Among migrants in high-income countries, maternal mortality and severe morbidity generally occur more frequently as compared to host populations. There is marked variation between groups of migrants and host countries, with much elevated risks in some groups and no elevated risk at all in others. Those without a legal resident permit are most vulnerable.
A reason for these elevated risks could be a different risk profile in migrants, but risk factors are unevenly distributed and not always present. Another reason is substandard care, which is identified more frequently in migrants, and comprises patient delays, for example, due to a lack of knowledge about the health system in the host country, and health worker delays, often compounded by communication barriers.
Improvements in family planning and antenatal services are needed, and audits and confidential enquiries should be extended to include maternal morbidity and ethnic background. This requires scientific and political efforts.
Highlights
- •
Maternal mortality and severe morbidity are generally more common in migrant women.
- •
Between migrant groups, risk increases vary from very high to none at all.
- •
Risk factors are unevenly distributed between migrant groups and not always present.
- •
Substandard care is more common, due to patient as well as health worker delays.
- •
Scientific and political efforts to improve quality and access are much needed.
Introduction
In our world, where peace and prosperity remain remote prospects for many, migration is – and will remain – one way to seek a better life. As a consequence, many health workers will encounter women with customs and cultures very different from their own. If they are to take up their task responsibly and act according to universal professional ethics, health workers attending to foreign-born women should be aware of the possible barriers to adequate care provision and try to overcome these.
This principle assumes foremost priority in the field of obstetrics, where cultural, linguistic or socio-economic barriers may lead to life-threatening delays in the provision of adequate health care. Maternal mortality is the most devastating consequence of such delay and has traditionally been used to monitor pregnancy outcomes around the globe. In high-income settings or small geographic areas, where maternal deaths are rare events, severe maternal morbidity has in recent times gained ground as an additional, and often a more useful, indicator.
Migration is a concept with a variety of forms and meanings. In this review, the type of migration addressed is that from low-income settings of conflict and economic hardship to high-income – particularly European, North American and Oceanian – settings. We have tried as much as possible to be specific about countries of origin and destination, since a failure to do so may carry a risk of stigmatization. This review is aimed particularly at obstetric care providers and policy-makers in high-income settings. Our objective is to equip such professionals with some tools to improve their services for migrants.
Paucity of data
There is a remarkable paucity of literature on maternal mortality and morbidity in migrant women. This illustrates that these women are generally not the primary study population of interest for researchers and policy-makers in high-income countries where most studies are conducted.
A PubMed search combining synonyms of ‘maternal morbidity and mortality’, ‘pregnancy complications’ and ‘migrants’, while excluding ‘fetal diseases’, rendered 548 articles on 14 June 2015 ( Fig. 1 ). However, based on titles and abstracts, only 51 of these articles primarily or secondarily addressed obstetric complications in migrant women. Many articles focused on neonatal rather than maternal outcomes, on post-partum depression and psychiatric conditions rather than somatic illness, and few papers specifically addressed the situation of foreign women in high-income settings. Snowballing revealed another three studies that were previously not identified, as well as renowned accounts of maternal mortality, such as the Confidential Enquiries into Maternal Deaths in the United Kingdom, which are not listed in Medline .
It is clear that, given the diversity of migrant groups and host settings as well as the complexities of maternal mortality and morbidity studies in general, the amount of documented evidence is not sufficient to give a comprehensive account of all migrant maternal health in high-income settings or do justice to this topic.
Paucity of data
There is a remarkable paucity of literature on maternal mortality and morbidity in migrant women. This illustrates that these women are generally not the primary study population of interest for researchers and policy-makers in high-income countries where most studies are conducted.
A PubMed search combining synonyms of ‘maternal morbidity and mortality’, ‘pregnancy complications’ and ‘migrants’, while excluding ‘fetal diseases’, rendered 548 articles on 14 June 2015 ( Fig. 1 ). However, based on titles and abstracts, only 51 of these articles primarily or secondarily addressed obstetric complications in migrant women. Many articles focused on neonatal rather than maternal outcomes, on post-partum depression and psychiatric conditions rather than somatic illness, and few papers specifically addressed the situation of foreign women in high-income settings. Snowballing revealed another three studies that were previously not identified, as well as renowned accounts of maternal mortality, such as the Confidential Enquiries into Maternal Deaths in the United Kingdom, which are not listed in Medline .
It is clear that, given the diversity of migrant groups and host settings as well as the complexities of maternal mortality and morbidity studies in general, the amount of documented evidence is not sufficient to give a comprehensive account of all migrant maternal health in high-income settings or do justice to this topic.
Definitions
According to the most recent definition developed by the World Health Organization (WHO), maternal mortality is defined as the death of a woman while pregnant or within 42 days of termination of pregnancy, irrespective of the duration and site of the pregnancy, from any cause related to or aggravated by the pregnancy or its management but not from accidental or incidental causes .
The definition of severe maternal morbidity is more problematic, since at present no universally applied definition or case criteria exist. In many studies, the terminology ‘severe acute maternal morbidity (SAMM)’ has been applied, but with varying case definitions. In recent years, WHO has set out to define ‘maternal near miss’ as a way to identify severe maternal outcomes in women who nearly died but survived a complication that occurred during pregnancy . However, ‘maternal near miss’ and the associated criteria have been shown to have some downfalls and will require additional adjustments and validation .
Therefore, in this review, we have included all articles that addressed any form of SAMM in a broader sense, in particular the most common diagnoses comprising SAMM. These are: (A) eclampsia and severe pre-eclampsia, the latter with different criteria used to define severity, such as an indication for induction of labour or caesarean section; (B) major obstetric haemorrhage, with different case definitions applied in different studies, based on the amount of the estimated blood loss, the number of units of blood transfused or surgical interventions; (C) sepsis and severe infection, including infections after (unsafe) abortion; and (D) obstructed labour, and particularly the presence of uterine rupture.
Maternal mortality among migrant women
A variety of studies conducted in high-income countries have consistently shown a higher risk of maternal mortality among migrant women ( Table 1 ). A large study pooling the results of 13 studies from six countries conducted between 1970 and 2013 including more than 42 million women and 4995 maternal deaths found that, compared with indigenous born women, the pooled relative risk in the case of migrant women in Western Europe was 2.00 [95% CI: 1.72–2.33] .
| Country/year | Maternities (M) or live births (LB) | Maternal deaths | Relative risk (RR) or Odds Ratio (OR) |
|---|---|---|---|
| Netherlands/1996–2005 | 2,269,506 LB | 302 | OR 2.1 (1.6–2.7) |
| United Kingdom and Ireland/2009–2012 | 3.182.890 M | 321 | RR 1.77 (1.39–2,24) |
| Spain/1999–2006 | 3.648.788 LB | 133 | RR 1.67 (1.22–2.33) |
| Switzerland/2000–2006 | 1,021,177 LB | 57 | OR 4.38 (1.88–10.55) |
| France/1996–2001 | 267 | OR 2.00 (1.42–2.80) | |
| West Germany/1980–1996 | 11,120.000 M | 1067 | RR 1.59 (1.37–1.84) |
| Italy/2000–2007 | 1.001.292 LB | 118 | RR 1.2 (0.7–2.1) |
| Sweden/1997–2007 | NA | 73 | RR 6.6 (2.6–16.5) |
| United Kingdom, Netherlands, Germany, France, Spain, Switzerland (1969–2008) | 42,290,456 | 4,995 | RR 2.00 (1.72–2.33) |
A closer look into the literature reveals that this increased risk is not equally divided among different migrant groups. In France, women from sub-Saharan Africa were at the highest risk of mortality (aOR 5.45 [3.29–9.00]). In this group, excess mortality was particularly high for those with sepsis (aOR 9.39 [2.47–35.70]) and hypertensive disorders (aOR 7.62 [3.24–17.39]) . Slightly older data from the United Kingdom, extracted from death registrations between 1970 and 1985, showed that women from the African and Caribbean Commonwealth countries had the highest relative risks (5.0 [3.8–6.5] and 4.8 [3.8–6.0], respectively, compared with women from England and Wales) .
In addition to the discrepancies between migrant groups, maternal mortality may also not be equally distributed between different regions within the host country. In Spain, excess mortality was particularly marked in two autonomous regions, Andalusia and Asturias . In France, it was found that maternal mortality was higher in non-nationals, and particularly higher in Paris, most likely due to a higher rate of suboptimal care .
Moreover, the perception that the mortality gap between natives and non-natives may be reduced over time as health systems develop was proven wrong in Switzerland, where between 1969 and 2006, the maternal mortality ratio among migrants showed a marked increase .
It must be noted that in all these studies, the group of migrants is rather heterogeneous, comprising people from very different socio-economic and geographical backgrounds. It is likely that those migrants who have recently migrated and do not have a residence permit in the receiving country are the most vulnerable group and therefore at a considerably elevated risk compared to other migrants. This hypothesis was shown to be true in the Netherlands, where maternal mortality among asylum seekers was found to be extremely high (RR 10.08 [8.02–12.83]) .
Besides, it must be said that the increased death risk among migrants is unlikely to be limited to diseases of pregnancy. In Sweden, based on the death registers of 27,957 women aged 15–49 years who died between 1988 and 2007, the total age-standardized mortality rate per 100,000 person years was significantly higher for women born in low-income countries (84.4), as compared with Swedish-born women (68.1) .
Unfortunately, the analysis of maternal mortality to date relies on a limited number of studies, mostly from European countries. Recently, however, some significant publications in the United States have addressed maternal mortality in that country including ethnic backgrounds of women who died ( Table 2 ). These figures indicate that foreign-born Hispanics and Asians are at elevated risks as compared to US-born Hispanics and Asians, but that this is not true for black women, who are at the highest risk, regardless of being born in or outside the country. Table 3 shows the maternal mortality ratios of different groups . Although not specifically addressing migrants, a US study found that abortion-related deaths were higher in black women (1.1 per 100,000 abortions), compared to Hispanic (0.5 per 100,000 abortions) and non-Hispanic white women (0.4 per 100,000 abortions) .
| Relative risks Reference: US-born women 1993–2006 | Relative risks Reference: US-born white women |
|---|---|
| Foreign-born white women: 0.83 (0.69–0.98) | Foreign-born white 0.83 (0.69–0.98) |
| Foreign-born Hispanic: 1.21 (1.07–1.36) | Foreign-born Hispanic 1.28 (1.18–1.38) |
| Foreign-born black: 0.92 (0.82–1.03) | US-born black Hispanic 1.06 (0.95–1.18) |
| Foreign-born Asian or Pacific: 1.37 (0.98–1.92) | Foreign-born black 3.55 (3.18–3.98) |
| US-born black women 3.87 (3.67–4.08) | |
| Foreign-born Asian or Pacific 1.31 (1.16–1.49) | |
| US-born Asian or Pacific 0.96 (0.69–1.31) |
| US-born white | 10.3 |
| Foreign-born white | 8.4 |
| US-born Hispanic | 10.3 |
| Foreign-born Hispanic | 12.3 |
| US-born black | 39.9 |
| Foreign-born black | 34.1 |
| US-born Asian or Pacific | 10.9 |
| Foreign-born Asian and Pacific | 11.7 |
Severe maternal morbidity among migrant women
With maternal mortality becoming a rare event in high-income countries, even among migrant women, maternal morbidity has increasingly been used as an additional indicator to monitor pregnancy outcomes and quality of care. Since maternal mortality may be regarded as the ultimate endpoint of severe maternal morbidity, it is hardly surprising that a similar pattern of higher risk is observed for maternal morbidity among migrants ( Table 4 ).
| Country/year | Maternities | Number of SAMM/near-miss | RR or OR |
|---|---|---|---|
| Netherlands, 2004–2006 | 358,874 | 2506 | RR 1.3 (1.2–1.5) |
| United Kingdom 2005–2006 | 775,186 | 686 | RR 1.58 (1.33–1.87) |
| Australia (Victoria) 2001–2010 | 636,042 | 1316 | OR 2.0 (1.45–2.75) |
| Canada (Ontario) 2001–2010 | 1.050,688 | 3062 | OR 1.5 (1.21–1.85) |
| Denmark 2001–2010 | 636,177 | 3085 | OR 1.8 (1.4–2.21) |
| Sweden 1998–2007 | 914,474 | 2655 | OR 2.3 (1.9–2.8) |
A large study that pooled the results of 2,322,907 deliveries in Victoria, Australia, Ontario, Canada and the whole of Denmark, of which 479,986 (21%) included migrant women, found that in all the three receiving countries, sub-Saharan African women in particular were consistently at higher risk of severe maternal morbidity (pooled aOR: 1.67; 95% CI: 1.43, 1.95) .
In a Swedish study that used nationwide data collected between 1998 and 2007, about 914,474 deliveries found 2.655 near-miss events (2.9 per 1000 deliveries) using the WHO organ failure and management criteria for maternal near miss, as well as ICD-10 codes. In comparison to Swedish-born women, women from low-income countries were at an increased risk of such near-miss events (OR 2.3, [1.9–2.8]), which was significant in all morbidity groups except for cardiovascular diseases and sepsis. Among the selected groups, the near-miss frequency was very high, particularly in Somalians ( n = 66, 9.1/1000), Eritreans ( n = 12, 7.2/1000) and Ethiopians ( n = 21, 6.6/1000) .
It is of utmost importance to differentiate between different groups of migrants within specific settings. In the Netherlands, France and the United Kingdom, several groups of migrant women were not at an elevated risk compared to native women, whereas others had markedly increased risks ( Table 5 ).
