Offspring birthweight and placental weight in immigrant women from conflict‐zone countries; does length of residence in the host country matter? A population study in Norway

We aimed to estimate differences in offspring birthweight and placental weight between Norwegian women and immigrants in Norway from countries with armed conflicts. We also studied whether length of residence in Norway was associated with offspring birthweight and placental weight.

Immigrants from non-western countries represent an increasing proportion of women who give birth in Europe and the USA, 6 and women from Africa and Asia give birth to offspring with lower birthweight than native women. [7][8][9][10][11] Such differences in birthweight have partly been attributed to genetic factors, and it has been suggested that ethnicity-specific fetal growth centiles should be used to better identify the pregnancies and offspring with true increased risk of adverse outcome. [12][13][14][15] A counter argument for ethnicity-specific growth centiles has been that the lower offspring birthweight in immigrants is caused by poor living conditions, and that poor living conditions for the mother compromise fetal growth. 16,17 If this is true, it is to be expected that improvements of deprived women's living conditions will lead to higher offspring birthweight. Hence, long-term residence in an affluent country for women emigrating from deprived regions could possibly increase their offspring's birthweight. However, a study from Sweden 18 found only marginal changes in offspring birthweight with the immigrants' duration of residence. Another study from the USA reported an initial decline in offspring birthweight after immigration. 19 Changes in offspring birthweight after emigration could possibly depend on the level of change in living conditions. 6 Particularly, women who live in an armed conflict region may have been exposed to deprived living conditions and thus be at high risk of giving birth to a low-birthweight infant. 20 The placenta is essential for fetal growth and development. 21 Placental dysfunction may therefore impair fetal growth, and a small placenta could indicate impaired placental function. [22][23][24] We are not aware of any studies of placental weight in immigrants or of changes in placental weight by duration of residence in a host country.
In the present study, we compared offspring birthweight and placental weight in women from Somalia, Iraq, and Afghanistan who gave birth in Norway, with Norwegian women. We also studied birthweight and placental weight according to length of residence in Norway.

| Study design and study sample
We performed a population-based registry study in Norway during the years 1999-2014 by using data from the Medical Birth Registry 25 and the Central Person Registry of Norway. 26 Almost all births in Norway take place in hospitals, and antenatal and obstetric care is free of charge. All births have been compulsorily notified to the Medical Birth Registry since 1967.
Women from countries with armed conflict who gave birth in Norway in gestational week 28 and beyond were included, as well as births to Norwegian mothers. Somali, Iraqi, and Afghan women represented the largest immigrant groups from countries with armed conflict, and women from these countries were therefore included.
Information about the women's country of birth and length of residence in Norway was obtained by individual linkage between the Medical Birth Registry and the Central Person Registry of Norway using the women's unique person identification numbers. All individuals with legal residence in Norway receive a unique person identification number. Women without legal residence, who give birth, receive a temporary person identification number. A woman was defined as Norwegian if both of her parents were born in Norway.
Placental weight has been reported to the Medical Birth Registry since 1999. Therefore, our study period started in 1999.
A total of 711 073 births were eligible for our study (Figure 1). We ex-

| Study factors
Our outcome variables were birthweight and placental weight, in grams. According to obstetric standards in Norway, the placenta is weighed at the obstetric ward with membranes and umbilical cord attached, within 1 hour after delivery. 27 Birthweight and placental weight may be associated with maternal age, parity, year of delivery, gestational age at delivery, maternal diabetes, and preeclampsia, and these factors may also be associated with country of birth. We therefore made adjustment for these factors, and the information was obtained from the Medical Birth Registry.
Maternal age at delivery was coded as <20, 20-34, or ≥35 years old, parity was coded as none, 1, 2, or 3 or more previous deliveries. Year of delivery was coded as 1999-2002, 2003-2006, 2007-2011, or 2011-2014. Gestational age at delivery was coded as 28-36 weeks, 37-41 weeks, or ≥42 weeks. Maternal diabetes was reported to the Medical Birth Registry as diabetes mellitus type 1, diabetes mellitus type 2, gestational diabetes, and unspecified diabetes, and was coded in our analyses as diabetes (yes or no). Preeclampsia was defined as blood pressure 140/90 mm Hg or higher after 20 weeks of gestation in addition to proteinuria (more than 0.3 g/24 h or +1 on dipstick).
Length of residence in Norway was coded as <5 years or ≥5 years.

| Statistical analyses
Mean birthweight and placental weight according to the women's country of birth were calculated, and differences in mean were tested by applying Student's t test. We used crude and multivariable linear regression analyses to estimate differences by country of birth in birthweight (in grams) and in placental weight (in grams), with 95% CI,

Key message
Immigrant women from Somalia, Afghanistan, and Iraq deliver offspring with lower weight than Norwegian women, but the difference is reduced by length of residence in Norway.
using births to Norwegian women as the reference group. In multivariable analyses, we made adjustments for maternal age, parity, year of delivery, gestational age at delivery, diabetes, and preeclampsia. First, we studied all women. Thereafter, we studied immigrant women with <5 years and ≥5 years of residence in Norway, separately.

| Ethical approval
This research project was approved by the Norwegian Regional Committee for Medical and Health Research Ethics (reference number 2012/1433).

| RE SULTS
Immigrant women from Somalia, Afghanistan, and Iraq represented an increasing proportion of the women who gave birth in Norway during our study period, from 1.3% in 1999-2002 to 3.9% in 2011-2014 (Table 1). Immigrant women were more often multiparous than Norwegian women. They had higher prevalence of diabetes, but lower prevalence of preeclampsia.

| Birthweight
Mean birthweight was 3578 g for offspring of Norwegian women, but in offspring of immigrant women, birthweight was 171 g lower, on average (Table 2; Figure 2). After adjustment for maternal age, parity, gestational week at delivery, year of delivery, diabetes, and preeclampsia, the difference in offspring birthweight between immigrant women and Norwegian women increased, and birthweight was estimated to be 206 g lower in immigrants than among Norwegian women. Offspring of Somali women had the lowest crude birthweight, at 185 g below the Norwegian mean, and the adjusted difference in birthweight was 245 g below the Norwegian mean (Table 2). Immigrant women with ≥5 years of residence in Norway had higher offspring birthweight than immigrant women with shorter length of residence (Table 2). Women from Afghanistan had the greatest increase in offspring birthweight according to length of residence, and mean offspring birthweight was 66 g higher in Afghan women with 5 or more years of residence compared with Afghan women with shorter length of residence (Table 2; Figure 2).

| Placental weight
Mean placental weight was 675 g in Norwegian women. Immigrant women had lower placental weight, on average at 7 g lower than the Norwegian mean (Table 2; Figure 2). After adjustment for the above study factors, the difference between immigrant women and Norwegian women increased, and mean placental weight was estimated to be 16 g lower in immigrant women. Afghan women had the lowest crude mean placental weight, at 17 g below the Norwegian mean. After adjustment, placental weight in Afghan women was estimated to be 20 g below the Norwegian mean (Table 2).
Immigrant women with ≥5 years residence in Norway had higher placental weight than immigrant women with shorter length of residence (Table 2; Figure 2). In immigrants, the placenta was 24 g less than the Norwegian mean after <5 years of residence, and the placenta was 7 g less than the Norwegian mean after ≥5 years of residence (17 g increase). Women from Afghanistan had the greatest increase in mean placental weight, an increase of 20 g.
Interestingly, mean placental weight in Afghan and Iraqi women with ≥5 years of residence in Norway was not significantly different from the Norwegian mean (Table 2; Figure 2). The increase in placental weight by length of residence was relatively larger than for birthweight.

| D ISCUSS I ON
We found that offspring birthweight and placental weight were lower among Somali, Afghan, and Iraqi women than among Norwegian women, but the difference was reduced by length of residence in Norway.
We are not aware of any previous population study that has compared birthweight and placental weight in immigrant women from conflict zones with native women in a European country.
We aimed to include all women from Norway, Somalia, We made adjustments for maternal age, parity, year of delivery, gestational age at delivery, diabetes, and preeclampsia, and such adjustment strengthened our findings. Offspring birthweight, and placental weight are known to be positively associated with maternal body mass index. 3,28,29 Hence, the lower offspring birthweight and placental weight in Somali, Afghan, and Iraqi women could be explained by lower body mass index in immigrant women.
Unfortunately, we had no information about maternal body mass index in our study.
Both birthweight and placental weight increase by gestational age at birth, 30 and immigrant women may have increased risk of preterm delivery. 31 Also, our findings could be explained by changes during our study period in the selection of women who have immigrated to Norway. However, adjustments for gestational age and year of delivery did not alter the direction of our estimates.
In agreement with our results, several previous studies report that immigrant women from developing countries give birth to infants with lower birthweight than European women. [32][33][34] In particular, women from conflict-zone countries seem to give birth to low birthweight infants. 7,20 We are not aware of any previous studies of placental weight in women who have recently immigrated. However, two studies from the USA found higher prevalence of placental growth restriction in African American women than among white women, 29,35 and an Australian study found higher placental weight in people of Asian ethnic origin than in those of European origin. 22 A Malaysian study compared women of Indian, Chinese, and Malay origin, and they found significant differences in offspring birthweight and placental weight between these groups. 36 A study of placentas in very-low-birthweight TA B L E 2 Crude and adjusted* difference in mean offspring birthweight and placental weight (in grams) between Norwegian women (reference) and immigrant women, according to the mother's country of birth and length of residence in Norway (<5 y or ≥5 y) Note: 95% CI are presented and Norwegian women are used as reference.
*Adjustments were made for maternal age, parity, year of delivery, gestational age at delivery, diabetes, and preeclampsia.
infants in the USA found no difference in placental weight according to ethnicity. 37 A Canadian study suggested increasing risk of preeclampsia according to length of residency in Canada. 38 We are aware of four previous studies that have addressed whether length of residence in a host country is associated with offspring birthweight. Neither a Swedish nor a Canadian study found significant trends in offspring birthweight by length of residence. 18,39 However, two studies from the USA found low offspring birthweight among the women with the shortest and the longest lengths of residence. 19,40 For those women, the standard of living may not have increased after immigration.
It is under discussion whether the observed differences in birthweight between ethnic groups are caused by genetic differences, 33,41,42 or by differences in intrauterine exposures. 16,32,43,44 A study from the USA found an association between low birthweight and foreign-born status, as well as with educational level. 45 Also, a Swedish study suggests that the mother's ethnicity does not sufficiently explain the differences in offspring birthweight between mothers born in Sweden and mothers born in developing countries. 46 A systematic review concluded that immigrant status in itself is not a good marker for risk of adverse perinatal outcomes. 6 Besides maternal nutritional status, exposure to stress may also influence birthweight. 47 Our findings suggest that settlement in a wel-

| CON CLUS ION
Immigrant mothers in Norway from Somalia, Afghanistan, and Iraq had lower offspring birthweight and placental weight than Norwegian mothers. This difference was reduced by length of residence in Norway. Our findings suggest that changes in living conditions may influence birthweight and placental weight, and that valid ethnicity-specific fetal growth centiles may be difficult to establish.

CO N FLI C T O F I NTE R E S T
There are no conflicts of interest.