The relationship between non-permanent migration and non-communicable chronic disease outcomes for cancer, heart disease and diabetes – a systematic review

Rapid economic development has resulted in sustained global migration in the past several decades. The United Nations Development Program estimates that nearly 800 million migrants have moved permanently to either another country or another location within their own country’s borders for the purpose of resettlement [1, 2]. The impact of migration on economics and development has been a long-standing focus of the global community [3, 4]. The spread of communicable diseases from influenza to HIV continues to highlight the public health implications of migrating communities on health and health systems [5]. In the context of an increasing global burden of non-communicable chronic diseases (NCDs) [6], migrating communities will add an additional complexity to the role of health systems in improving population health.

The vast majority of NCD research among migrants has concentrated on identifying and addressing health needs and improving the healthcare of “permanent” migrants, i.e. immigrants (those who have resettled in another country) or rural-to-urban migrants (who have resettled within their country’s border) [7, 8]. Among NCDs, research has primarily focused on cancer, cardiovascular diseases, including diabetes, as they are the leading contributors to death worldwide [9‐12]. With this large body of literature, some evidence suggests immigrants have better health outcomes than local populations, known as the ‘healthy migrant effect’ [13]. The suggestion is that there may be a selection effect, wherein healthier migrants are more likely to immigrate to another location, and thereby have better health outcomes than the local, non-migrant population [14, 15]. Studies have demonstrated, for example, lower cardiovascular disease and risk factors among immigrants as compared to locally born populations [13, 14, 16]. In contrast, other evidence suggests that immigrants continue to have poorer outcomes compared to local populations [7, 8, 17‐19], such as higher rates of diabetes among immigrants in Canada or among Afro-Caribbean immigrants as compared to non-immigrant white populations in different settings globally [20‐22]. Similarly, research on rural-to-urban migrants has documented this same equivocal trend in NCD rates of disease across differing rural populations [23‐25]. These differences in the burden of NCDs and their complications between permanent migrant populations and non-migrant populations are important to understand to guide necessary prevention and management strategies [26, 27].

Another large body of research focuses on identifying and reducing NCD disparities in healthcare, specifically NCD screening disparities between permanent migrants and other populations [28]. A study in Canada, for example, demonstrated varying rates of diabetes screening in immigrants, with some sub-populations of immigrants having lower screening rates than the non-immigrant population [29]. Other studies of cancer disparities have demonstrated differential rates of cancer diagnosis in immigrant populations as compared to non-immigrant populations. Furthermore, many studies have demonstrated lower rates of cancer screening among immigrant populations as compared to other population groups [30, 31].

As migration is a dynamic process not often captured by literature on permanent migrants, there is a need to better understand the sub-populations often aggregated into the category of “migrants” [32]. Some researchers have explored migration-related attributes beyond permanent migration and their relationship to NCD outcomes to shed light on health differences between migrant populations and other populations [2, 33‐35]. These attributes include, but are not limited to, length of time outside of place of origin; distance from place of origin; or the unit of migration (e.g., an individual or an entire family) [36]. Type of migration, other than “permanent” migration, is also a migration-related attribute that can influence health outcomes [36‐45]. The “non-permanent” migrant populations (such as those who are temporary, circular, return, or double-leap/secondary migrants) are important to examine because: 1) with globalization, the volume of “non-permanent” migrant population is likely to grow; and 2) they likely represent a heterogeneous population who are at risk for NCDs [32, 38, 46].

We hypothesized that non-permanent migrants would have higher rates of disease and lower rates of chronic disease screening, as the dynamic livelihoods of non-permanent migrants may be a signal of greater vulnerability as compared to other migrant populations, based on World Health Organization conceptual model for the influence of social and structural determinants of health on the health of individuals [47]. Different than permanent migrants, the geographic mobility of non-permanent migrants may be a manifestation of the influence of contextual factors in different localities (such as discrimination, legal status, economic opportunity, available housing, etc.) and may independently affect health and healthcare outcomes, positively or negatively. The identification of greater rates of disease in non-permanent migrant populations as compared to other populations is essential to inform strategies to prevent and manage NCDs for these populations across geographic contexts. Moreover, identification of lower rates of screening in non-permanent migrant populations would necessitate a reassessment of the delivery of healthcare services to these populations.

As a step to advancing scientific understanding of these populations and their healthcare needs, we conducted a systematic review focused on the health of non-permanent migrants. Specifically, we examined the literature to compare non-permanent migrants to other populations for the NCDs, specifically cancer, heart disease, and diabetes, which are the main NCD contributors to mortality globally [9‐12]. We were interested in rates of disease and screening as initial steps to understating health and healthcare needs. Since we anticipated variation in how non-permanent migrant populations were described in the literature, we also sought to describe how non-permanent migration was defined and measured.

The search of databases provided a total of 11,492 citations (Fig. 1). Our review of grey literature and non-indexed journals resulted in zero papers for inclusion. Twelve articles met our inclusion criteria [40, 53‐63]. The number of abstracts and full texts reviewed at each stage are shown in Fig. 1.

This systematic review of the literature found 12 studies that compared non-permanent migrants with another population to examine differences in NCD outcomes for cancer, heart disease, and diabetes. Our main finding was that NCD outcomes of non-permanent migrants were similar to or worse than other populations. However, these equivocal findings likely reflect other important conclusions of this review, such as the heterogeneity of 1) study designs, 2) non-permanent migrant groups by migration attributes and 3) spatial and temporal measurements of migration.

Some studies found significant differences in NCD prevalence between non-permanent migrant populations and other population groups. For example, of the four studies on heart disease, all documented higher rates of self-reported heart disease among the non-permanent migrants as compared to other groups [58‐61]. In contrast, of the five studies on diabetes, only one study demonstrated that non-permanent migrants had higher prevalence than other non-migrant groups; the other studies showed no difference [40, 53, 54, 58, 62]. In light of the variation in study designs, measurements, and research contexts, it is challenging to make strong conclusions about differences between non-permanent migrant groups and other population groups. However, differences within non-permanent migrant groups and between these groups and others likely exist, as past literature on permanent migrants suggests multifactorial influences on disease risk including genetics, environmental exposures, and social-behavioral factors [35, 64]. For non-permanent migrants, we expect there may be factors affecting health risk and outcomes, complicated by the additional influence of having resided in different geographies, with varying numbers of movements, over different periods of time. Non-permanent migrants may also have social and family ties across geographies that influence health risks and outcomes [45, 65‐68]. However, the strength of these ties may differ as a result of varying migration attributes, such as the migration unit (i.e. the individual or family), distance from one’s network, cultural identities and broader socioeconomic dynamics in countries of origin [69‐71]. We infer that these contextual and social-level factors in these different geographies confer unique risks and opportunities that affect an individual’s health [38, 72‐74]. Future work should investigate the influence of these factors on the health behaviors and practices of non-permanent migrants in different geographic locations.

The included studies represented diverse migrant populations in their motivations and their migration-related spatial and temporal attributes. For example, the studies represented a broad range of motivations that led to an individual’s non-permanent migrant status, including for labor reasons (e.g. farmworkers) and lifestyle (e.g. nomadic populations), that would affect outcomes [54‐56, 58‐61]. The geographic borders crossed were both regional and national and are important to identify and understand, as living in one context or another confers different opportunities and challenges for migrants. Moreover, if measured, the time period for the measurement of migration ranged from seasonal to discrete counts of a migration event [40, 57, 59, 60]. This issue has been discussed in migration literature, where the temporal attributes of migration such as frequency, duration, and timeframe are not consistently measured, thereby hindering opportunities for generalization and comparison in migration research [75].

A review of how the included were indexed within the bibliographic databases suggests a need for more granular and standardized vocabulary around non-permanent migration. Three of the 12 articles were retrieved from keyword searches with use of the terms “migrant” or “farmworker” [54, 55, 63], eight were retrieved based on their categorization into the broad medical subject headings like “Emigration and Immigration” and “Transients and Migrants” [40, 53, 56‐59, 61, 62]. These findings are consistent with the broad International Office for Migration (IOM) definition of migration [2, 76]. A few other definitions exist that may serve to provide a framework for this population, but these are also limited. For example, per IOM, a short-term migrant is “a person who moves to a country other than that of his or her usual residence for a period of at least three months but less than a year, except in cases where the movement to that country is for purposes of recreation, holiday, visits to friends or relatives, business or medical treatment” [50, 77]. This definition excludes migrants who are regionally mobile within national geographic borders, those who may remain in one locality for more than one year, or those who have no usual residence. Refinement and standardization of the language used to describe non-permanent migration will enable further examination of the health behaviors and health outcomes of these populations.

There are several limitations to this systematic review. Our own a priori definition of “non-permanence” based on the number of moves across a border suggesting a non-lasting residence is a limited, but important, facet of non-permanence. Our premise was that these individuals were geographically mobile, and therefore may also be residentially mobile, to differentiate from the concept of “permanent” migrant whose definition is tied to having a usual residence. However, the concept of residence/residential mobility limits our understanding of non-permanence. ‘Residence’ is less central to the concept of non-permanent migration as compared with permanent migration, as non-permanent migrants may have no usual residence or no residence at all [49]. The definition utilized here demands further development and investigation to capture the diversity of different non-permanent migrant populations, to highlight the different motivations for migration, and to better understand the influence of contextual factors in different locations. However, these findings reflect the current limitations and challenges in this field of study, where there is wide variation in vocabulary, definitions, and measurement across disciplines and limited availability of data [35]. Article quality of included studies also varied and limited the ability to generalize or make inferences about the relationship between non-permanent migration as compared to other populations and NCD disease and screening rates. As prior research studies have noted, variations in terminologies, design, and quality challenges comparisons across studies and underscores the need of further systematization to establish standards of practice [35].

The limited number of studies on non-permanent migrants is not surprising. Some studies on non-permanent migrant populations (such as labor migrants, nomadic populations, and farmworkers) that were excluded from our final analysis did not include a comparison group [78‐83]. Each migrant population has distinct individual and societal factors that influence health and health outcomes, which makes finding an appropriate comparison group challenging [35, 84]. The limited number of studies on non-permanent migrants from low- or middle-income countries likely reflects our inclusion criteria, such as requiring a comparison group or outcomes of interest [59, 81, 85, 86]. Though it is possible we did not capture all articles for inclusion into our review because of the diversity of terms for migration, to minimize this possibility, we created a comprehensive search string, reviewed articles in the references list of included articles, and reviewed relevant grey literature. Moreover, our analysis focused on cancer, heart disease, and diabetes, yet non-permanent migrants are affected by other non-communicable diseases, such as chronic obstructive lung diseases and obstetric diseases, that are influenced by social and structural determinants of health [87, 88]. Future work can investigate these populations in low- and middle-income countries and expand outcomes of interest.

More standardized measurements for the spatial and temporal attributes of non-permanent migration and explicit documentation of definitions are needed [57, 59, 62]. Consistent use of a multilevel framework for migration may inform standardization of variable selection and definitions used. In peer-reviewed literature on permanent migrants, several conceptual models have been used to define variables that influence individuals’ health outcomes [89‐91]. These models could be adapted for research on non-permanent migrants to account for the multiple contexts in which non-permanent migrants have lived and how these contexts influence their health and health seeking behaviors. Researchers could also begin to consistently use the term “non-permanent” to describe these populations in distinct contrast to permanent migrants. With greater standardization, we can then incrementally build upon the limited knowledge about the different typologies of non-permanent migration, advance the science around migrating populations, and improve public health systems’ abilities to plan for and care for these populations.

The confluence of non-permanent migration in the context of the growing burden of NCDs is an important area of investigation that requires greater examination. As research examining non-permanent migrants’ health grows, standardized vocabulary, definitions, and measurements of migration attributes are key mechanisms for organizing data to facilitate indexing and synthesis of results across multiple contexts. This work has implications not only for the health of migrant populations but also for the public health systems that care for these populations. Currently, health prevention, screening, and treatment protocols may vary across these locations, and health information about these non-permanent populations is not often accessible in different regional or international areas [92, 93]. Public health and healthcare systems may need to develop different strategies to minimize health risks of NCDs for non-permanent migrants across geographies.