Introduction

Population Mobility and the Health of Migrants

Human migration is not a new phenomenon, but it has changed significantly in volume and nature with the advance of globalization, including the growth of international transport and communication, and the interlinked destinies of nations following the recent economic recession [1]. In the US, the size of the immigrant population in 2012 was 40.8 million, corresponding to a 13 % share of the total population [2]. This share has been increasing in recent years. In Europe, Eurostat data indicate that the contribution of net migration has exceeded natural population growth since 1992, peaking in 2003 [3]. Since then, migration flows seem to have decreased. In total, compared with the year just before the economic crash (2007) immigration to EU Member States is estimated to have decreased by 6 % and emigration to have increased by 13 % [4]. Currently, Europe is facing new challenges with thousands of war refugees seeking asylum. As a result, population mobility is among the leading policy issues of the twenty-first century.

A topic of surpassing interest to both sending and receiving countries is the health status of migrants [4]. A contentious political issue is whether migrants impose an economic cost to the receiving countries through increased medical, educational, and social safety net expenditures. This topic is debated everywhere, whether it be in Hong Kong where the Chief Executive recently imposed a ban on mainland Chinese immigrants moving there in order to deliver babies (to qualify for citizenship), or in California counties where citizen referenda have attempted to deny health care for undocumented farm laborers. Nonetheless, information on migrant health in many countries remains scarce. Information about the health of migrants in Europe differs by region, which makes it difficult to monitor and improve migrant health [4].

We sought to review the published literature about migration and health contrasting the United States with the Europe region, since both have been regions with a traditional history of migration yet have differing results in terms of health outcomes (e.g. healthy migrant effect widely observed in the US compared to Europe). We aim to observe which features can influence health outcomes among immigrants. We suggest that the health outcomes of the current asylum seekers in Europe will depend on the characteristics of host countries. Our hypothesis is that in both regions the primary drivers that affect the health of migrants will depend not only on the migrants’ profiles, but also on the migration regimes within receiving countries. The migration regime—understood as the system of laws, regulations, policies and institutions within each country—will have a profound impact on the lives of migrants. To illustrate the contrasting literatures on health and migration we focus on one particular health outcome for which our systematic review uncovered a substantial number of studies, viz. reproductive health outcomes. We specifically focus on low birthweight and small for gestational age (SGA). Using the example of reproductive health outcomes we aim to discuss differences between countries regarding health outcomes of immigrants and to elucidate why these differences arise. We also discuss some of the data challenges, the impact of the migration regime and the social environment of migrants in the receiving countries, and what we need to know in order to improve the evidence base for policy making in these challenging times.

The focus of our review is not directed toward a gendered analysis of migration, e.g. how differences in the profile of migrants can be highly patterned by gender, such as the flow of (virtually exclusively) female domestic workers from Southeast Asia to Hong Kong, or the migration of almost exclusively male manual labor from East Asia to the United Arab Emirates. A gendered analysis of migration and health necessitates a consideration of the economic structures, gender relations (rights, laws, structures of power), and gender-based cultural norms prevailing in both the sending countries as well as receiving countries. This is beyond the scope of the present review. Instead our focus on reproductive outcomes of migrant women was primarily driven by the consideration that: (a) there is a large literature on the subject, and (b) reproductive outcomes are a sensitive “mirror” of social conditions confronted by migrants in their host countries.

Methods

We conducted a literature search in Pubmed and Embase. Our search follows the PRISMA guidelines and we used the two databases to retrieve only articles in the published scientific literature. The search did not include low and middle-income countries since our focus was on studies from the US and the EU. The inclusion criteria to consider the articles in our analysis were: abstract available and information about low birth weight (LBW) or slow for gestational age (SGA) in migrants. From the total of articles we obtained, we selected 63 from the United States and 51 from Europe. In a second round, we read all the articles selected and excluded those according to the following criteria: articles written in a different language than English, do not include LBW or SGA as an outcome, intervention studies aimed at decreasing LBW or SGA, economic evaluations, do not provide data to compare migrants and native-born. After reading all the articles selected we finally included 38 articles from the United States and 30 articles from Europe. Information about the literature search strategy, the inclusion and exclusion criteria and the manuscripts included in the study are given on Table 1 and Fig. 1.

Table 1 Literature search strategy in Pubmed and Embase
Full size table
Fig. 1
figure 1

Flow chart of the review process. Inclusion criteria first round: abstract available and information about low birth weight (LBW) or small for gestational age (SGA) in migrants. Exclusion criteria second round: articles written in a different language than English, do not include LBW or SGA as an outcome, intervention studies aimed at decreasing LBW or SGA, economic evaluations, do not provide data to compare migrants and native-born

Full size image

Results

The Healthy Migrant Effect in the US and Health Inequalities in Europe

The US literature on migration and health has been heavily dominated by discussions of the healthy migrant effect and the so-called “Latino health paradox”. Studies have been devoted to dissecting the reasons behind the apparent paradox that immigrants—particularly Mexican immigrants—have been found to have better health outcomes (e.g. lower rates of LBW) compared to the native-born, in spite of their lower socio-economic backgrounds. The “Latino health paradox” in the United States has been in turn ascribed to three different strands of explanation: (a) immigrant self-selection (i.e. those who are fit and healthy tend to migrate for work); (b) the “salmon bias” which posits that immigrants return home after they become sick (to be cared for by their relatives), and (c) the “ethnic enclave” hypothesis, which conjectures that immigrants are protected as a result of settling into residential areas with high immigration concentration, thereby providing them social integration (a kind of bonding social capital) at the same time as insulating them from the deleterious exposure to discrimination from society at large [5]. More recent literature has, however, begun to question the generalizability of the Latino health paradox, pointing out that when we examine the fine-grained detail of immigration from different sending countries, not all migrant groups from Latin America experience better health outcomes compared to the native born [6, 7]. Clearly, the “paradox” is contingent on many factors, including the specific country of origin of migrants, as well as the specific health outcome under consideration.

Pregnancy Outcomes

Table 2 is a summary of the results of articles published with data from the United States. The healthy migrant effect in pregnancy outcomes is reported in 21 articles, observed mostly in Latinas. However, the articles from Collins and Shay [40] and from Guendelman and English [9] show that second generation Latinas or those with a longer duration of residence in the US have worse pregnancy outcomes, suggesting that the healthy migrant effect tends to wane over time [8, 9]. In addition, the works from Kaufman et al. [10] or Rosenberg et al. [11] suggest that Puerto Ricans have worse pregnancy outcomes [10, 11]. In the case of immigrant black populations, two articles show worse pregnancy outcomes [12, 13]. Among Asian people, six articles show that pregnancy outcomes are not better compared to US borns [14–19]. Importantly, Kelaher and Jessop [20] and Reed et al. [18] show that there are no important differences between documented and undocumented migrants [18, 20].

Table 2 Studies with US data included in the review (N = 38)
Full size table

Table 3 is a summary of the results with articles published with data from Europe. The results of these articles differ compared to the results of the analysis of American data. First, two countries report results that support a healthy migrant effect in pregnancy outcomes: Spain and Belgium. In Spain, the results reported by Speciale and Regidor [21] suggest that the LBW outcomes differ considering the groups of migrants, and that some groups of migrants have better LBW results compared to native-born [21]. Garcia-Subirats et al. [22] found that Spanish mothers have higher risk of moderate LBW, while migrants have higher risk of very LBW [22]. The other reports published with Spanish data did not replicate these observations. In the case of Belgium, three reports observed a healthy migrant effect, while Racape et al. [23] conclude that this effect depends on the origin of migrants [23]. Jacquemyn et al. [24] report there is no healthy migrant effect in Belgium when the native-born are compared with Moroccan and Turkish immigrants [24]. Small et al. [26] found a healthy migrant effect in Somalis compared to the native-born in Belgium, Canada, Finland, Norway and Sweden, and Lalchandani et al. [25] conclude that there are no differences between Irish native-born and refugees in terms of LBW [25, 26]. In conclusion, European data only report a healthy migrant effect in terms of pregnancy outcomes in two countries, and Somalis show a healthy migrant effect compared to native-borns from several European countries. There are no differences between refugees and the native-born in Ireland. Surprisingly, there is a lack of studies where the immigrant categories are specified. Hence, we conclude that little is known about what to expect in terms of pregnancy outcomes from asylum seekers.

Table 3 Studies with European data included in the review (N = 30)
Full size table

Discussion

The results of our review highlight the differences in the reproductive health outcomes of migrants comparing the USA to the European region. Differences in pregnancy outcomes not only derive from the characteristics of the migrant population, but also stem from differences in how immigrants are defined in each society, as well as the migrant regime of each region. These results may be extrapolated to other health outcomes, and the data challenges apply to all the studies related to migration.

Who is a Migrant? Problems with Definition

During the process of looking for differences between regions we have found that part of the studies do not take into account the reasons for migration, which makes the comparisons between countries and health outcomes even more difficult [17, 19, 27]. In addition, to make sense of cross-national comparisons of migration and health we need to take into consideration the fact that each country defines migrants differently. Each country also has a unique history of migration flows. It is influenced by factors such as labor migration, historical links between countries of origin and destination, and established networks in destination countries [4]. The collection, interpretation and comparability of data about migrants and their health status is difficult. In the case of the US, for example, the Behavioral Risk Factor Surveillance System surveys of the Center for Disease and Control prevention (CDC) do include immigrants. However, given that the sampling frame of the surveys is through telephone surveys, the CDC loses populations that do not have access to landlines. Similarly in Europe, data are incomplete to meet the needs of public health policy or health-care provision [28]. Bhopal [28] concludes that existing data do not usually provide a national perspective as they are mostly from local studies [28]. In addition, only the first and second generation and another country of birth define migration status. Nevertheless, the EU immigration portal offers different definitions related to migration [29]. First, a migrant is considered ‘a broader-term of an immigrant and emigrant that refers to a person who leaves from one country or region to settle in another, often in search of a better life’. The definition of immigration is the following: ‘In EU context, the action by which a person from a non-EU country establishes his or her usual residence in the territory of an EU country for a period that is, or is expected to be, at least 12 months’. Going further, countries within Europe have different definitions of who is a migrant. In Germany, people who immigrated after 1950 and their descendants are described as people with immigrant background. The same also happens in Israel. Both countries adopted this definition after the Second Word War. In contrast to this approach, in the UK migrants are broadly defined as ‘foreign born’ [30]. Data collection is still guided by national legislative, administrative and policy needs, and follows national definitions and classifications, just as the determination of citizenship, residency and immigration in the EU remains to a large extent a national responsibility [31].

Migration Regimes

The composition of migrants varies across time and place, according to the migration regime that happens to be in place. In the case of Europe, several different categories of migrants can be distinguished: asylum-seekers and refugees, victims of trafficking, students, migrant workers, and reunified family members [4]. In the studies we include in our literature search about pregnancy outcomes, little is known about the different categories of migrants included in the investigations. However, this information is crucial to compare different migrant groups that would probably have quite different pregnancy outcomes. It poses a problem because these groups have specific health needs and may face particular legal or other barriers in accessing health services [32]. The results of Lalchandani et al. [25] do not find differences in pregnancy outcomes among native Irish and refugees [25]. This unexpected result makes the study of this population even more important now that Europe is facing a big challenge with thousands of refugees seeking asylum. In some countries, migrants face major barriers in accessing health services, whereas others are more integrative and less restrictive [33]. Furthermore, even within distinct categories of migrants, there is bound to be great variation in the problems faced [31]. The health of migrants also depends to a large degree on the specifics of the host country. We have observed that some countries in Europe do not accept asylum seekers and have denied the reception of immigrants. Others have accepted the reception of asylum seekers but with some reservations.

In the case of the United States, an important distinction (the issue that dominates public discourse) is between documented and undocumented migrants. The articles retrieved in our literature review with United States data rarely specify the legal status of migrants, which makes the comparability of groups difficult again. However, the pregnancy outcomes of documented and undocumented migrants can be very different. Unauthorized immigrant workers have been an important source of low-skilled labor supply to the United States economy for many decades. The persecution of unauthorized immigrants, but not employers of undocumented migrants, is the expression of the complex sociopolitical migration regime of the US.

In the case of Europe, the citizenship structure varies greatly between Member States. From the distinguished categories of migrants, the most important distinction is between regular or irregular migrants. As Rechel et al. [4] pointed out, the situation is further complicated through short-term, circular, and return migration [4]. In addition, the variety of policies and the diversity of socioeconomic and living conditions of the European host countries make the situation of migrants even more complex. Considering the possible differences between countries, those that have to face the most difficult situations are undocumented migrants and asylum-seekers.

Policies of the different host countries may have different impacts on the health of migrants. The political attention paid to the health of migrants is also related to prevailing attitudes towards migrants and immigration in the hosting countries. Some countries in Europe have based their policies on restriction and control. Asylum-seekers and refugees frequently face a hostile reception in their receiving countries, as we have recently observed in Europe with refugees from Syria and Iraq looking for asylum. These attitudes towards immigration constitute the migration regime, are specific of the host country and will determine the citizenship rights of migrants as well. Overall, the health of migrants (e.g. their pregnancy outcomes) will depend on it.

Conclusions

We observe opposite patterns regarding reproductive outcomes among immigrant populations compared to native-born in the United States versus the Europe region [9, 23]. One explanation for the US/Europe difference may be ascribed to the composition of migrants in the two regions. However, only a few of the studies retrieved talk about the immigrant categories. Migrants to the US have been predominantly labor migrants and therefore strongly selected for health. However, this profile changed significantly when the Immigration and Naturalization Services allowed families of migrants to join their working relatives after the approval of the 1065 Immigration Act. Consequently, older relatives seeking to be reunited with their children could enter the country on a legal basis. This change in the profile of immigrants is reflected by the changing health status of immigrants entering the United States.

By contrast, immigrants to Europe have been suggested to be much more heterogeneous, as we have pointed out, and some groups (e.g. refugees) are less selected with respect to health status compared to labor migrants. However, we are unable to draw any definitive conclusions since only a few studies distinguish between migrant types (e.g. refugees, undocumented migrants). There is a need for studying the health outcomes of this community.

In addition, the definition of “immigrant” varies by country, which adds further complexity. For example, in the US, second-generation migrants are not classified as “immigrants”. They are, by definition, citizens. However, in some studies they are still considered part of a group ethnically different. US studies show that the Latino advantage in birth outcomes disappears in second generation migrants. This trend towards convergence in health outcomes can be due to a possible acculturation effect, as the trends in obesity suggest [34].

Our literature search has pointed out the data challenges the European region has to face with regard to data collection and comparability of this data. Emerging reports of a healthy migrant effect in Europe need further investigation. But it is necessary to include the time migrants have been living in the new host country, the reasons for migration and continue with mortality studies, to investigate if the mortality rates tend to convergence over time, as has been suggested [4]. Nonetheless, these results will be subject to different categories of migrants and the migration regime of each country. The new political winds have put up barriers to make this access even more difficult than before. The current asylum seekers in Europe will face different challenges depending on the countries they are going to live in. Health inequalities are expected to be even bigger in those countries that reduce entitlements for undocumented migrants.

Social epidemiologists point out that the social environment of the new hosting country will have an effect on the health of migrants. To change the rules of the migration regime will change the profile of migrants, and their health related outcomes. In short, a specific understanding of the “migration regime” is required in order to properly understand the complex and evolving nature of the relationship between migration and health. Nevertheless, we expect that the association between migrant status and health will differ according to the background forces that shape migration patterns. As migration trends indicate, there are important period differences regarding who migrates and why, and the results concerning migrants and health will completely depend on it [3]. Hence, we expect that refugees from Syria and Iraq seeking from asylum in Europe will have different health outcomes depending on the countries they are going to live in.

Limitations of this Study

An important issue is that some articles with United States data do not talk about migrants or foreign-born since they investigate health related outcomes of Latinos and only have included the concept Latino as a key word. In this case, they have not been included in our literature search. We discarded the ISI Web of Knowledge for the literature search, since we only wanted to include manuscripts related to health.