Between 2009 and 2018 there were 708,475 women who birthed a singleton baby in Victoria. The demographics of the study population are presented in the Additional file 1, Table 1. Most women were born in Australia (67%; n = 461,903). The remaining women were migrants from South East Asia and Sub-Saharan Africa (17.9%; n = 126,640); the Americas, Europe, and Oceania (9.1%; n = 64,195); and North East Africa, North Africa, and the Middle East (7.9%; n =55,737). The mean birthweight for the Victorian study population (3387.39g, SD = 543.4g) was 59.4g (95%CI: 58.1 to 60.7, p < .001) higher than the Australian 2017 population mean birthweight of 3328g [44].
A smaller proportion of migrant women than Australian born women were not in a relationship (6.7% vs 14.6% respectively); under the age of 20 years (0.8% vs 2.5%); and SEIFA advantaged (36.9% vs 41.6%). Migrant women were also more likely than Australian born women to be birthing their first baby (45.3% vs 43% respectively); classified as short height (12.5% vs 28%); develop more gestational diabetes managed with diet (9.5% vs 4.0) and insulin (6.1% vs 2.8%); and experience more suspected fetal growth restriction (5.8% vs 4.3%). The majority of women in the study population were non-smokers. Those that did smoke were more likely to be Australian born women rather than migrant women, both before 20 weeks gestation (3.8% vs 13.4%); and after 20 weeks gestation (1.9% vs 8.5%). These findings were statistically significant at p < 0.001, and are presented in the Additional file 1, Table 2.
Maternal Region of Origin and SGA
There was a total of 61,564 SGA neonates in the study population, resulting in an overall SGA rate of 8.7%. A higher proportion of SGA was identified for migrant women at 11.3% (n=27,815) compared to 7.3% (n=33,749) for Australian born women. We found significant variation in the distribution of SGA across maternal regions of origin (p < 0.001). The proportion of SGA for migrant women from South East, South Central Asia, and Sub-Saharan Africa 14% (n= 17,687); North East, North Africa, and the Middle East 10% (n=5,563) was higher than for women from Australia 7.3% (n=33,749); the Americas, Europe, and Oceania 7.1% (n=4,565). After controlling for confounding factors, regression analysis identified migrant women were 1.48 times more likely to birth an SGA baby than women born in Australia (95%CI: 1.45 to 1.51, p <.001), see Table 1.
Table 1.
Adjusted odds ratio (95%CI) independent association between SGA and migrant/non migrant women.
Variable
|
SGA aOR (95%CI)
|
Marital Status
|
In a Relationship
|
1
|
|
Not in a Relationship
|
1.06 (1.03 to 1.10) ***
|
Age Group (years)
|
< 20
|
0.78 (0.72 to 0.85) ***
|
|
20 to 35
|
0.91 (0.89 to 0.94) ***
|
|
35+ r
|
1
|
Parity
|
Primiparous
|
2.05 (2.01 to 2.10) ***
|
|
Multiparous r
|
1
|
BMI
|
Underweight
|
2.06 (1.96 to 2.17) ***
|
|
Healthy Weight
|
1.28 (1.26 to 1.31) ***
|
|
Overweight r
|
1
|
Height
|
Tall r
|
1
|
|
Average
|
1.63 (1.59 to 1.68) ***
|
|
Short
|
2.77 (2.69 to 2.85) ***
|
Smoking before 20wks
|
Yes
|
1.13 (1.06 to 1.20) ***
|
|
No r
|
1
|
Smoking after 20wks.
|
Yes
|
2.20 (2.05 to 2.37) ***
|
|
No r
|
1
|
Gestational Age at 1st Visit
|
Before 11 weeks r
|
1
|
|
12 to 23 weeks
|
1.07 (1.04 to 1.09) ***
|
|
24 weeks /no care
|
1.25 (1.21 to 1.30) ***
|
Type 1 Diabetes
|
Yes r
|
1
|
|
No
|
2.98 (2.29 to 3.89) ***
|
Pre-existing Hypertension
|
Yes
|
1.26 (1.15 to 1.37) ***
|
|
No r
|
1
|
Gestational Diabetes - Insulin
|
Yes r
|
1
|
|
No
|
1.29 (1.22 to 1.36) ***
|
Preeclampsia
|
Yes
|
1.44 (1.35 to 1.53) ***
|
|
No r
|
1
|
HELLP Syndrome
|
Yes
|
1.68 (1.38 to 2.04) ***
|
|
No r
|
1
|
Suspected FGR
|
Yes
|
10.25 (9.97 to 10.54) ***
|
|
No r
|
1
|
Maternal SEIFA
|
Advantaged r
|
1
|
|
Average
|
1.05 (1.02 to 1.08) **
|
|
Disadvantage
|
1.05 (1.02 to 1.07) ***
|
Maternal Region of Origin
|
Australia Born Women r
|
1
|
|
Migrant Women
|
1.48 (1.45 to 1.51) ***
|
Note. 1. Birthweight Adjusted for Sex and Gestational Age. 2. r = reference. 3. ***p <0.001, ** p < 0.01.
|
Independent association between SGA and maternal region of origin.
Multivariate modelling was performed to identify the differences in SGA between maternal regions of origin groups, see Table 2. A statistically significant independent association between SGA and maternal region of origin groups was identified. In comparison to Australian born women, migrant women from the Americas, Europe, and Oceania were more likely to birth an SGA child (aOR 1.06, 95%CI: 1.02 to 1.12, p = .003); women from North Africa, North East Africa, and the Middle East were 1.4 times more likely to birth an SGA child (95%CI: 1.35 to 1.45, p < .001) and women from South East Asia, South Central Asia, and Sub-Saharan Africa were 1.75 times more likely to birth an SGA child (95%CI: 1.70 to 1.80, p < .001).
Table 2:
Adjusted odds ratio (95%CI) independent association between SGA and maternal region of origin.
Maternal Region of Origin
|
SGA aOR (95%CI)
|
Australia r
|
1
|
Americas, Europe, Oceania
|
1.06 (1.02 to 1.12) *
|
North & North East Africa, Middle East
|
1.40 (1.35 to 1.45) ***
|
South East, South Central Asia, Sub Saharan Africa
|
1.75 (1.70 to 1.80) ***
|
Note. 1. Birthweight Adjusted for Sex and Gestational Age. 2. r = reference. 3. ***p <0.001. * p < 0.05. 4. Confounding variables included, marital status, maternal age group, parity, BMI, height group, smoking before or after 20 weeks gestation, gestational age at first visit, type 1 diabetes pre-existing hypertension, gestational diabetes -insulin, pre-eclampsia, HELLP syndrome, suspected FGR.
Socioeconomic status has previously been identified as a strong influencing factor in the association with SGA. Therefore, additional regression analysis was undertaken to assess for collinearity with the independent variable region of origin. A series of modelling stratified by SEIFA advantaged, average, and disadvantaged was performed to confirm the relationships between SGA and region of origin (see Table 3.). The independent association between maternal region of origin and SGA was statistically significant across all SEIFA groups except for average and disadvantaged women from the Americas, Europe, and Oceania. These findings indicate for migrant women in Victoria, the risk of SGA did not follow a classic socioeconomic gradient of disadvantage, confirming maternal region of origin was a much stronger predictor of SGA than maternal socioeconomic status.
Table 3.
Independent association between SGA and maternal region of origin stratified SEIFA advantage, average, disadvantaged.
SGA OR (95%CI)
|
Maternal Region of Origin
|
SEIFA Advantaged
|
SEIFA Average
|
SEIFA Disadvantaged
|
Australia r
|
1
|
1
|
1
|
Americas, Europe, Oceania
|
1.15 (1.08 to 1.22) ***
|
1.01 (0.92 to 1.11)
|
0.99 (0.92 to 1.06)
|
North & North East Africa, Middle East
|
1.38 (1.30 to 1.47) ***
|
1.40 (1.29 to 1.53) ***
|
1.40 (1.32 to 1.49) ***
|
South East, South Central Asia, Sub Saharan Africa
|
1.83 (1.75 to 1.91) ***
|
1.84 (1.73 to 1.95) ***
|
1.67 (1.61 to 1.74) ***
|
Note. 1. Birthweight Adjusted for Sex and Gestational Age. 2. r = reference. 3. ***p <0.001. 4. Confounding variables included, marital status, maternal age group, parity, BMI, height group, smoking before or after 20 weeks gestation, gestational age at first visit, type 1 diabetes pre-existing hypertension, gestational diabetes -insulin, pre-eclampsia, HELLP syndrome, suspected FGR.
Being born SGA was also associated with several maternal characteristics, medical conditions, and pregnancy complications, see Table 1. For the total study population, lower risks of SGA were associated with younger age: a risk reduction of 23% at < 20 years (aOR 0.78, 95%CI: 0.72 to 0.85, p < 0.001), and 11% at age 20 to 35 years (aOR 0.91, 95%CI: 0.89 to 0.94, p < .001), compared with women over 35 years of age. Women birthing their first baby were twice as likely (aOR 2.05, 95%CI: 2.01 to 2.10, p <.001) to birth an SGA child than women birthing subsequent babies. Shorter women (< 25th centile of the study population) were 2.77 times more likely (95%CI: 2.69 to 2.85, p < .001), and average height women (add the centile range here to make it clear) 1.63 times more likely (95%CI: 1.59 to 1.68, p < .001), to birth an SGA baby than tall women (> 75th centile of the study population). Being underweight doubled a woman’s odds of birthing an SGA baby (aOR 2.06, 95%CI: 1.96 to 2.17, p < .001) compared to women who were overweight or obese.
Women not accessing care until after 24 weeks gestational or not at all were 25% more likely to birth an SGA child than women accessing maternity care before 12 weeks gestation (95%CI: 1.21 to 1.30, p < .001). The risk of SGA increased in proportion to the severity of hypertensive disorders from 26% more likely with pre-existing hypertension, 44% more likely with pre-eclampsia, to 68% more likely with HELLP syndrome (p < .001). Both type 1 diabetes (aOR 2.98, 95%CI: 2.29 to 3.89, p < .001), and gestational diabetes treated with Insulin (aOR 1.29, 95%CI: 1.22 to 1.36, p < .001), were protective, decreasing the odds of SGA compared to women without these conditions. Women with suspected fetal growth restriction were over 10.25 times more likely to birth an SGA child than women not suspected of fetal growth restriction (95%CI: 9.97 – 10.54, p < .001).