Abstract

Adverse life events are major causes of declining health and well-being, but the effects vary across subpopulations. We analyze how the intersection of migration status and sex relates to two main adverse life events—job loss and divorce—thereby affecting individual health and well-being trajectories. Using data from the German Socio-Economic Panel (1984–2017), we apply descriptive techniques and individual fixed-effects regressions to analyze how job loss and divorce influence the health of immigrants and nonimmigrants. Our results support the hypothesis that immigrants suffer more from adverse life events than nonimmigrants in both the short and the long run. Relative to nonimmigrants, immigrants have a health advantage at younger ages, which becomes a disadvantage at older ages, and this faster decline at older ages is particularly steep among immigrants who experience adverse life events. These results help explain the vanishing health advantage of immigrants by showing that they are exposed to a double disadvantage over the life course: immigrants are more likely than nonimmigrants to suffer from adverse life events, such as job loss, and these events typically have a larger impact on their health. Our findings are the first to provide evidence regarding the consequences of different adverse life events and how they relate to the intersection of migration status and sex. Moreover, our results highlight the importance of intersectional analyses in research on immigrant health.

Introduction

Although immigrants are usually in good health when they arrive in the receiving countries owing to health selection, they face challenges in preserving their initial health capital. Some of the largest risk factors contributing to health deterioration among immigrants are their social, economic, and labor market disadvantages relative to nonimmigrants. These sources of disadvantage accumulate over the life course, before and after migration, and they intersect with one another to exacerbate their consequences on immigrants’ health and well-being (Viruell-Fuentes et al. 2012). The disadvantages immigrants experience can be attributed not only to structural factors, such as their immigrant status and gender (Viruell-Fuentes et al. 2012) and socioeconomic conditions, but also to their greater exposure to adverse life events relative to nonimmigrants (Leopold et al. 2017).

Job loss and divorce are major life course events that clearly and strongly affect health and well-being (Paul et al. 2018; Sbarra 2015; Schaller and Stevens 2015). However, most of the literature examining health and well-being as functions of structural differences and adverse events considers them to be distinct processes. Recent research has started documenting the effects of job loss on immigrants’ well-being (Leopold et al. 2017), but this discussion still lacks knowledge of immigrants’ health. Intersectionality provides a sound theoretical and empirical framework for conceptualizing interactions among the social, economic, and labor market risk factors and their impact on health (Bauer 2014). In particular, it provides theoretical support for studying the consequences of the interplay of structural differences when examining immigrant health (Viruell-Fuentes et al. 2012).

This study examines the role of adverse life events in the health gap between immigrants and nonimmigrants while focusing on Germany as an important case study, given its prominence as a receiving country in Europe. Our analysis uses data from the German Socio-Economic Panel, a representative survey of the general and the immigrant population in Germany. We focus on two adverse events—job loss and divorce—and examine sex and migration status as structural risk factors. To our knowledge, no study has tested the interplay of multiple layers of structural differences and adverse life events to examine their effects on immigrant health and well-being.

We study (1) whether the relationship between age and migration status differs depending on having experienced adverse events during the life course; (2) whether the short- and long-term health consequences of experiencing a job loss or a divorce differ between immigrants and nonimmigrants; and (3) whether experiencing both adverse events (job loss and divorce) has a stronger impact on health.

Theoretical Background

Immigrants often have better health than nonimmigrants in receiving societies, which is paradoxical because their disadvantaged background and poorer socioeconomic status would be expected to yield poorer health. The immigrant health advantage is usually explained by health selectivity, also known as the “healthy immigrant effect”: the healthiest and strongest individuals are the most likely to leave their origin countries and successfully reach their destination countries. The health advantage, however, vanishes relatively quickly during their stay (Goldman et al. 2014; Loi and Hale 2019). Evidence suggests that the reasons for losing the health advantage are acculturation, negative assimilation, and poor socioeconomic conditions (Lechner and Mielck 1998; Palloni and Arias 2004; Ronellenfitsch and Razum 2004).

Immigrants are known to be disadvantaged in many spheres of personal and societal life in receiving countries. For example, they occupy lower skilled positions in the job market and often have more stressful and physically demanding jobs, including the so-called three-D (dirty, dangerous, and difficult) jobs (Orrenius and Zavodny 2013).

A growing body of research recognizes that health disparities are driven by the complex interplay between structural factors and other sources of social disadvantage (Bauer 2014; Mandelbaum 2020). Individuals’ lives are structured unequally (Graham 2007), and inequalities in health are driven by social inequity and are structured across socioeconomic factors (Bauer 2014). The intersection of multiple layers of disadvantage widens disparities with age and over time as adverse exposures accumulate during the life course (Bowleg 2012). Therefore, unitary approaches are insufficient to explain the complexity of the social world, particularly the complicated set of factors underlying persistent health inequalities. Unitary approaches to studying these inequalities assume that single categories operate additively and are layered atop one another (Bauer 2014). Intersectionality, by contrast, recognizes and seeks to explain how multiple social identities—such as race, migration status, and gender—intersect at the micro level to reflect interlocking systems of privilege and oppression at the societal level (Atewologun and Mahalingam 2018; Bowleg 2012). Hence, intersectionality examines inequalities simultaneously across multiple dimensions and social groups (Atewologun and Mahalingam 2018).

Scholars have recognized that intersectionality is exceptionally well suited for investigating health disparities, given its emphasis on how multiple systems of oppression (e.g., racism, classism, sexism) simultaneously affect health (Bauer and Scheim 2019; Bowleg 2012; Green et al. 2017; Hankivsky 2012; Harari and Lee 2021; Viruell-Fuentes et al. 2012). However, progress in integrating intersectionality into quantitative health disparity research has been slow owing to several limitations, as a recent systematic review noted (Harari and Lee 2021). These limitations include the lack of use of life course perspectives and the narrow measurements of intersectional groups, focusing primarily on racial and ethnic differences and less on migration background. In this article, we address these limitations by focusing on health disparities by migration background and adopting a life course perspective.

Adverse events are negative experiences throughout the life course that clearly and strongly affect health and well-being (Schaller and Stevens 2015). These events, which include job loss (Leopold et al. 2017; Lucas et al. 2004; Paul and Moser 2009) and divorce (Lucas 2005), are highly stressful when they occur and as they accumulate over the individual's life course. Nevertheless, they may affect individual health and well-being differently over the short and long term. The path linking these adverse life events to health and well-being is complex. Job loss is associated with long-term losses of earnings and psychosocial assets, as well as with social withdrawal, family disruption, lower job quality, and declines in psychological and physical well-being (Brand 2015). Job loss affects an individual's health and well-being not only by causing a reduction in the person's earnings and material conditions but also through strong psychological effects that lower the individual's general quality of life (Paul et al. 2018; Schaller and Stevens 2015).

Unemployment generally has a greater impact on men than women. Western societies have seen a decrease in the conventional male breadwinner family, in which a husband works full-time and a wife is a full-time caregiver (Trappe et al. 2015). However, men still more often identify as workers and family providers, and women might more frequently become stay-at-home parents or homemakers (Leopold et al. 2017). Therefore, the relationship between job loss and health might be patterned by sex.

As Leopold et al. (2017) noted, the negative consequences of unemployment differ across social groups (Paul and Moser 2009). Poverty, loss of social status linked to unemployment, and family strain are the main drivers of the detrimental effects of job loss, which disproportionally affects immigrants relative to nonimmigrants (Leopold et al. 2017:232).

Divorce is one of the most stressful life course events (Bloom et al. 1978). A broad range of research has found that marital separation and divorce are linked to a high risk of experiencing poor health outcomes, including hospital-diagnosed infectious disease (Nielsen et al. 2014), cardiovascular morbidities (Alviar et al. 2014), and early death (Sbarra and Coan 2017; Shor et al. 2012). The mechanisms behind the link between divorce and poor health include a lack of social and financial resources, cognitive and affective experiences, and harmful health behaviors. However, the literature has noted that individual differences that predict marital dissolution also help predict a person's health after divorce (Sbarra and Coan 2017). Regarding well-being, research suggests that individuals habituate to divorce quite rapidly, reporting that they have greater life satisfaction 5–10 years after divorce than before it (Clark et al. 2008).

The consequences of divorce differ significantly by sex. Women are more vulnerable to the economic consequences of divorce (Leopold 2018). After divorce, women have greater declines in income (de Vaus et al. 2017) and a higher risk of poverty (Smock et al. 1999). However, men are more affected by the negative consequences of divorce on health and well-being (Shor et al. 2012) and mortality (Berntsen and Kravdal 2012; Sbarra et al. 2011). A study of the economic, housing and domestic, health and well-being, and social consequences of divorce (Leopold 2018) found that women and men did not differ much; when sex differences emerged, they were primarily short term. However, the study noted that the apparent contradiction with expected sex differences might be explained by the fact that some of these factors hint at opposing forces that might balance each other (Leopold 2018).

Because of their initial health advantage, immigrants could be resilient to the effect of adverse life events on health. However, as their duration of stay increases, immigrants experience increased exposure to structural disadvantage; the positive traits owing to health selectivity at arrival could thus be whittled down, leaving immigrants vulnerable to the negative impacts of adverse life events on health. On the other hand, because divorce is less common among immigrants (see Table 1 in the Results section), it might be more culturally sanctioned than for nonimmigrants. Therefore, experiencing divorce may have a more negative impact on the health of immigrants versus nonimmigrants. Furthermore, differences in marital status and the likelihood of divorce between immigrants and nonimmigrants might be driven by the characteristics of the immigrant's migration background, such as the reason for migration and the country of origin. For instance, individuals who immigrated for family reunification could be more predisposed to stay married if their residence status depends on it. Depending on the country of birth, divorce might be more or less culturally accepted and, therefore, common. If so, the consequences of divorce could be especially hard for immigrants.

Studying the consequences of adverse events on health and well-being requires a consideration of the timing of these events and the duration of their effects. Previous research has shown that some individuals experience only transient effects on their well-being after significant (positive or negative) life events and return relatively quickly to stable well-being levels (Luhmann et al. 2012). Further, well-being varies considerably across groups both in the period immediately after the event and over the longer term (Luhmann et al. 2012).

Our study builds on and extends previous theoretical knowledge and empirical evidence about the consequences of adverse life events on immigrant health in several important ways. First, we seek to understand how adverse life events intersect to produce or exacerbate health disparities. Most previous research on adverse life events and their effects on immigrant outcomes analyzed only one adverse life event at a time, such as the effects of job loss on an immigrant's well-being (Leopold et al. 2017), while neglecting the intersection of multiple layers of social disadvantage and the impact of the accumulation of disadvantage in different social areas. Leopold and colleagues (2017) studied well-being trajectories and found a steeper deterioration in well-being among immigrants than nonimmigrants. We use a similar approach to study health by comparing the health trajectories of immigrants and nonimmigrants. However, we extend this approach to provide a comprehensive picture of the health consequences of the intersectional accumulation of disadvantage (adversities) in different social spheres by migration status and sex and by the adversities that individuals experience. A second novelty of our study is that we consider individuals’ health trajectories before and after these events to analyze both the short-term and the long-term effects of experiencing multiple adversities, and consider the individuals’ health status before the adversities occurred. By focusing on German-born individuals and immigrants aged 18–64, we examine the main relationship between health and job loss, as well as the relationship between health and divorce and their interplay with migration status. We also examine how experiencing or not experiencing adverse events affects individuals’ health trajectories over the life course and the related age trajectory crossover.

Given that the health gap between immigrants and nonimmigrants reverses over the life course, such that immigrants have better health at younger ages but poorer health at older ages, we formulate three hypotheses.

  • Hypothesis 1: The immigrant health disadvantage at older ages is worsened by adverse events experienced during the life course: immigrants who experienced adverse events will have worse health at older ages than nonimmigrants who experienced adverse events, particularly when more than one adverse event occurs during the life course.

  • Hypothesis 2: The consequences of job loss and divorce on health differ between immigrants and nonimmigrants: immigrants have stronger consequences on health after experiencing these adverse events; the short-term and the long-term impacts of adverse events on health differ between immigrants and nonimmigrants, such that immigrants face more long-term negative health consequences than nonimmigrants after experiencing adverse events.

  • Hypothesis 3: The immigrant disadvantage in the consequences of job loss and divorce is further exacerbated when experiencing both events during the life course.

Data and Methods

We use data from the 1984–2017 waves of the German Socio-Economic Panel (G-SOEP). The SOEP study is a representative, longitudinal study of private households in Germany. These data provide information on all surveyed household members. The survey participants include Germans living in former West and East Germany, foreigners, and recent immigrants to Germany. The survey covers household composition, occupational biographies, employment, earnings, and health and satisfaction indicators. Immigrants were resampled to account for the changes in German society in 1994/1995, 2013, 2015, and 2016. New samples were added in 1998, 2000, 2002, 2006, 2009, 2011, and 2012.

Key Measures

Our key outcomes are self-rated health (on a scale from 1 to 5), satisfaction with one's health (on a scale from 0 to 10), and well-being (on a scale from 0 to 10). Information on self-rated health was collected by asking, “How would you describe your current health?” The response options were “very good,” “good,” “satisfactory,” “poor,” and “bad.” Information on satisfaction with one's health was collected by asking, “How satisfied are you today with the following areas of your life: Health? Please answer on a scale from 0 to 10, where 0 means completely dissatisfied and 10 means completely satisfied.” Information on well-being was collected by asking, “How satisfied are you with your life, all things considered? Please answer on a scale from 0 to 10, where 0 means completely dissatisfied and 10 means completely satisfied.” We report results for self-rated health in this article and provide results for the other two outcomes in the online appendix.

Migration status is defined by place of birth: individuals born outside of Germany are classified as immigrants, and individuals born in Germany are classified as nonimmigrants (the reference category in all analyses). We focus on two adverse events: job loss and divorce. We measure job loss as the individual's change from employment to registered unemployment between waves, and we measure divorce as the individual's change from being married to being divorced between waves. In both cases, we focus on the first episode of the event occurring during the observation window. We measure the first change observed, meaning that the information from the first wave is excluded from the analyses. When analyzing the two adverse events simultaneously, we first focus on job loss as the main event while considering whether individuals also divorced before experiencing job loss over their life course. In separate models, we focus on divorce as the main event in the context of having previously experienced job loss.

Empirical Approach

We use random-effects models clustered by the individual ID to describe the age trajectories of health. We use individual fixed-effects linear models to estimate the within-individual response change in health following the adverse life events of job loss and divorce. By estimating the within-individual change, fixed-effects models implicitly control for all possible unobserved confounding characteristics as long as those characteristics do not change over time (Allison 2009). We stratify the sample by sex, and we focus our analyses on the adult population (aged 18–64) because we aim to study individuals at risk of being married (those aged 18 or older) and of working age, who are therefore likely to still be active in the job market (younger than 65). Limiting our analyses to individuals younger than 65 partly limits (although does not fully correct for) the so-called salmon bias. This bias can occur because some immigrants return to their country of origin frequently at older ages, contributing to the overestimation of the share of the immigrant population in good health relative to that of the nonimmigrant population (Turra and Elo 2008).

Testing Hypotheses 1

We estimate differences in immigrants’ and nonimmigrants’ health and well-being trajectories across ages with random-effects models and on the three outcomes. To do so, we estimate a model that includes age; migration status (German-born vs. foreign-born); and an interaction between migration status, having ever experienced the event, age, and quadratic and cubic terms of age. This strategy allows us to test the hypothesis of the differential relationship between adverse events during the life course, age, and health depending on migration status. We hypothesize that individuals who experienced such adverse events and have the social disadvantage of being foreign-born will experience a steeper and faster decline in health by age:

Yit=αi+a=13βaageta+γmigration_statusi+δeventi+θmigration_statusieventi+a=13μaagetamigration_statusi+a=13ρaagetaeventi+a=13φaagetamigration_statusieventi+εit,
(1)

where Yit is the outcome of person i at time t; β is the individual random effect; ageta,a=1,2,3, are linear, quadratic, and cubic terms of age; age, event (ever experienced job loss or not), and migration status are interacted with each other and together; and εit denotes the random error of person i at time t.

Testing Hypotheses 2

We use fixed-effects models to estimate the individual's change in health and well-being following changes in employment status (employed to unemployed) and in marital status (married to divorced) by migration status. To do so, we test a model with linear, quadratic, and cubic terms of age and lags since the adverse event interacted with migration status. We test a model in which the adverse event is job loss and a separate model in which the adverse event is divorce. We aim to test the hypothesis that the consequences of job loss and divorce on immigrants’ and nonimmigrants’ health and well-being differ over the short and long run.

Yit=αi+a=13βaageta+l=55γllagitl+l=55δllagitlmigration_statusi+εit,
(2)

where Yit is the outcome of person i at time t; ageta are linear, quadratic, and cubic terms of age; and αi is the individual fixed effect. Lagitl,l=5,. . .,5, are dummy factors of lags indicating when the event occurred in year intervals [–10, −5], (−5, −4], (−4, −3], (−3, −2], (−2, −1], (−1,0], (0,1), [1,2), [2,3), [3,4), [4,5), [5,15], where (−1,0] is the time of the event, [–10, −5] is used as the reference, and the lags are interacted with migration status. We define the short-term effect as the observed change in health within the first year and the long-term effect as the trajectory of change after two or more years.

Testing Hypotheses 3

We add another term to the interaction tested in the previous step, including the lag since job loss, migration status, and divorce that occurred before job loss in models that test job loss as the main adverse event. In a model that tests divorce as the main adverse event, we include an interaction between the lag since divorce, migration status, and job loss that occurred before divorce. We aim to test the hypothesis that the intersection between adverse events (job loss and divorce) and multiple strata of social disadvantage (being foreign-born, unemployed, and divorced) accelerates health deterioration.

Yit=αi+a=13βaageta+l=55γllagitl+l=55δllagitlmigration_statusicovariate+εit.
(3)

The terms in Eq. (3) are the same as those in Eq. (2), except for the additional interaction term.

Results

Table 1 summarizes the descriptive characteristics of the sample used to study job loss as the main adverse event. Table 2 describes the sample used to study divorce as the main event. As expected, we find that immigrants are, on average, younger than nonimmigrants, irrespective of sex. The proportion of individuals who experienced job loss during the observation period, calculated as the first observed event on the full sample of participants, is higher among immigrants than among nonimmigrants: 20.2% and 14.6% among immigrant and nonimmigrant men, respectively, and 17.7% and 15.0% among immigrant and nonimmigrant women, respectively (see Table 1).

However, immigrants are less likely to experience divorce than nonimmigrants, irrespective of sex (see Table 2): among the full sample of participants, 5.9% of nonimmigrant men and 8.2% of nonimmigrant women divorced, compared with only 1.7% of immigrant men and 3.6% of immigrant women. Immigrants are also less likely than nonimmigrants to be employed, but more likely to be married. The largest groups of immigrants in Germany are from Turkey, Poland, and Italy.

Among individuals who experienced job loss, immigrants have a higher chance of staying unemployed (54.0% of immigrant men and 47.7% of women were not reemployed during the follow-up) than nonimmigrants (37.2% of men and 34.4% of women; see Table 1). Immigrants are also less likely to be reemployed within the first year after a job loss (∼30% irrespective of sex) than nonimmigrants (39.8% of men and 37.7% of women). The average time spent in unemployment is 1.94 years for nonimmigrants and 1.93 years for immigrants.

Within the full sample of participants, among immigrants, 1.7% of men and 3.6% of women divorced (measured as the first observed event), compared with 5.9% and 8.2% of nonimmigrant men and women, respectively (see Table 2). Among those who divorced, 68.0% of nonimmigrant men and 74.5% of nonimmigrant women never remarried, as opposed to 78.8% and 82.4% of immigrant men and women, respectively. The average time remaining unmarried is 4.93 years for nonimmigrants and 3.78 years for immigrants.

Among individuals experiencing job loss, a higher share of nonimmigrants divorced before experiencing job loss (10.2% men, 16.2% women) than immigrants (4.4% men, 12.7% women; see Table 1). Among individuals who divorced, a higher share of immigrants experienced job loss before divorcing (32.7% men, 30.8% women) than nonimmigrants (19.4% men, 24.0% women; see Table 2).

Age Trajectories of Self-rated Health and Sex Differences

Figure 1 shows the self-rated health trajectories by age for individuals who never experienced job loss during the observation period and for individuals who experienced job loss at least once, for men and women separately (the full model is shown in Table A1; all tables and figures designated with an “A” are available in the online appendix). Immigrant men who did not experience job loss (panel a) have a health advantage relative to nonimmigrants at younger ages. However, the two trajectories cross at around age 58, with an emerging immigrant health disadvantage at older ages (60+). Among men who experienced job loss (panel b), we observe the same pattern of a health advantage turning into a disadvantage with age, but we also see a much steeper decline in the trajectories of both immigrants and nonimmigrants and, most importantly, a wider immigrant–nonimmigrant health gap.

Women have lower overall levels of self-rated health than men across all ages, regardless of migration status. Like men, women who did not experience job loss (panel c) have a lower overall decline in health than women who experienced job loss (panel d). The age at which the two trajectories cross over is much lower among women (around 40) than men, irrespective of their job loss experience. Note that the immigrant–nonimmigrant gap at older ages is wider among women than among men, irrespective of job loss experience.

Figure 2 shows the self-rated health trajectories by age for individuals who did not divorce during the observation period (panels a and c) and for individuals who divorced at least once during the observation period (panels b and d) by sex (the full model is shown in Table A2). In contrast to our findings for job loss as the main event, we do not observe large differences in the steepness of the health trajectories by age between men who never divorced and those who divorced. However, we observe a younger age at crossover among men who divorced (around 40) relative to those who did not (∼48), and a larger immigrant–nonimmigrant gap after crossover.

Yet, we observe higher levels of self-rated health at all ages among women who did not divorce than among those who did. At younger ages, immigrant women who did not divorce have a health advantage, whereas women who divorced have similar health levels irrespective of migration status. At older ages, the health gap between immigrants and nonimmigrants is much wider among women who never divorced, which is mainly due to the better health conditions of nonimmigrant women who never divorced.

Short- and Long-Term Self-rated Health Trajectories After Job Loss

Figure 3 shows the short- and long-term self-rated health trajectories by time before and after job loss (the full model is shown in Table A3). The job loss event occurs between Time 1 before the event and Time 0.

In panel a, we observe a declining health trajectory before job loss for immigrant and nonimmigrant men. The trajectories have a similar trend irrespective of migration status but diverge starting roughly three years before the event. After the event, immigrants’ trajectory keeps declining, whereas nonimmigrants’ trajectory starts to recover. The two trajectories keep diverging, with nonimmigrants almost returning to pre–job loss health levels and immigrants experiencing further health declines up to four years after the job loss. Four years after the event, immigrants start recovering but do not reach pre–job loss health levels. Among women, the self-rated health of immigrants after job loss is more similar to that of nonimmigrants. Thus, we do not observe wide gaps among women as we did for men. However, we still see a steeper decline with no recovery more than five years after job loss among immigrant women.

Short- and Long-Term Self-rated Health Trajectories After Divorce

Figure 4 shows the short- and long-term self-rated health trajectories by time before and after divorce for men (panel a) and women (panel b) (the full model is shown in Table A4). The event occurred between Time 1 before the event and Time 0. For men, we do not observe a notable drop in self-rated health after divorce. Overall, immigrant men have lower levels of health than nonimmigrant men, although the difference is not statistically significant. However, nonimmigrant men's self-rated health increases in the years after divorce, reaching higher levels than before divorce. For women, immigrants and nonimmigrants experience similar decreases in health after the event of divorce.

Figure 5 shows the short- and long-term self-rated health trajectories by time before and after job loss for individuals who never divorced before experiencing job loss and for those who divorced before experiencing job loss, by sex (the full model is shown in Table A3). For men who did not divorce before experiencing job loss (panel a), who thus did not cumulate these two adverse events during their life course, we observe a pattern similar to that observed for job loss only (see Figure 2, panel a). For men who divorced before experiencing job loss, who thus cumulated two adverse events during the observation period (panel b), we see a much faster-declining trend for both populations. We also observe a gap between immigrants and nonimmigrants, with health levels declining faster among immigrants than nonimmigrants. In this case, neither population reaches pre–job loss health levels. Among women (panels c and d), we do not observe gaps between immigrants and nonimmigrants, irrespective of the prior divorce experience.

Figure 6 shows the short- and long-term self-rated health trajectories by time before and after divorce for individuals who never experienced job loss and for those who experienced job loss before divorcing, by sex (the full model is shown in Table A4). For men who did not experience job loss before divorcing (panel a), we observe a pattern similar to that for divorce only: an increasing trend after divorce among nonimmigrant men and a decreasing trend after divorce among immigrant men (see Figure 4, panel a).

The estimates for men who experienced job loss before divorcing (panel b) are less stable, and the pattern does not differ between immigrants and nonimmigrants. Among women (panels c and d), we do not observe notable differences between immigrants and nonimmigrants, irrespective of their job loss experience before divorce. Instead, we see a more steeply declining overall trend for both immigrant and nonimmigrant women who experienced both adverse events (panel d).

Robustness Checks

We ran additional analyses to ensure the robustness of our results. First, we considered information on the two adverse events in the first wave, including being unemployed and being divorced in the first wave as an event. We found patterns very similar to those shown here but with stronger magnitudes of the associations. Second, we compared our main results with analyses conducted on an additional health outcome: satisfaction with one's health (see Tables A5–A8) and well-being (Tables A9–A12). The results for satisfaction with one's health were qualitatively comparable with those observed for self-rated health. When we compared the results for well-being with those from another study (Leopold et al. 2017), we found expected patterns. Furthermore, we included an additional analysis of the consequences of subsequent reemployment to examine whether reemployment is differentially beneficial for immigrants and nonimmigrants. We observed that immigrant men benefit less from reemployment than nonimmigrants; we found the opposite for women (Figures A1–A3).

Discussion

In a context where immigration is a structural phenomenon with an increasing trend, as in Germany, health trajectories with age and over time might be important drivers of integration. Studying life course trajectories and the role of adverse events can help us better understand the mechanisms behind immigrants’ vanishing health advantage and health deterioration. Immigrants are exposed to a double disadvantage over the life course: they are more likely than nonimmigrants to suffer from certain adverse life events (e.g., job loss), and the impact of such events on their health tends to be larger than among nonimmigrants. Our findings are the first to provide evidence of the consequences of different adverse life events and their interplay with structural differences, such as migration status and sex. Thus, our results highlight the importance of intersectional analyses in research on immigrant health.

In this article, we expanded on prior research on the association between job loss and well-being (Leopold et al. 2017). We provided several innovations that can help explain why experiencing adverse life events contributes to the health gap between immigrants and nonimmigrants and why the immigrant health advantage disappears with age and over time. We estimated immigrants’ and nonimmigrants’ overall health trajectories surrounding two important adverse life events: the loss of employment and divorce.

We found evidence supporting all our hypotheses. First, we showed that the immigrants in our sample are aging while being in poorer health than nonimmigrants. We thus extended the literature that reported a faster decline in immigrants’ satisfaction with their health (Ronellenfitsch and Razum 2004) and self-rated health (Gubernskaya 2015). Results show that immigrants have higher self-rated health than nonimmigrants but that this advantage declines with age (see Figure 1 and Table A5). This pattern is consistent with the idea that immigrants are resilient to the effect of adverse life events on health at younger ages owing to health selectivity. However, over time and with increased exposure to structural disadvantages in the receiving society, immigrants become more vulnerable than nonimmigrants to the negative impacts of adverse life events on health.

Second, we documented that the migration status difference in declining health and well-being trajectories by age is exacerbated when an adverse life event occurs over the life course. This finding is a novel contribution to the literature on the mechanisms that produce migrant status differences in declining health with age. We provided evidence that having such detrimental experiences could accelerate declines in health by age among individuals in a disadvantaged position in society, as immigrants tend to be.

Third, we examined the short- and long-term trajectories of health after experiencing adverse life events by migration status and sex, highlighting that job loss is more detrimental to health for immigrants than for nonimmigrants. We found that the economic and social shocks of job loss differentially impacted nonimmigrants and immigrants. Further, it was not only the process of aging in good or poor health that was influenced by such events but also the trajectory of health by the time since the events. The accumulation of psychosocial and economic disadvantages during the life course was reflected in differences in the likelihood of aging in poor health and in the chances of experiencing short- and long-term changes in health.

Fourth, we showed that these mechanisms differed between men and women. The gap in the declining health trajectories by age was larger among women than among men, but the negative relationship between job loss and health was much stronger for immigrant men (relative to nonimmigrant men) than for immigrant women (relative to nonimmigrant women). We did not find a similar pattern for the adverse event of divorce. The literature has extensively shown that women age in poorer health than men despite surviving longer (Crimmins et al. 2011). Adding to that literature, we observed aging in poorer health when comparing women and men, as well as when including migration status in the equation.

In addition, we found that immigrant women were aging in poorer health than both men and nonimmigrant women. This result supports Hypothesis 1 and provides evidence of an intersectional impact of sex and migration status on declining health by age. One explanation for this pattern is that immigrant women are in a particularly disadvantaged position relative to immigrant men and nonimmigrant women (Donato et al. 2014). For example, immigrant women are, on average, poorer, overrepresented in the informal job market (De Jong and Madamba 2001; Donato et al. 2014), less educated, and more subject to the detrimental effects of traditional gender roles and social norms (Kanas and Müller 2021; Khoudja and Fleischmann 2017). Being more exposed to the detrimental effects of traditional gender roles might imply, for instance, that immigrant women are more likely than men or nonimmigrant women to face challenges in balancing family care (children and parents) with work.

Limitations

This study is not without limitations. First, we could not account for the immigrants’ country of birth because of the low sample size. Some of the observed patterns for immigrants might vary across countries of birth. The fixed-effects models partly accounted for that heterogeneity, given that country of birth is a time-invariant characteristic. However, there might still be compositional differences in each group of immigrants that could have been considered if the sample size had allowed us to run a stratified analysis. Moreover, there could be additional relevant time-varying explanatory factors that we did not account for. Second, although our sample restriction to individuals only up to age 64 implicitly limited the bias caused by return migration at retirement ages, we could not fully account for the potential bias created by out-migration in general. Third, we defined immigrants on the basis of country of birth, even though individuals born abroad but who had lived in Germany since they were very young (second-generation immigrants) might have more similar health outcomes to nonimmigrants. Fourth, individuals entered the observation period when they started participating in the survey, and the adverse events were calculated as the first observed change in employment and marital status between waves. Thus, we considered only those adverse events occurring after an individual started participating in the survey rather than all events over the individual's life course. However, we believe this condition produced a downward bias, and the observed patterns would have been further exacerbated if we had included more adverse life course events in the analyses. Fifth, the intersectional perspective would require including an additional interaction term with sex in the analyses. Instead, we used sex as a stratification factor. Including an interaction with the focal variable and sex is qualitatively very similar to stratifying the analysis and estimating the role of the focal variable within levels of sex. We acknowledge that the advantages of using an interaction include the ability to test its significance and estimate the modifier's impact on the outcome, neither of which is possible using stratification. However, because of low sample sizes and the related interpretability of our results, including an additional interaction term was not possible.

Conclusions

Given that our results suggest that multiple layers of disadvantage and adversities experienced over the life course interact with one another to exacerbate the health gaps between immigrants and nonimmigrants, the next question is what steps could be taken at the societal level to reduce these gaps. Our findings suggest that immigrants who experienced job loss in the receiving country, especially if they had previously experienced marriage dissolution, constitute a specific risk group who face worse consequences on a variety of health dimensions. The accumulation and the intersection of various forms of socioeconomic disadvantage during the life course have particularly harmful consequences on the health and well-being of individuals living in a foreign country, who have, on average, smaller extended social and family networks and are at a higher risk of experiencing discrimination and racialization. In addition to these aspects related to the societal sphere, immigrants are also more likely than nonimmigrants to have low wages, to be employed in less prestigious positions, and to be poor (Orrenius and Zavodny 2013). Experiencing job loss, divorce, or both events during the life course can have further detrimental consequences on immigrants’ health.

Improving immigrants’ health outcomes, especially for those at higher risk of living in conditions of social disadvantage and those who experience multiple adversities in the receiving country, is an important policy goal. The working-age population we studied will soon enter ages at which the risk of health frailties increases, and the intersectional role of adversities and the structural health disadvantages we observed in this population are likely to be exacerbated by aging. If receiving countries do not design and implement policies to address the social and economic disadvantages of immigrants, their poor health could become a public health issue.

Acknowledgments

Silvia Loi was supported by the European Union (ERC Starting Grant, MigHealthGaps, 101116721). Peng Li was supported by grants to the Max Planck–University of Helsinki Center from the Max Planck Society (Decision No. 5714240218), Jane and Aatos Erkko Foundation, Faculty of Social Sciences at the University of Helsinki, and Cities of Helsinki, Vantaa, and Espoo; and the European Union (ERC Synergy, BIOSFER, 101071773). Mikko Myrskylä was supported by the Strategic Research Council, FLUX consortium, Decision Nos. 345130 and 345131; by the National Institute on Aging (R01AG075208); by grants to the Max Planck–University of Helsinki Center from the Max Planck Society (Decision No. 5714240218), Jane and Aatos Erkko Foundation, Faculty of Social Sciences at the University of Helsinki, and Cities of Helsinki, Vantaa, and Espoo; and the European Union (ERC Synergy, BIOSFER, 101071773). The views and opinions expressed are, however, those of the authors only and do not necessarily reflect those of the European Union or the European Research Council. Neither the European Union nor the granting authority can be held responsible for them.

References

Allison, P. D. (
2009
).
Quantitative applications in the Social Sciences: Vol. 160. Fixed effects regression models
.
Thousand Oaks, CA
:
SAGE publications
.
Alviar, C. L., Rockman, C., Guo, Y., Adelman, M., & Berger, J. (
2014
).
Association of marital status with vascular disease in different arterial territories: A population based study of over 3.5 million subjects
.
Journal of the American College of Cardiology
,
63
(
Suppl. 12
), A1328. https://doi.org/10.1016/S0735-1097(14)61328-0
Atewologun, D., & Mahalingam, R. (
2018
).
Intersectionality as a methodological tool in qualitative equality, diversity and inclusion research
. In Booyson, L. A. E., Bendl, R., & Pringle, J. K. (Eds.),
Handbook of research methods in diversity management, equality and inclusion at work
(pp.
149
170
).
Cheltenham, UK
:
Edward Elgar Publishing
.
Bauer, G. R. (
2014
).
Incorporating intersectionality theory into population health research methodology: Challenges and the potential to advance health equity
.
Social Science & Medicine
,
110
,
10
17
.
Bauer, G. R., & Scheim, A. I. (
2019
).
Methods for analytic intercategorical intersectionality in quantitative research: Discrimination as a mediator of health inequalities
.
Social Science & Medicine
,
226
,
236
245
.
Berntsen, K. N., & Kravdal, Ø. (
2012
).
The relationship between mortality and time since divorce, widowhood or remarriage in Norway
.
Social Science & Medicine
,
75
,
2267
2274
.
Bloom, B. L., Asher, S. J., & White, S. W. (
1978
).
Marital disruption as a stressor: A review and analysis
.
Psychological Bulletin
,
85
,
867
894
.
Bowleg, L. (
2012
).
The problem with the phrase women and minorities: Intersectionality—An important theoretical framework for public health
.
American Journal of Public Health
,
102
,
1267
1273
.
Brand, J. E. (
2015
).
The far-reaching impact of job loss and unemployment
.
Annual Review of Sociology
,
41
,
359
375
.
Clark, A. E., Diener, E., Georgellis, Y., & Lucas, R. E. (
2008
).
Lags and leads in life satisfaction: A test of the baseline hypothesis
.
Economic Journal
,
118
,
F222
F243
.
Crimmins, E. M., Kim, J. K., & Solé-Auró, A. (
2011
).
Gender differences in health: Results from SHARE, ELSA and HRS
.
European Journal of Public Health
,
21
,
81
91
.
De Jong, G. F., & Madamba, A. B. (
2001
).
A double disadvantage? Minority group, immigrant status, and underemployment in the United States
.
Social Science Quarterly
,
82
,
117
130
.
de Vaus, D., Gray, M., Qu, L., & Stanton, D. (
2017
).
The economic consequences of divorce in six OECD countries
.
Australian Journal of Social Issues
,
52
,
180
199
.
Donato, K. M., Piya, B., & Jacobs, A. (
2014
).
The double disadvantage reconsidered: Gender, immigration, marital status, and global labor force participation in the 21st century
.
International Migration Review
,
48
(
Suppl. 1
),
335
376
.
Goldman, N., Pebley, A. R., Creighton, M. J., Teruel, G. M., Rubalcava, L. N., & Chung, C. (
2014
).
The consequences of migration to the United States for short-term changes in the health of Mexican immigrants
.
Demography
,
51
,
1159
1173
.
Graham, H. (
2007
).
Unequal lives: Health and socioeconomic inequalities
.
Berkshire, UK
:
Open University Press
.
Green, M. A., Evans, C. R., & Subramanian, S. V. (
2017
).
Can intersectionality theory enrich population health research?
Social Science & Medicine
,
178
,
214
216
.
Gubernskaya, Z. (
2015
).
Age at migration and self-rated health trajectories after age 50: Understanding the older immigrant health paradox
.
Journals of Gerontology, Series B: Psychological Sciences and Social Sciences
,
70
,
279
290
.
Gurak, D. T., & Dean, G. (
1980
).
The remarriage market: Factors influencing the selection of second husbands
.
Journal of Divorce
,
3
(
2
),
161
173
.
Hankivsky, O. (
2012
).
Women's health, men's health, and gender and health: Implications of intersectionality
.
Social Science & Medicine
,
74
,
1712
1720
.
Harari, L., & Lee, C. (
2021
).
Intersectionality in quantitative health disparities research: A systematic review of challenges and limitations in empirical studies
.
Social Science & Medicine
,
277
,
113876
. https://doi.org/10.1016/j.socscimed.2021.113876
Kanas, A., & Müller, K. (
2021
).
Immigrant women's economic outcomes in Europe: The importance of religion and traditional gender roles
.
International Migration Review
,
55
,
1231
1264
.
Khoudja, Y., & Fleischmann, F. (
2017
).
Labor force participation of immigrant women in the Netherlands: Do traditional partners hold them back?
International Migration Review
,
51
,
506
541
.
Lechner, I., & Mielck, A. (
1998
).
Die verkleinerung des "healthy-migrant-effects": Entwicklung der morbidität von ausländischen und Deutschen befragten im Sozioökonomischen Panel 1984–1992 [Decrease in the “healthy migrant effect”: Trends in the morbidity of foreign and German participants in the 1984–1992 Socio-Economic Panel]
.
Gesundheitswesen
,
60
,
715
720
.
Leopold, L., Leopold, T., & Lechner, C. M. (
2017
).
Do immigrants suffer more from job loss? Unemployment and subjective well-being in Germany
.
Demography
,
54
,
231
257
.
Leopold, T. (
2018
).
Gender differences in the consequences of divorce: A study of multiple outcomes
.
Demography
,
55
,
769
797
.
Loi, S., & Hale, J. M. (
2019
).
Migrant health convergence and the role of material deprivation
.
Demographic Research
,
40
,
933
962
. https://doi.org/10.4054/DemRes.2019.40.32
Lucas, R. E. (
2005
).
Time does not heal all wounds: A longitudinal study of reaction and adaptation to divorce
.
Psychological Science
,
16
,
945
950
.
Lucas, R. E., Clark, A. E., Georgellis, Y., & Diener, E. (
2004
).
Unemployment alters the set point for life satisfaction
.
Psychological Science
,
15
,
8
13
.
Luhmann, M., Hofmann, W., Eid, M., & Lucas, R. E. (
2012
).
Subjective well-being and adaptation to life events: A meta-analysis
.
Journal of Personality and Social Psychology
,
102
,
592
615
.
Mandelbaum, J. (
2020
).
Advancing health equity by integrating intersectionality into epidemiological research: Applications and challenges
.
Journal of Epidemiology and Community Health
,
74
,
761
762
.
Nielsen, N. M., Davidsen, R. B., Hviid, A., & Wohlfahrt, J. (
2014
).
Divorce and risk of hospital-diagnosed infectious diseases
.
Scandinavian Journal of Public Health
,
42
,
705
711
.
Orrenius, P. M., & Zavodny, M. (
2013
).
Immigrants in risky occupations
. In Constant, A. F. & Zimmermann, K. F. (Eds.),
International handbook on the economics of migration
(pp.
214
226
).
Cheltenham, UK
:
Edward Elgar Publishing
.
Palloni, A., & Arias, E. (
2004
).
Paradox lost: Explaining the Hispanic adult mortality advantage
.
Demography
,
41
,
385
415
.
Paul, K. I., Hassel, A., & Moser, K. (
2018
).
Individual consequences of job loss and unemployment
. In Klehe, U.-C. & van Hooft, E. A. J. (Eds.),
Oxford handbook of job loss and job search
(pp.
57
85
).
New York, NY
:
Oxford University Press
.
Paul, K. I., & Moser, K. (
2009
).
Unemployment impairs mental health: Meta-analyses
.
Journal of Vocational Behavior
,
74
,
264
282
.
Ronellenfitsch, U., & Razum, O. (
2004
).
Deteriorating health satisfaction among immigrants from Eastern Europe to Germany
.
International Journal for Equity in Health
,
3
,
4
. https://doi.org/10.1186/1475-9276-3-4
Sbarra, D. A. (
2015
).
Divorce and health: Current trends and future directions
.
Psychosomatic Medicine
,
77
,
227
236
.
Sbarra, D. A., & Coan, J. A. (
2017
).
Divorce and health: Good data in need of better theory
.
Current Opinion in Psychology
,
13
,
91
95
.
Sbarra, D. A., Law, R. W., & Portley, R. M. (
2011
).
Divorce and death: A meta-analysis and research agenda for clinical, social, and health psychology
.
Perspectives on Psychological Science
,
6
,
454
474
.
Schaller, J., & Stevens, A. H. (
2015
).
Short-run effects of job loss on health conditions, health insurance, and health care utilization
.
Journal of Health Economics
,
43
,
190
203
.
Shor, E., Roelfs, D. J., Bugyi, P., & Schwartz, J. E. (
2012
).
Meta-analysis of marital dissolution and mortality: Reevaluating the intersection of gender and age
.
Social Science & Medicine
,
75
,
46
59
.
Smock, P. J., Manning, W. D., & Gupta, S. (
1999
).
The effect of marriage and divorce on women's economic well-being
.
American Sociological Review
,
64
,
794
812
.
Trappe, H., Pollmann-Schult, M., & Schmitt, C. (
2015
).
The rise and decline of the male breadwinner model: Institutional underpinnings and future expectations
.
European Sociological Review
,
31
,
230
242
.
Turra, C. M., & Elo, I. T. (
2008
).
The impact of salmon bias on the Hispanic mortality advantage: New evidence from Social Security data
.
Population Research and Policy Review
,
27
,
515
530
.
Viruell-Fuentes, E. A., Miranda, P. Y., & Abdulrahim, S. (
2012
).
More than culture: Structural racism, intersectionality theory, and immigrant health
.
Social Science & Medicine
,
75
,
2099
2106
.

Supplementary data