Counting the non-existing: causes of death of undocumented migrants in Brussels-Capital Region (Belgium), 2005–2010

Since 2005, research on health and access to care of undocumented migrants has been steadily increasing [34]. The main focus of this literature has been on the legal restrictions to healthcare access [4, 18, 29], on lower healthcare utilization [31] and on barriers that prevent the effective access to healthcare by undocumented migrants in Europe, Canada and the US [5, 10, 16]. These studies resulted in recommendations for policy changes to increase access to healthcare for this population and in proposing strategies to promote a better integration into mainstream healthcare institutions [29].

Yet, so far there is limited knowledge on the impact of an undocumented status on general health. There are no epidemiological data establishing the association between health policies and physical health outcomes of undocumented migrants [18]. Studies on chronic morbidity and especially mortality studies of undocumented migrants are rare. Most study designs are qualitative and much of the available research are reports, conducted by NGO’s. Because of their undocumented status, they are often not or not adequately registered in medical files, nor included in national databases [34] and they do not appear in international morbidity or mortality comparisons [13].

Reports such as those of the United Nations Monitoring Committee on Economic, Social and Cultural Rights describe the ill-health of undocumented migrants without referring to formal data (UN, 2000), information that is often reproduced in other publications [28]. Rousseau et al. [27] describe case reports of deaths that are considered to have been caused by problems in access to healthcare due to migration status concerns, as ‘wrongful deaths’. In addition, a lot of research in this area focuses on the illnesses of specific subgroups such as border crossers [33] or on specific pathologies and health issues. Especially, the effect of the migration status on mental health [16, 18, 30] and particularly vulnerable groups such as children and pregnant women have been studied. These studies explore prematurity, low birth weight and neonatal mortality [3, 9], work related accidents [16] or treatment of infectious diseases, such as hepatitis B/C, HIV and tuberculosis [8, 34]. Yet, focusing on specific pathologies does not provide us with insight into the impact of limited access to healthcare on overall health. Another approach is to measure the perceived health [5, 13] of undocumented migrants and use it as an indicator of health. These studies show that undocumented migrants more often report their health as poor or very poor compared to residents. Therefore, based on studies of Kaplan (1996) [12] and DeSalvo [6] that suggest a good general correlation between self-rated health and objective (and/or medical) indicators of health, they assume that morbidity and mortality among undocumented migrants is higher than among residents.

To sum up, information on morbidity and especially mortality of undocumented migrants, continues to be a major gap in knowledge about the health of undocumented migrants. This is puzzling, since one could either argue that their health and mortality is worse or better than that of legal residents, depending on which literature one draws upon. On the one hand, the hardships related with anti-immigration policies and the limited access to health services are perceived to result in worse health and higher mortality, especially from avoidable causes. On the other hand, according to the healthy migrant hypothesis, the all-cause and chronic-disease mortality of first-generation migrants is generally known to be lower than that of the host population [21, 25]. A good general health might be an alternative explanation for low utilisation rates of healthcare services. In order to inform policymakers adequately and to build up an inclusive healthcare system, the use of accurate health data is crucial.

Recently, Wahlberg et al. [32] were the first to determine national figures on causes of death of undocumented migrants, based on data of the Swedish causes of death register (CDR). They established that the mean age at death was much lower for undocumented migrants in Sweden compared to Swedish residents; and that they had an increased risk of dying from external causes and circulatory system diseases, and a lower risk of dying from neoplasms.

Belgium is well suited to study mortality by migrant background, as it has one of the largest proportions of immigrants in Europe [7]. The proportion of both documented and undocumented migrants in Belgium is largest in Brussels-Capital region [11, 23]. Belgium has a publicly financed healthcare system that covers all residents and documented migrants, whereas undocumented migrants can only access the public healthcare system after going through a complex parallel administrative procedure. Several federal and non-governmental reports state that there are many barriers for undocumented migrants to access healthcare in Belgium ([24]; Chauvin, 2009 [5]; Roberfroid et al. 2015 [26]), including the complexity of the administrative procedure, financial barriers and fear of arrest (Chauvin, 2009 [5]; Roberfroid et al. 2015 [26]).

The primary objective of this study is to establish whether the mortality pattern of undocumented migrants in the Brussels-Capital Region, when controlled for age and sex, differs from that of regular residents on the one hand and documented migrants on the other. Our study tests the findings of Wahlberg et al. [32]. Yet, our study also provides a more detailed analysis by performing age-adjustments (which were lacking in the work by Wahlberg) and by analysing a wider range of underlying causes of death. Furthermore, we compare patterns of causes of death in undocumented migrants not only with Belgian residents, but also with documented migrants.

About two thirds in the undocumented group were males, which is consistent with the composition of the sample in an earlier multi-country survey about the subjective health amongst undocumented migrants in Belgium [5]. Consistent with findings in the general population [1], our data reveal important sex differences in cause-specific mortality among undocumented migrants. The top 3 countries of origin in the undocumented group in our sample were Morocco, Algeria and Congo (Tables 1 and 2). Amongst the so-called third-countries, these nationalities are the most important countries of origin of undocumented migrants in Belgium [20].

For undocumented males, we found a significantly higher risk of dying from cardiovascular diseases compared with both Belgian residents and documented migrants. Deaths from heart diseases are considered to be avoidable, as they are both amenable and preventable. The increased mortality risk from cardiovascular diseases might be related to a reduced access to primary prevention or a reduced access to treatment for male undocumented migrants (Commission on Social Determinants of Health, 2008 [15];). Alternatively, the differences in cardiovascular mortality between male undocumented migrants and Belgian residents/documented migrants might partially be related to socio-economic factors, such as a difference in level of education between undocumented men and legal residents [14]. Unfortunately, our data did not provide sufficient data on socio-economic status to probe into this hypothesis; nor information on differences in lifestyle factors (such as smoking habits) that are likely to impact cardiovascular mortality.

When comparing mortality in male undocumented migrants to that of male Belgian residents we also found a decreased risk of death from neoplasms and suicide. However, this difference was not confirmed when comparing to documented migrants. Wahlberg et al. [32] also found a decreased risk of death from neoplasms for undocumented migrants compared to Swedish residents. Since mortality from neoplasms is comparable between documented and undocumented males, the decreased risk of death from neoplasms of undocumented migrants versus Belgian residents might point at a specific mortality profile of migrants -the healthy migrant effect- rather than to the undocumented status [21, 25].

Our findings also show a lower risk of suicidal death in both female and male undocumented migrants compared to Belgian residents. When comparing the risk of suicidal deaths with that of documented migrants, there were no statistically significant differences. These findings are remarkable, given the important body of literature suggesting an association between undocumented status and poor mental health outcomes such as depression, post-traumatic stress disorder and anxiety [16, 18, 30]. This literature is mainly based on qualitative research. Our findings rather seem to support the healthy migrant hypothesis, and more specifically ‘the migrant hope hypothesis’, when it comes to suicide risks of migrants. The latter hypothesis states that the hope that comes with migrating to a more ‘developed’ country engenders resilience towards otherwise intolerable conditions [2]. However, a lower suicide mortality risk doesn’t necessarily mean better mental health among undocumented migrants. The lower suicide mortality risk might also wear off with increasing duration of stay in Belgium. In any case, our findings point to the need to combine findings from quantitative and qualitative research to understand the mental health status of undocumented migrants better.

More than one quarter of the male undocumented migrants died from an external cause of death. A possible explanation for the increased risk of death from external causes and accidental falls among undocumented men versus documented migrants, is that working conditions of male undocumented migrants are worse than those of documented migrants who are likely to have a higher socioeconomic status. Undocumented migrant men in Brussels play an important role in the informal economy, mainly in construction. Safety issues in this sector have repeatedly been reported [22]. Undocumented women mainly work in childcare and domestic work, which could explain the gender difference [36]. Contrary to Wahlberg et al. [32], who found an increased risk of death from external causes among both male and female undocumented migrants, we surprisingly found a lower risk of death from external causes amongst female undocumented migrants, compared to Belgian women.

In contrast to male undocumented migrants we do not find statistically significant differences between female undocumented migrants and Belgian women/documented migrants for most underlying causes of death. There are several possible explanations for these different findings in undocumented males and females. First, the sample size of the female undocumented migrants was relatively small. Consistent with the higher number of male undocumented migrants in Belgium [5], the male sample was twice as large as the female sample. Another possible explanation for the important gender differences is the different access to healthcare and shelters, which has been described in qualitative research. Despite equal health rights, undocumented women seem to have better access to healthcare and social care through informal practices identifying women as victims (Author, forthcoming). Possibly, the differences also reflect a difference in the population characteristics or differences in living and working conditions between undocumented males and females. These different factors could (partially) account for the gendered differences in mortality risks. Similarly, earlier research shows that migration-related health inequalities can affect man and women differently [17].

This study shows important differences in the mean age at death and underlying causes of death in undocumented migrants compared to Belgian residents and documented migrants.

Most importantly, male undocumented migrants in Belgium have a statistically significant increased risk of dying from cardiovascular diseases. Our study does not allow to determine whether this is caused by lifestyle factors, reduced access to primary prevention, reduced access to treatment, or a combination of these. This requires further research.

Nevertheless, our analyses on causes of death provide information on lacunas in health services. Cardiovascular deaths are considered a source of avoidable mortality. This excess risk of cardiovascular death could be avoided through better access to healthcare and improving wider determinants of health such as lifestyle factors, daily living conditions and socio-economic status. These findings indicate that, to address the issue of poor health among undocumented migrants, we must rather improve general healthcare provision for undocumented migrants and focus on social determinants of health, rather than focussing on improving health services for specific illnesses or particularly ‘vulnerable’ groups.