更改国籍对布鲁塞尔移民的围产期死亡率及其原因的影响
Judith Racape, Myriam De Spiegelaere, Michèle Dramaix, Edwige Haelterman and Sophie Alexander
2015-05-19
全文
Objectives
Perinatal mortality rates vary between ethnic groups but the relation with immigrant status is unclear. Previous research suggested that birth outcomes may either improve or deteriorate with duration of residence, depending on the migrant group. The objectives of this study are to describe and measure inequalities in pregnancy outcomes, perinatal mortality and causes of perinatal deaths according to current citizenship versus national origin of the mother, in Brussels.
Study design
This is a population-based cohort study using data from linked birth and death certificates from the Belgian civil registration system. The data relate to all babies born between 1998 and 2008, whose mothers were living in Brussels, irrespective of the place of delivery. We used a logistic regression to estimate the odds ratios (ORs) for the association between mortality, causes of deaths and nationality.
Results
Women from Morocco, sub-Saharan Africa and Turkey experience an 80% excess in perinatal mortality (p < 0.0001) compared to Belgians, but this excess of perinatal mortality is not observed for mothers with Belgian citizenship at delivery. For sub-Saharan African women, this excess is caused mainly by immaturity-related conditions and reflects a high rate of preterm deliveries, low birth weight and a low socio-economic level. Moroccan and Turkish mothers have favourable pregnancy outcomes that persist after adopting Belgian nationality, but they experience a strong excess of perinatal mortality, mainly due to congenital anomalies and asphyxia or unexplained deaths prior to the onset of labour.
Conclusion
In Brussels, perinatal mortality varies according to nationality but those differences do not persist after adopting Belgian nationality. The explanation of this positive effect is probably due to a mix of determinants such as acculturation, use of health services or cultural contexts. Further analysis should help to better understand the results observed.
1. Introduction
Perinatal outcomes have been extensively studied among immigrant women in different countries, showing conflicting results [1] and [2]. Some studies indicate immigrant status as a risk factor, showing higher rates of perinatal mortality, prematurity and low birth weight [3] and [4], but perinatal outcomes vary according to the mother's country of origin and the receiving country [5]. In contrast, other studies show immigrant status as a positive factor. A number of them refer to the “epidemiological paradox”, showing that despite their low socioeconomic status, immigrants have good pregnancy outcomes. The relation with the immigrant status is unclear, however, and several studies in the United States suggest that health outcomes differ by generational status and duration of residence. Longer duration of residence in Mexican immigrants is associated with a change in health status and birth outcomes [6]. Birth outcomes may either improve or deteriorate with length of residence depending on the migrant group [1]. A common framework for the interpretation of these patterns is provided by the acculturation theory, which is a sociocultural process in which members of one cultural group adopt the beliefs and behaviours of another group [7].
A previous study relating perinatal mortality to the nationality of the mother at delivery in Brussels has shown a clear pattern of inequalities [2]. The aim of the present study was to investigate the impact of adopting Belgian citizenship on perinatal outcomes among immigrants in Brussels. In Belgium, nationality can be acquired in three ways: statement, marriage or naturalisation. Statement and marriage are open to adults who meet a series of requirements (related to residence, number of years of residence, place of birth, attachments to Belgium). For naturalisation, a main residence, plus a legal residence in Belgium for at least three years (2 years for refugees) is required. The procedure takes at least 18 months but does not depend on cultural or language knowledge, or on medical conditions [8]. The objectives of this study were to describe and measure inequalities in pregnancy outcomes, perinatal mortality and causes of perinatal deaths according to the current citizenship, versus national origin, of mothers in Brussels in 1998–2008.
return to article outline
2. Materials and methods
2.1. Population and data
This study is a population-based cohort study using data from linked birth and death certificates from the Belgian civil register. Data used in the present study relate to all babies born from 1998 to 2008 whose mothers were living in Brussels, irrespective of place of delivery. We have selected the most represented nationalities in Brussels except European countries because in our previous work we did not find any significant difference for perinatal mortality compared to Belgian mothers [2].
In Belgium, it is a legal requirement to register all live births and all fetal deaths from 22 completed weeks of gestation. Birth and death certificates are established by medical doctors and midwives for every birth and death in the territory, including asylum seekers and illegal residents. Causes of death, as recorded on death certificates, are coded following the International Classification of Diseases, 10th Revision (ICD-10).
2.2. Exposures and outcomes definitions
We used maternal nationality at mother's own birth and at delivery of her child, as stated by the parents on the birth certificate. We defined naturalisation as the adoption of Belgian nationality by the mother. We grouped nationalities into seven categories, depending on the mothers’ own births (Belgian, Moroccan, Turkish and sub-Saharan African) and naturalisation at delivery. Our results do not show the “Others” category, due to heterogeneity in nationalities. Socioeconomic level was defined on the basis of the number of incomes from declared employment in the household. We considered a birth to be preterm when the duration of gestation was below 37 completed weeks, and a low birth weight when the weight of the baby was below 2500 g. A low Apgar score is defined by an Apgar score below 7 at 5 min.
We computed the rate of perinatal mortality (fetal deaths at 22 weeks of gestation or over to death at 0–6 days of life per 1000 live births). We used Wigglesworth's classification to categorise the underlying causes of perinatal death and grouped the causes into five categories: congenital anomalies, asphyxia and unexplained deaths prior to the onset of labour, intrapartum events, conditions consequent upon immaturity, and other specific conditions [9].
2.3. Statistical analysis
We used logistic regression to estimate the odds ratios (ORs) for the association between nationality and perinatal mortality, adjusted for maternal age, parity, multiple births, and number of incomes from declared employment. We presented the odds ratios derived from the logistic regression and the p-value corresponding to the Wald χ2. The Hosmer and Lemeshow test was used to check the goodness of fit of the model. The level of significance was set at α = 0.05, and all of the analyses were performed using Stata 11 software.
return to article outline
3. Results
3.1. Characteristics of the mothers in Brussels
The mother's own birth characteristics are described in supplementary data. Of a total of 173,514 newborns, 67.2% had a mother of non-Belgian nationality. In our population study (n = 83,622), Belgian mothers were the most represented (47.7%) followed by Moroccan naturalised Belgian (18.1%) and Moroccan (14.8%) (Table 1). The proportion of babies born to families without income from declared employment was very high for mothers from sub-Saharan Africa (60.3%). In contrast, the proportion of babies born to families with two incomes was higher for Belgians (64.7%). Sub-Saharan African naturalised mothers were on average older and had the highest proportion of women above 40 years old (6.1%). Mothers from Turkey were younger with the highest proportion of mothers below 20 years old (8.8%). Mothers with three or more births before the index birth were much more represented for Moroccan naturalised mothers (21.1%). Multiple births were slightly higher for sub-Saharan African naturalised mothers (4.1%).
3.2. Birth weight, preterm birth and low Apgar score according to nationality
Table 2 shows that mothers from Morocco (naturalised and not naturalised) have significantly fewer low birth weight babies and fewer preterm babies than Belgian mothers (p < 0.0001). The same is true for mothers from Turkey, who also have significantly fewer low birth weight babies than Belgians. This difference is particularly striking for naturalised Turkish mothers (p = 0.001). In contrast, a significantly higher risk of preterm birth was observed for mothers from sub-Saharan Africa (naturalised and not naturalised) compared to Belgian-born mothers (p < 0.05).
We observed a significantly greater proportion of low Apgar scores among non-naturalised mothers from Morocco and sub-Saharan Africa compared to Belgians (p < 0.0001). Conversely, among Turkish naturalised mothers the rate is significantly lower than for Belgians.
3.3. Perinatal mortality according to nationality
Mortality rates vary widely according to nationality (Table 2). Babies of mothers who have the same nationality at their own birth and at delivery, from Morocco, sub-Saharan Africa and Turkey experience a striking excess of perinatal mortality (p < 0.0001 compared to Belgians). For the same three origins, however, when mothers have adopted Belgian nationality, perinatal mortality is similar to Belgian-born mothers.
Table 3 shows the results of the logistic regression analyses. There is no significant interaction between nationality and the number of incomes from declared employment, parity, multiple births and mother's age. Adjusting for multiple births, mother's age and parity had a small impact on the odds ratio for sub-Saharan women (not shown). The adjustment for the number of parents with an income slightly decreases the OR for babies of mothers from Morocco and Turkey. In contrast, the perinatal mortality rate for babies of mothers from sub-Saharan Africa is no longer significantly different to the Belgian rate, after adjustment. For Moroccan and sub-Saharan African naturalised mothers the crude and adjusted OR for perinatal mortality is not significantly different compared to the Belgians. For Turkish naturalised mothers the adjusted OR for perinatal mortality is lower than for Belgian mothers (p < 0.001).
3.4. Perinatal death by causes
Compared to Belgians, perinatal deaths by all causes except immaturity were significantly higher for babies of Moroccan mothers (Table 3). Mothers from Turkey experience a significant excess of perinatal mortality by congenital anomalies, asphyxia and unexplained death prior to the onset of labour and immaturity compared to Belgian mothers. Among mothers from sub-Saharan Africa, only mortality from asphyxia and unexplained death prior to onset of labour and from conditions consequent upon immaturity are significantly increased compared to Belgian mothers. Adjustment for age of the mother, parity, multiple births, and number of incomes in the household do not substantially change the findings. For Moroccan, sub-Saharan African and Turkish naturalised mothers, none of the crude and adjusted OR for perinatal causes of death was significantly different from that for Belgian mothers.
4. Comment
4.1. Perinatal mortality among immigrants
This study confirms that the association between nationality at mother's birth and birth outcomes is not uniform but depends on the migrant subgroup [1] and [10]. As observed in our previous study, the excess of perinatal mortality for sub-Saharan mothers is mainly explained by an excess of preterm birth, low birth weight and a low socio-economic level [2]. The excess of mortality by conditions consequent upon immaturity among sub-Saharan African mothers may be linked to the high rate of preterm birth. This may in turn be explained by a higher incidence of hypertension, diabetes and infection and has been widely discussed in our previous work [2]. Other studies indicate that sub-Saharan African origins are positively associated with adverse perinatal outcomes [4] and [11].
This study observed a similar pattern for Turkish and Moroccan mothers. Similar results have shown a higher rate of perinatal death by congenital anomalies among Moroccan and Turkish mothers [12], [13], and [14]. This increased rate of death from congenital anomalies may be related to two components: on the one hand, marriage between cousins or between people who come from close communities, and on the other hand, differences in utilisation of antenatal screening services. For migrant populations, Stoltenberg et al. [15] show a twice higher risk of congenital malformations among children whose parents were first cousins in Norway. For differentials in utilisation of screening services, and also presumably in choice of pregnancy termination in the case of major congenital anomaly, this has also been described in France [16]. The excess of perinatal mortality due to congenital anomalies does not persist after naturalisation, presumably due to change of behaviour.
4.2. Adequate access to and utilisation of (perinatal) health care services
A further component of the differences between perinatal outcomes among migrants may be cultural differences, such as lifestyle factors (nutrition, folic acid), and poor use of prenatal care [2]. Several studies from different countries suggest that the increased risk of other adverse perinatal outcomes is associated with late or inadequate prenatal and obstetric care due to the difficulty in accessing public health services [17].
One of the important differences among maternal risk groups is their access to preventive health care during the prenatal period. Mothers who acquire Belgian nationality may be more familiar with the language, the health care system and other relevant aspects of the host society. Indeed, cultural and linguistic barriers may limit access to health services during pregnancy and the lack of cultural mediators in prenatal clinics may lead to inequalities in pregnancy surveillance, especially for new immigrants [18]. Increasing participation in a host society is related to more health services utilisation, and language ability appears to play a central role in the uptake of health care and on the effect on pregnancy outcomes [19].
4.3. Proxy measure of acculturation (naturalised Belgian citizens)
In our study, we observe a favourable effect of naturalisation on birth outcomes and perinatal mortality for Moroccan, Turkish and sub-Saharan African mothers. Our results follow up on the study of Cacciani et al. [4] showing that migrant status is a risk factor for adverse perinatal health but the adverse outcomes decrease over time among immigrants. They suggest improvement policies are adopted, in order to increase accessibility to mother–child health services. Bollini et al. [3] have also suggested that integration policy is an important determinant of birth outcomes in migrant populations. In this study, Belgium is categorised as a country with permissive integration policies where steps are taken to acknowledge cultural differences and specific needs in order to facilitate the entry of new immigrants.
Nevertheless, birth outcomes may either improve or deteriorate with duration of residence, depending on the migrant group [1]. Troe et al. [20] have shown that among Turkish mothers in the Netherlands, infant mortality as well as perinatal and congenital causes of death increase with lower age at immigration. This trend, however, completely differs among the Surinamese migrants, suggesting a better integration in Dutch society, with adequate access and utilisation of perinatal health care.
In our study, Moroccan, Turkish and sub-Saharan African mothers may be similar to the Surinamese in the Netherlands, due to their integration in the host country. A common explanation for this integration is the acculturation theory, which is a multidimensional phenomenon that has different effects on pregnancy outcomes and nationality [19]. In our study, our proxy measure of acculturation (naturalisation) is not associated with a decrease in protective cultural behaviours among Moroccan and Turkish mothers but with a decrease in perinatal mortality. Health outcomes of immigrants tend to converge towards the level observed in the host population, presumably via changes in health-related behaviours. More particularly, acculturation is associated with the use of general practitioner care among Turkish and Moroccan migrants in the Netherlands [21]. It is also associated with significant changes in health behaviour, health and morbidity for ethnic minority groups [22].
4.4. Limitations of the study
Contrary to the mother's nationality at delivery, we observed a large number of missing data for nationality at mother's own birth (29.8%). Nevertheless, nationality percentages are consistent with data from CEPIp in Brussels in 2008 [23] and missing data are mainly Belgian mothers. There is some evidence that when the family is Belgian, the category “nationality of origin” is not always filled in. Our results may also be biased by misclassification. Autopsies are rarely conducted among migrants [24], which could lead to a possible overestimation of the number of cases of unexplained deaths that occur prior to the onset of labour, as well as an underestimation in the other causes of death categories.
Because of the >30% missing value for the education and occupation variables derived mainly from death certificates, and to avoid bias, we used the number of incomes within the household as an indicator of the socioeconomic level. This indicator has been validated in the evaluation of precarious situations [25] but it is not a complete indicator of the socioeconomic level. We also have to be careful in our interpretation of the OR value in the categories “intrapartum events” and “conditions consequent upon immaturity” because there were few subjects and the large 95% confidence interval limits our statistical interpretation. Finally, an important item of data which would be very useful in order to study acculturation is the duration of stay of migrants.
5. Conclusion
Perinatal mortality and causes of perinatal deaths vary according to nationality in Brussels but the excess of perinatal mortality for Moroccan, sub-Saharan African and Turkish mothers does not persist after adopting Belgian nationality. The demographic trajectories of immigration, acculturation and use of health services or cultural contexts are probably different according to the mothers’ nationalities.
Acknowledgement
We thank Ms. Emmanuelle Rivière for editing the manuscript.
References
[1] A.J. Gagnon, M. Zimbeck, J. Zeitlin, et al. Migration to western industrialised countries and perinatal health: a systematic review. Social Science and Medicine. 2009;69:934-946 Crossref.
[2] J. Racape, M. De Spiegelaere, S. Alexander, M. Dramaix, P. Buekens, E. Haelterman. High perinatal mortality rate among immigrants in Brussels. European Journal of Public Health. 2010;20:536-542 Crossref.
[3] P. Bollini, S. Pampallona, P. Wanner, B. Kupelnick. Pregnancy outcome of migrant women and integration policy: a systematic review of the international literature. Social Science and Medicine. 2009;68(3):452-461 Crossref.
[4] L. Cacciani, S. Asole, A. Polo, et al. Perinatal outcomes among immigrant mothers over two periods in a region of central Italy. BMC Public Health. 2011;11:294 Crossref.
[5] M. Gissler, S. Alexander, A. Macfarlane, et al. Stillbirths and infant deaths among migrants in industrialized countries. Acta Obstetricia et Gynecologica Scandinavica. 2009;88:134-148 Crossref.
[6] S. Guendelman, P. Buekens, B. Blondel, M. Kaminski, F.C. Notzon, G. Masuy-Stroobant. Birth outcomes of immigrant women in the United States, France, and Belgium. Maternity and Child Health Journal. 1999;3:177-187 Crossref.
[7] M. Lopez-Class, F.G. Castro, A.G. Ramirez. Conceptions of acculturation: a review and statement of critical issues. Social Science and Medicine. 2011;72:1555-1562 Crossref.
[8] Rapport statistique et démographique 2010: Migrations et populations issues de l’immigration en Belgique. DEMO & Centre pour l’égalité des chances et la lutte contre le racisme; 2011. p. 35.
[9] E.N. Hey, D.J. Lloyd, J.S. Wigglesworth. Classifying perinatal death: fetal and neonatal factors. British Journal of Obstetrics and Gynaecology. 1986;93:1213-1223 Crossref.
[10] M.L. Urquia, R.H. Glazier, B. Blondel, et al. International migration and adverse birth outcomes: role of ethnicity, region of origin and destination. Journal of Epidemiology and Community Health. 2010;64:243-251 Crossref.
[11] B. Essen, B.S. Hanson, P.O. Ostergren, P.G. Lindquist, S. Gudmundsson. Increased perinatal mortality among sub-Saharan immigrants in a city-population in Sweden. Acta Obstetricia et Gynecologica Scandinavica. 2000;79:737-743
[12] T.W. Schulpen, J.E. van Steenbergen, H.F. van Driel. Influences of ethnicity on perinatal and child mortality in the Netherlands. Archives of Disease in Childhood. 2001;84:222-226 Crossref.
[13] S. Anthony, H. Kateman, R. Brand, et al. Ethnic differences in congenital malformations in the Netherlands: analyses of a 5-year birth cohort. Paediatric and Perinatal Epidemiology. 2005;19:135-144 Crossref.
[14] M. David, J. Pachaly, K. Vetter. Perinatal outcome in Berlin (Germany) among immigrants from Turkey. Archives of Gynecology and Obstetrics. 2006;274:271-278 Crossref.
[15] C. Stoltenberg, P. Magnus, A. Skrondal, R.T. Lie. Consanguinity and recurrence risk of stillbirth and infant death. American Journal of Public Health. 1999;89:517-523 Crossref.
[16] B. Khoshnood, P. Pryde, B. Blondel, K.S. Lee. Socioeconomic and state-level differences in prenatal diagnosis and live birth prevalence of Down's syndrome in the United States. Revue d Epidemiologie et de Sante Publique. 2003;51:617-627
[17] B. Essen, B. Bodker, N.O. Sjoberg, et al. Are some perinatal deaths in immigrant groups linked to suboptimal perinatal care services?. BJOG. 2002;109:677-682
[18] M. Malin, M. Gissler. Maternal care and birth outcomes among ethnic minority women in Finland. BMC Public Health. 2009;9:84 Crossref.
[19] J.A. Cobas, H. Balcazar, M.B. Benin, V.M. Keith, Y. Chong. Acculturation and low-birthweight infants among Latino women: a reanalysis of HHANES data with structural equation models. American Journal of Public Health. 1996;86:394-396 Crossref.
[20] E.J. Troe, A.E. Kunst, V. Bos, I.M. Deerenberg, I.M. Joung, J.P. Mackenbach. The effect of age at immigration and generational status of the mother on infant mortality in ethnic minority populations in the Netherlands. European Journal of Public Health. 2007;17:134-138 Crossref.
[21] T. Fassaert, A.E. Hesselink, A.P. Verhoeff. Acculturation and use of health care services by Turkish and Moroccan migrants: a cross-sectional population-based study. BMC Public Health. 2009;9:332 Crossref.
[22] E. Wiking, S.E. Johansson, J. Sundquist. Ethnicity, acculturation, and self reported health. A population based study among immigrants from Poland, Turkey, and Iran in Sweden. Journal of Epidemiology and Community Health. 2004;58:574-582 Crossref.
[23] Minsart A, Willen GDSM, Van de Putte S, Verdoot C, Englert Y. Donnees perinatales en Region Bruxelloise – Annee 2008. Centre d’Epidemiologie Perinatale.
[24] S.J. Gordijn, J.J. Erwich, T.Y. Khong. The perinatal autopsy: pertinent issues in multicultural Western Europe. European Journal of Obstetrics, Gynecology, and Reproductive Biology. 2007;132:3-7 Crossref.
[25] T. Roesems, G. Feyaerts. Brometre social – Rapport bruxellois sur l’etat de la pauvrete. Obseravtoire de la sante et du social de Bruxelles-capitale. 2009;:60-63 Chapitre 6 Commision communautaire commune
Footnotes
a CR Epidemiology, Biostatistic and Clinical Research, School of Public Health, Free University of Brussels, Brussels, Belgium
b Health and Social Observatory of Brussels, Brussels, Belgium
Corresponding author at: CR Epidemiology, Biostatistic and Clinical Research, School of Public Health, Free University of Brussels, CP598, Route de Lennik 808, 1070 Brussels, Belgium. Tel.: +32 2 555 40 47; fax: +32 2 555 40 49.