.44)	*1.20 (1.17-1.23)
Baby hospital stay greater than 7 days(d)	*1.80 (1.75-1.86)	*1.17 (1.13-1.21)

a) Adjusted for explanatory variables listed at Table 2

b) Includes liveborn babies only.

c) Includes liveborn singleton babies only.

d) Includes liveborn babies born in hospital and discharged home.

* = p <0.0001

Summary

Aboriginal and Torres Strait Islander mothers had a higher risk of smoking during pregnancy, obesity, pre-existing diabetes and a postnatal hospital stay of 7 days or more, when compared with non-Indigenous mothers, for both unadjusted and adjusted risk ratios.

However, when comparing the relative difference between Aboriginal and Torres Strait Islander and non-Indigenous mothers, this difference was reduced after adjusting for confounding factors. This suggests that a large component of the difference in risks between the two groups is associated with the differences in the explanatory variables entered into the model.

Interestingly, some outcomes for Aboriginal and Torres Strait Islander mothers showed a slightly increased relative difference in risk when comparing unadjusted and adjusted risk ratios, including gestational diabetes and pre-existing hypertension. This was also seen for the labour and birth interventions of caesarean section birth – which showed a similar reduction in the difference in risks for both Aboriginal and Torres Strait Islander and non-Indigenous mothers after adjusting for confounding – and episiotomy.

The pattern seen for both babies of Aboriginal and Torres Strait Islander mothers and Aboriginal and Torres Strait Islander babies was the same, although risk ratios differed slightly. Both babies of Aboriginal and Torres Strait Islander mothers and Aboriginal and Torres Strait Islander babies had a higher risk of pre-term birth, being of low birthweight, being small for gestational age and having a hospital stay of 7 days or more, when compared with babies of non-Indigenous mothers and non-Indigenous babies.

However, after controlling for the effect of explanatory variables included in the model, the adjusted differences (risk ratios), of these outcomes are reduced for both groups. These findings indicate that the differences in perinatal outcomes between babies of Aboriginal and Torres Strait Islander mothers and babies of non-Indigenous mothers, and Aboriginal and Torres Strait Islander babies and non-Indigenous babies, are largely associated with the explanatory variables entered into the model.

Where the estimated relative difference remains high for an Aboriginal or Torres Strait Islander mother or baby after adjusting for potential confounding, this is potentially due to limitations in data availability – such as alcohol consumption during pregnancy, which is available but has not been collected long enough to include in the modelling – rather than a reflection of intrinsic risk in the population.