KEY CONCEPTS IN THE RELATIONSHIP BETWEEN MIGRATION AND HEALTH

3.2.1 Selective Migration

Most types of migration have been described as highly selective and, for a variety of reasons previously discussed in Chapter 2, people who move often have been those who are most able to do so from a health status point of view (Carballo, Divino and Zeric, 1998). Selective migration refers to the evidence that international migrants might not be a random sample from their home countries (Feliciano, 2005; Palloni and Arias, 2004; Marmot and Adelstein, 1984). Research has suggested that most prime-aged migrants moved in search of better labour market opportunities and, because they had the motivation and resources to undertake a move, they were "positively selected". This positive selection has meant that migrants might have been more educated and in better psychological and physical health than non-migrants (Rubalcava et al. 2008; Razum, Zeeb and Rorhman, 2000).

In Latin America, little research in this field has been conducted and most is over 20 years old (Cotlear, 1984; Draper, 1985; Torrealba, 1991; Gurak, 1996). The few studies tended to support the idea of selective international migration, with the need for migrants to improve their current economic situation the most relevant reason for mobility. Permanent and temporary/ seasonal migration was selective, in that healthy adults were more able to move to another region for work. This may also be the case nowadays in Chile, as a growing proportion of young adults have immigrated in the last few years (see Chapter 2).

3.2.2 The Healthy Migrant Effect

As argued by Fennelly (2005), there has been a growing body of literature describing the healthy migrant phenomenon in different countries around the world, especially in Europe and the US.

This phenomenon has been related to the fact that on many measures, first generation immigrants are often healthier than the host-born residents who share similar ethnic or racial backgrounds (Neria, 2002; Fuentes-Afflick, Hessol and Eperez-Stable, 1999; Singh and Siapush, 2001; Muening and Fahs, 2002). Several studies have tested the healthy migrant effect in in different geographical and social migratory contexts, and findings tend to support the real existence of selective migration and the healthy migrant effect in some immigrant groups.

92

The healthy migrant effect has been observed in the USA, in Canada (Hyman, 2001), Australia (Autralian Institute Health, 2000) and countries in Western Europe (Razum, Zeeb and Rorhman, 2000; Toma, 2001; Swerdlow, 1991). For example, foreign-born residents had lower levels of obesity, hypertension, diabetes, cardiovascular diseases and serious psychological distress than US-born residents (Dey and Lucas 2004; Fennelly, 2005). Immigrants also showed a

significantly lower risk of adult and infant mortality (Singh and Siahpush, 2001; Fennelly, 2005) with longer life expectancies than their native born counterparts, and lower rates of breast and cervical cancer, sexually transmitted infections, heart disease, diabetes, teenage pregnancy, suicide, and tobacco and alcohol consumption (Fennelly, 2005). Despite these findings, some authors have questioned the existence of selective migration, as conflicting results have been found in some studies (Rubalcava et al., 2008). Four main arguments have been made in this matter and are presented in Table 3.2.

Table 3.2 Four arguments against the healthy migrant effect Argument Explanation

Inappropriate comparison group The appropriate comparison group to test the hypothesis should be non migrants from the home country rather than natives in the receiver country.

Inadequate timing Studies have typically examined the health of migrants after they moved to the host country rather than prior to migrating.

Self-report bias in multicultural

settings Most of the existing research has relied on self-reported health, information that has been shown to depend upon other broader variables like cultural factors, ethnicity, and access to health care (Crimmins, 2005; Ren and Amick, 1996).

Denominator bias The lower mortality among immigrants compared to the host population in registered studies could largely be explained by inaccurate denominator figures (Weitoft et al., 1999, Abraido-Lanza, 1999). Mortality and morbidity rates would be

underestimated since a relevant proportion of migrants might have already moved despite being counted as registered in the host health care system or not being reached and counted due to their

undocumented status (Gee, Kibayashi and Prus, 2003; Razum, Zeeb and Rorhman, 2000; Swerdlow 1991; Kliewer, 1992; Razum 1999).

Large denominators lead to an underestimationof morbidity and mortality estimates among immigrants in a country.

93 3.2.3 The paradox of assimilation

It has been observed that over time the migrant health advantage diminishes dramatically in some countries. In what Rumbaut (1997) called the “paradox of assimilation,” the length of time that an immigrant spent in the US has been correlated with increases in adolescent risk

behaviours (Fennelly, 2005), low birth weight infants (Fuentes-Afflick, 1999; Peak and Weeks, 2002), anxiety and depression (Finch and Vega, 2003), cancer (Fennelly, 2005), general

mortality (Singh and Siapush, 2001; Muening, 2002), life expectancy, risk of illness, patterns of deteriorating health, cardiovascular disease, body mass index, and hypertension (Lassetter and Callister, 2009).

Lassetter and Callister (2009) stated in a recent review of the literature on the health of voluntary migrants to Western societies, that multiple factors could explain variability found in health outcomes in this population. These included length of residence and acculturation, disease exposure, life style and living conditions, risky behaviours, healthy habits, social support networks, cultural and linguistic barriers, experiences with racism, and levels of awareness of cultural health practices among health care providers. Overall, migration is a dynamic, extended process, with effects extending years beyond physical relocation. Systemic change has been required, including health policies that ensure equity for migrants, culturally appropriate health promotion, and routine assessment of migration history, cultural health practices, and disease exposure (Lassetter and Callister, 2009).

3.2.4 The Latino paradox

Many studies have shown that lower socioeconomic status (SES) is related to poor health, in terms of both morbidity and mortality (Adler, 1994; Abraido-Lanza, 1999). In the USA, compared with non-Latino Whites, Latinos have had higher poverty rates, less education, and less health insurance, but despite this they have a lower all-cause mortality rate (Markides, 1986). Latinos have lower income-adjusted mortality rates for cancer, cardiovascular disease, and all-cause mortality relative to non-Latino Whites (Sorlie et al., 1993; Kaufman et al., 1998).

Other studies have shown that Latinos exhibit better health than Whites, for outcomes that include birth weight and infant mortality, even after adjusting for socioeconomic status (Dubowitz et al., 2007; Abrams and Guendelman, 1995; Norman, Boyle and Rees, 2005;

Norman et al., 2004; Marmot, 1981; Sorlie et al., 1993). In the US, Latino mortality has stood in sharp contrast to that of African Americans, who, like Latinos, have had a lower SES profile

94

than Whites but a higher mortality rate (Abraido-Lanza, 1999). Thus, Latino mortality has presented an epidemiologic paradox, defined by several authors as the mortality advantage of Hispanic adults relative to non-Hispanic Whites, despite the lower socioeconomic status of Hispanics (Rubalcava et al., 2008; Hummer et al., 2000; Sorlie et al., 1993). Similar findings have been recently reported in the UK (Shaw & Pickett, 2011).

3.2.3 The salmon bias

A relevant hypothesis on migration and health is salmon bias, which proposes that reflecting the desire to die in one's birthplace, many immigrants return to their country of birth after temporary employment, retirement, or becoming seriously ill (Adler, 1994; Pablos-Mendez, 1994). As declared by Abraido-Lanza (1999), because foreign deaths have not been tabulated in US mortality statistics, some individuals were rendered statistically immortal, resulting in an artificially low Latino mortality rate (see Table 3.2 on the denominator bias). Evidence has suggested that the salmon bias hypothesis could be plausible. Sorlie and colleagues (1993) estimated return migration rates of various foreign-born groups based on data from a program requiring immigrants to submit yearly address reports to the Immigration and Naturalization Service in the US. Lower- and upper-bound return migration estimations (assuming a 50% and 100% response rate for filing address reports) ranged from 15.6% to 56.2% for Mexicans, 52.4%

to 72.5% for South Americans, and 49.6% to 69.5% for Central Americans and Caribbean persons (excluding Cubans).

95