Can Differing Occupational Class Positions Explain Migrant Health Inequalities? Differences in Trajectories of Subjective Health Between Migrants and Native Germans over Time

Holz, Manuel; Mayerl, Jochen Article — Published Version Can Differing Occupational Class Positions Explain Migrant Health Inequalities? Differences in Trajectories of Subjective Health Between Migrants and Native Germans over Time KZfSS Kölner Zeitschrift für Soziologie und Sozialpsychologie Provided in Cooperation with: Springer Nature Suggested Citation: Holz, Manuel; Mayerl, Jochen (2025) : Can Differing Occupational Class Positions Explain Migrant Health Inequalities? Differences in Trajectories of Subjective Health Between Migrants and Native Germans over Time, KZfSS Kölner Zeitschrift für Soziologie und Sozialpsychologie, ISSN 1861-891X, Springer Fachmedien Wiesbaden GmbH, Wiesbaden, Vol. 77, Iss. 1, pp. 27-52, https://doi.org/10.1007/s11577-025-00985-3 This Version is available at: https://hdl.handle.net/10419/323561 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. http://creativecommons.org/licenses/by/4.0/deed.de ABHANDLUNG https://doi.org/10.1007/s11577-025-00985-3 Köln Z Soziol (2025) 77:27–52 Can Differing Occupational Class Positions Explain Migrant Health Inequalities? Differences in Trajectories of Subjective Health Between Migrants and Native Germans over Time Manuel Holz · Jochen Mayerl Received: 4 December 2024 / Accepted: 21 January 2025 / Published online: 11 March 2025 © The Author(s) 2025 Abstract Migrants living in postindustrial countries are confronted with various socioeconomic challenges, including lower incomes, extended working hours, and lower occupational statuses than natives. Although health disparities linked to occupational positions have frequently been documented, they remain a relatively unexplored factor in the explanation of health gaps over time between migrants and native populations. To address this issue, we utilized longitudinal data spanning from 2002 to 2018 from the German Socio-Economic Panel to investigate disparities in physical health–related quality of life across different migrant and native German cohorts and their associations with occupational class position. Our findings reveal that overall lower occupational class positions can account for the health disparities observed between migrants and native Germans. Further, our study unveils complex relationships between initial health conditions (intercepts), changes in health over time (slopes), region of origin (European migrants, non-European migrants, and native Germans), and gender. These nuanced outcomes underscore the importance of adopting approaches that consider both region of origin and gender when seeking to enhance working conditions and facilitate access to the labor market for diverse populations. Keywords Health trajectory · Migration · Inequality · Occupational position · Latent growth · Curve analysis M. Holz · J. Mayerl Institute for Sociology, Faculty of Behavioral and Social Sciences, Chemnitz University of Technology Thüringer Weg 9, 09126 Chemnitz, Germany E-Mail: [email protected] J. Mayerl E-Mail: [email protected] K 28 M. Holz, J. Mayerl Können unterschiedliche berufliche Klassenpositionen die gesundheitlichen Ungleichheiten von Migranten erklären? Unterschiede in subjektiven Gesundheitsverläufen zwischen Migranten und Deutschen ohne Migrationshintergrund im Längsschnitt Zusammenfassung Migranten, die in postindustriellen Ländern leben, sehen sich mit verschiedenen sozioökonomischen Herausforderungen konfrontiert. Darunter fallen niedrige Einkommen, lange Arbeitszeiten und niedrige berufliche Positionen. Während gesundheitliche Ungleichheiten in Verbindung mit beruflichen Positionen häufig dokumentiert wurden, bleiben sie ein relativ unerforschter Faktor bei der Erklärung von längsschnittlichen Gesundheitsunterschieden zwischen Migranten und einheimischen Bevölkerungen. Um dieses Problem anzugehen, wurden Längsschnittdaten aus dem Zeitraum von 2002 bis 2018 aus dem Sozio-oekonomischen Panel (SOEP) verwendet. Die Unterschiede in der gesundheitsbezogenen Lebensqualität zwischen verschiedenen Migranten-Kohorten und deutschen Kohorten ohne Migrationshintergrund werden im Zusammenhang mit der beruflichen Klassenzugehörigkeit untersucht. Unsere Ergebnisse zeigen, dass insgesamt niedrigere berufliche Klassenzugehörigkeiten die beobachteten gesundheitlichen Unterschiede zwischen Migranten und einheimischen Deutschen erklären können. Darüber hinaus zeigt unsere Studie komplexe Beziehungen zwischen anfänglichen Gesundheitszuständen (Intercepts) und Veränderungen der Gesundheit im Zeitverlauf (Slopes) in Bezug auf die Herkunftsregion (europäische Migranten, nicht europäische Migranten und Deutsche ohne Migrationshintergrund) und das Geschlecht. Die Ergebnisse unterstreichen die Bedeutung integrativer Ansätze, die sowohl die Herkunftsregion als auch das Geschlecht berücksichtigen, wenn es darum geht, die Arbeitsbedingungen zu verbessern und den Zugang zum Arbeitsmarkt für diverse Bevölkerungsgruppen zu erleichtern. Schlüsselwörter Gesundheitsverläufe · Migration · Ungleichheit · Berufliche Position · Latente Wachstumskurven 1 Introduction Migrant populations often enter host countries with a health advantage compared to the native population, a phenomenon commonly known as the “healthy migrant effect” (HME; McDonald and Kennedy 2004). However, over time, this advantage diminishes, leading to a transition into a state of health disadvantage or convergence with native health levels. Whereas there is more consensus and tested explanations for the first part of the HME, tested interpretations and descriptions for the second part are scarce (Rellstab et al. 2016;Yang2021). To enrich our understanding of health trajectories following migration, we draw on aspects from life-course perspectives on socioeconomic health disparities (Engelhardt-Woelfler and Leopold 2020). Within this comprehensive framework, three distinct scenarios of health trajectories may emerge: divergence, convergence, and continuity (Leopold and Engelhardt 2011;Prus2007). In our case, divergence occurs K Can Differing Occupational Class Positions Explain Migrant Health Inequalities? Differences... 29 when health disparities between migrants and natives increase over time, suggesting an increase in migrant health disadvantages or a decline in migrant health advantages compared to native populations. Convergence, on the other hand, implies a leveling off of health differentials, with migrant health gradually aligning with that of native populations. Continuity indicates that health disparities persist over time, with migrant health remaining consistently above or below native levels without substantial change. In our study, our primary objective is to explain these potentially differing trajectories of postmigration migrant health. Although poverty is often considered a “risk factor” for health disparities, we argue that it cannot be fully understood without considering its relationship to social class, which is closely tied to the work environment (Scambler and Higgs 1999). The working conditions of migrants have emerged as a critical yet underexplored mechanism for longitudinal health inequalities that warrants closer examination. Concurrently, migrants face various challenges in the host country’s labor market, including lower employment rates and occupational status, underemployment, and reduced earnings (Berwing 2019; Damelang et al. 2021; Kalter and Granato 2018; Stanek and Ramos 2013). The working conditions in labor market positions occupied mainly by migrants have severe detrimental effects on migrant health. Migrant workers face disproportionate occupational health hazards such as heavy lifting; vibrations from machinery; long periods of standing; exposure to extreme temperatures, noise, and chemical toxins; the need to work at a high pace; employment arrangements without contracts or with limited control over free days; and psychosocial strain (Oldenburg et al. 2010; Ronda Pérez et al. 2012; Simmons et al. 2022). Hence, we hypothesize that the deterioration in migrant health occurs at a more accelerated rate compared to that of native populations, and we attribute this variance to the comparatively lower occupational class positions held by migrants. This study contributes to the existing body of literature by examining whether occupational class position can explain migrant health disparities over time, and therefore it can deliver an empirically tested explanation for the health decline observed in the HME phenomenon.1By utilizing data from the German Socio-Economic Panel (SOEP) and applying latent growth curve analysis within a structural equation modeling framework, this study establishes occupational class position, using the Erikson–Goldthorpe–Portocarero (EGP) classification of social classes (Erikson and Goldthorpe 1993), as a mediator between migration background and physical health outcomes. Moreover, we differentiate between different regions of origin (European vs. non-European migrants) and conduct the analysis separately by gender to gain more nuanced insights regarding the relationship between migration background, occupational class, and health over the course of time. 1Hence, our emphasis is not on recent migrants (those with a short duration since migration) but rather on migrants as a whole, as our primary interest lies in understanding the health-decline aspect of the HME. K 30 M. Holz, J. Mayerl Fig. 1 Health trajectories in the context of the healthy migrant effect 2 Conceptual Background The context of this analysis is the so-called HME (McDonald and Kennedy 2004). In this well-documented phenomenon, it is observed that migrants generally have better health upon arrival in the host country compared to the native population. However, this advantage tends to diminish over time, eventually leading to a state of health disadvantage. The primary element of this phenomenon, which is the initial health advantage, has received considerable attention and delivered sufficiently tested interpretations. These explanations primarily revolve around healthy self-selection prior to migration (Fuller-Thomson et al. 2016), selective remigration (Di Napoli et al. 2021), and culture-specific health-protective behaviors (Lee et al. 2013). For the second part of the HME phenomenon, the disproportionate health decline, explanations and evidence are either rare or inconclusive (Rellstab et al. 2016;Yang 2021). Our focus in this study is therefore the subsequent part of the HME. The convincing, yet untested, reasons for postmigration health trajectories being either similar or worse than the native health trajectories have been attributed mainly to acculturation and the adoption of unhealthy host country–specific lifestyles (Edberg et al. 2011; Fujishiro and Hoppe 2020; Namer and Razum 2018). Among the many factors contributing to migrant health disparities, the working conditions experienced by migrants are a critical and underexplored determinant accounting for migrant health decline (Corna 2013). Notably, the employment context is a very well-researched factor in explaining diverging health outcomes (Lahelma et al. 2005; Hiesinger and Tophoven 2019; Landsbergis 2010; Gonzalez-Mulé and Cockburn 2017; Geyer and Peter 1999). Given the long latency periods between the socioeconomic conditions describing the migrant–native differences in long-term health trajectories, such as the occupational context (Clumeck et al. 2009; Johnson et al. 1996), life-course perspectives in socioeconomic health disparities provide a useful framework (Corna 2013; Edberg et al. 2011; Fujishiro and Hoppe 2020; Leong et al. 2014). There are typically three scenarios, or hypotheses, concerning the development of health differences between social groups over time: convergence, divergence, or continuity/status maintenance (Engelhardt-Woelfler and Leopold 2020; Leopold and Engelhardt 2011). These trajectory types were initially formulated for disparities in health due to different socioeconomic positions. In this context, we apply this K Can Differing Occupational Class Positions Explain Migrant Health Inequalities? Differences... 31 concept to migrant health disparities. Considering the healthy migrant phenomenon, all scenarios start with a migrant health advantage (Fig. 1). However, in scenario a, the divergence trajectory, the health of migrants deteriorates constantly at a higher rate than the health of nonmigrants (or “natives”), resulting in an increasing difference in the trajectories over time. In scenario bof convergence, the rate of health deterioration is initially higher for migrants than for nonmigrants, but the health trajectories approximate each other and become statistically indistinguishable over the course of time. In the continuity scenario c, both groups experience a similar rate of health deterioration over time, yet a persisting health advantage exists for one group, and the trajectories stay parallel in the long run; i.e., health inequalities stay within a constant range and neither grow nor diminish when the number of observation periods increases. The divergence trajectory is primarily explained by the concept of cumulative advantage/disadvantage. Due to their position in the socioeconomic strata, individuals experience a path-dependent accumulation of dis/advantages. Prior states or positions amplify subsequent conditions, and the payoffs from advantages or consequences from disadvantages increase over time (Engelhardt-Woelfler and Leopold 2020). In other words, higher socioeconomic positions can translate into less hazardous working conditions, leading to reduced periods of harmful exposure over time (Prus 2007). The disadvantages in the case of migrant workers are represented as these workers being disproportionately more often exposed to occupational health hazards, as outlined in the introduction (Oldenburg et al. 2010; Ronda Pérez et al. 2012; Simmons et al. 2022). Therefore, differences in working conditions translate into growing health disparities between social groups, i.e., between migrants and the native population. The convergence trajectory is often explained by the “age-as-leveler” hypothesis (Corna 2013; Engelhardt-Woelfler and Leopold 2020). This perspective posits that initial health inequalities are “buffered” by biological, social class–independent health deterioration in later stages of life. For the migrant–native health differences, it can be argued that in addition to the biological “buffering,” cultural factors can be at play, which lead to convergence. A potential explanation can be derived by elements from cumulative inequality theory (Ferraro et al. 2009). This perspective highlights that individuals can be exposed to both health-promoting and healthdeteriorating factors at the same time. In this case, hazardous working conditions can indeed increase the rate of health decline, but culture-specific factors such as better diet (Darmon and Khlat 2001), closer family relations (Arends-Tóth and van de Vijver 2008), and religiosity (Aukst-Margeti´ candMargeti ´ c2005) can mitigate the negative effects of low socioeconomic position on health, eventually leading to similar health levels. Another reason for converging health trajectories stems from methodological issues, namely selective panel mortality. Due to health participation bias in surveys, unhealthier individuals exit the panel (or die) at a disproportionate rate. As a result, the survey population in later years becomes more similar (Shaw and Krause 2002). According to the continuity perspective, health disparities remain relatively stable over time. This perspective suggests that once health disparities are established due to socioeconomic status, they persist throughout the life course, with the migrant and K 32 M. Holz, J. Mayerl native lines remaining parallel (Engelhardt-Woelfler and Leopold 2020). Building similarly on cumulative inequality theory, cultural factors can buffer the effects of working conditions on health decline, although in this scenario these factors cannot cancel out this effect entirely. Regarding the evidence of the possible health trajectories, there is some evidence for convergence (Beckett 2000), yet divergence seems to be the most common pattern of life-course socioeconomic health disparities, indicating that the cumulative advantage/disadvantage perspective is the most probable one (Leopold and Engelhardt 2011; Chandola et al. 2007; Shaw and Krause 2002;Prus2007; Sacker et al. 2011). Yet in these studies, the concept of socioeconomic position used is mostly in the form of education. Job context or exposure has not been applied as a moderator between age/time and health. One consistent finding across studies regarding health trajectories of migrants and ethnic minorities (Edberg et al. 2011; Jatrana et al. 2018; Shuey and Willson 2008) is the existence of a health disadvantage in the form of faster health decline of these groups compared to the native majority population. Another aspect that emerges is the heterogeneity of migrant health inequality. In the work of Edberg et al. (Edberg et al. 2011), migrants from countries characterized by higher economic inequality tend to exhibit a more pronounced postmigration health deterioration than migrants from countries with lower economic inequality. Similarly, in a study on health trajectories of migrants in Australia, Jatrana et al. (2018) found that migrants from English-speaking countries showed persisting health advantages in the life course compared to natives, while migrants from non–English-speaking countries showed health disadvantages that increased over time. It is therefore crucial to differentiate between different groups regarding health trajectories. Further, the mitigating effect of socioeconomic position on health decline does not show a clear trend regarding minority status. In a study by Assari (2017), educational attainment buffered health deterioration only for Black men, and Shuey and Willson (2008) found that belonging to an ethnic minority group is associated with faster deterioration in health compared to the majority population. However, socioeconomic position serves as a buffer against this decline, although the buffering effect appears to be more pronounced among the white majority population. 2.1 Occupational Class as a Measure of Working Conditions Bringing the aforementioned topics all together into one research program is challenging in terms of data availability. In order to increase sample sizes and observation periods, we considered data from the German SOEP to be the most appropriate for our purpose. The existing job exposure matrices, based on International Standard Classification of Occupations (ISCO) codes, as given by Kroll (Kroll 2011), deliver a relatively direct measure of job exposure per occupation (job exposure in this sense is the exposure to either physical strain, such as heavy lifting or carcinogenic agents, or psychosocial strain). Yet the number of missing values created by matching this exposure measure with respondent occupational information (Brünger et al. 2020) hinders a longitudinal analysis with a high number of observation periods and differentiation between migrant and gender groups. We propose that another ISCO-based K Can Differing Occupational Class Positions Explain Migrant Health Inequalities? Differences... 33 measure, namely occupational class, as outlined in the EGP classification of social classes (Erikson and Goldthorpe 1993) can effectively serve as a proxy for assessing job exposure, while exhausting larger parts of the dataset. Theoretically, within the framework of power dynamics, individuals in higher social class positions, as given in the EGP scheme, tend to exert greater control over their working conditions and tasks. Highly educated professionals, being equipped with specialized knowledge, can avoid and mitigate instances of exploitation (McCartney et al. 2019). Furthermore, there exists a presumed gradient along the dimensions of social class concerning job security and levels of underand unemployment that subsequently influence cultural capital, risk behavior, and overall health outcomes (Scambler and Higgs 1999). Although the empirical evidence does not strictly adhere to a linear relationship between the EGP scheme and job exposure, it is observed that average job exposure tends to increase from EGP class I to VII. This is particularly notable for exposure to hazardous physical conditions, limited job control, and inadequate social support (Krokstad et al. 2002; Schrijvers et al. 1998). Similar associations have been demonstrated in German data, where heavy lifting, exposure to vibrations, and contact with chemical toxins were found to be more prevalent in higher EGP class values (and thus lower social class positions) compared to lower EGP class values (Dragano et al. 2016). Alternative ISCO-based metrics, such as the International Socio-Economic Index of Occupational Status (ISEI) and the Standard International Occupational Prestige Scale (SIOPS), tend to emphasize financial factors or intangible rewards such as prestige. However, we contend that the relational dimensions of the EGP better encapsulate work exposure than ISEI and SIOPS do. To ensure the robustness of our findings, we recalculated our models using ISEI and SIOPS as primary mediating variables (see Online Appendix). 3 Expectations Overall, we expected occupational class to act as a mediator for the effect of migration background on health. Considering the possible trajectories (as in Fig. 1) and the evidence, we expected divergence to be the dominant pattern of health trajectories between migrants and natives (H1). This divergence can be explained by differing occupational class positions; i.e., the difference in health trajectories between migrants and native Germans either vanishes or decreases significantly when occupational class position is controlled for (H2). Further, migrants are expected to find themselves more frequently in lower occupational class positions (H3). Because migrants exhibit diverse backgrounds in terms of selection, culture, and skills, the impact on their health trajectories can vary significantly. European migrants in Germany, for instance, often come from countries with high proficiency in German as a second language (Auswärtiges Amt [Federal Foreign Office] 2015), and their educational qualifications tend to hold more value in the German job market compared to migrants from non-European countries (Kogan 2011). Conversely, nonEuropean migrants tend to display certain health-promoting behaviors such as lower rates of alcohol abuse disorders (Cook et al. 2021), higher vegetable consumption (Kleiser et al. 2010), and greater religiosity (van Tubergen and Sindradóttir 2011), K 34 M. Holz, J. Mayerl but they are also more prone to have experienced trauma (Nesterko et al. 2019) compared to their European counterparts. We therefore investigated the distinction between European and non-European migrants in an exploratory manner. Regarding gender disparities in the workplace, women frequently encounter harassment(Baderetal.2018), are more likely to have temporary contracts with low wages (Campos-Serna et al. 2013), and face additional stressors such as role overload (Pearson 2008). Conversely, men are more commonly exposed to hazardous working conditions leading to workplace accidents (Rommel et al. 2016)andare less inclined to adopt health-protective behaviors (Hiller et al. 2017). Given the varying stressors and risks, predicting the physical health trajectories for each gender becomes challenging. Moreover, considering that identical job titles may involve different tasks and levels of exposure based on gender (Messing et al. 1994), we conducted separate analyses for women and men, as recommended in the literature on occupational health (Messing et al. 2003). 4Data We used individual data from the German SOEP, a longitudinal study with random sampling conducted by the German Institute for Economic Research (DIW) since 1984. The SOEP involves over 12,000 private households in Germany and is collected annually. This dataset is particularly suitable for our purposes—first, due to its large sample size, and second, because it includes an oversampling of migrant respondents from (South)Eastern Europe and Southwest Asia (Brücker et al. 2014). The respondents in this dataset are aged 17 years and older. The health variables, collected as repeated measures, are available biannually. In order to increase sample sizes while at the same time including as many time points as possible, we analyzed the waves 2002–2018 (nine time points). We restricted respondent age to a maximum of 49 years in the initial wave to exclude cases of respondents who potentially drop out of the sample due to retirement. The final sample consists of n= 13,049 respondents (11,386 native Germans, 1082 European migrants, 581 non-European migrants). For the main outcome variables, we used the physical health scale of the 12Item Short-Form Health-Related Quality of Life Questionnaire (Ware et al. 1996). Physical health is measured as general health status, limitations in climbing up stairs and conducting everyday activities, the presence of severe physical pain in the previous 4 weeks, limitations in achievement due to physical health, and general limitations due to physical health (see Online Appendix for exact wording and scales). We classify migrants as individuals who were not born in the Federal Republic of Germany. Native Germans are categorized as respondents whose birth and the birth of both of their parents occurred in Germany. This analysis does not take into account individuals with an indirect migration background, in which only one K Can Differing Occupational Class Positions Explain Migrant Health Inequalities? Differences... 41 Table 4 Parameter estimates from conditional latent growth models for latent slopes of physical health Dependent variable= latent slope of the latent factor physical health Women Men Model 1 Model 2 Model 3 Model 1 Model 2 Model 3 Variable European (reference: native German) –0.026** (0.009) –0.078 –0.020* (0.009) –0.058 –0.019* (0.009) –0.057 –0.007 (0.010) –0.021 –0.000 (0.010) –0.001 0.003 (0.010) 0.007 Non-European (reference: native German) –0.021 (0.014) –0.040 –0.010 (0.014) –0.020 –0.010 (0.009) –0.019 –0.034* (0.016) –0.064 –0.024 (0.016) –0.046 –0.027 (0.016) 0.007 Occupational class – 0.013*** (0.002) 0.135 0.009*** (0.003) 0.092 –0.010*** (0.002) 0.145 0.007*** (0.002) 0.106 Education – – 0.004** (0.001) 0.087 ––0.004* (0.002) 0.085 Age – – –0.123*** (0.021) –0.155 –––0.147*** (0.023) –0.188 Mean latent slope –0.047*** (0.002) –0.077*** (0.006) –0.066*** (0.006) –0.051*** (0.002) –0.073*** (0.004) –0.065*** (0.005) Coefficients are represented as b (standard error) b*, with bunstandardized coefficient, b* standardized coefficient ***p≤0.001, **p≤0.01, *p≤0.05, (quasi-)metric variables mean centered K 42 M. Holz, J. Mayerl Germans, while the pattern for non-European migrants reflects more continuity. With the utilization of a singular physical health indicator and the absence of growth process modeling, there is no indication that the prevailing pattern is divergence. Yet there are differences in the trajectories that require further analysis. 6.2 Multivariate From Table 2we can inspect model-fit measures of each regression model. In the unconditional growth model, Model 0, the CFI is at 0.959, the RMSEA at 0.028 and the SRMR at 0.067. These values are in line with the thresholds in the literature and indicate an adequate model fit (Kline 2023; Marsh et al. 2009). We therefore conclude that the linear growth pattern is applicable to the latent variable physical health, i.e., restricting factor loadings of the latent slope parameter to the values as reported in the methods section reflects the data structure sufficiently. Subsequent models similarly show an adequate data fit. Figure 43displays the health trajectories derived from the growth curve models. Compared to the univariate case depicted in Fig. 3, the pattern for women is clearer, resembling a divergence trajectory. Additionally, there is no discernible contrast between European and non-European women regarding their physical health trajectory, as both groups experience a steeper decline compared to native German women. Conversely, among men, the pattern is more varied. The disparity between European men and native German men aligns more with a continuity pattern, while non-European men exhibit a clear divergence scenario. Notably, for men, the latent intercepts, or starting points, are consistently higher than those for women. Overall, women consistently have lower health trajectories than men, with migrant women having the lowest trajectory. In the unconditional growth model, Model 0, the variances of latent slopes (women: 0.005 [standard error, SE,= 0.00], men: 0.005 [SE= 0.00]) and intercepts (women: 0.338 [SE= 0.008], men: 0.296 [SE= 0.008]) significantly differ from 0 (p≤0.001). This suggests that respondents exhibit substantial deviation in both the means of latent intercepts (women: 4.149 [SE= 0.009], men: 4.266 [SE= 0.009]) and means of latent slopes (women: –0.049 [SE= 0.002], men: –0.052 [SE= 0.002]), indicating the presence of variance that requires explanation. In the multivariate analysis with the latent intercept of physical health as dependent variable (Table 3), we find a significant (p≤0.05) negative effect for nonEuropean women (b= –0.086, SE= 0.040, b*= –0.032) compared to native German women (Model 1). This effect disappears (p> 0.05) as soon as occupational class is introduced to the model (Model 2) and stays this way after controlling for further covariates (Model 3). No significant difference (b= –0.102 SE= 0.060, b*= –0.025) can be found in the latent intercept between non-European women and native German women (Model 1). This result stays stable after controlling for occupational class (Model 2) and covariates (Model 3). Put differently, when we examine the beginning of the physical health trajectory, it does not match the depiction in Fig. 1. We do not see an initial health benefit 3The estimates for this plot can be inquired in Tables 3and 4. K Can Differing Occupational Class Positions Explain Migrant Health Inequalities? Differences... 43 Table 5 Parameter estimates from conditional latent growth models for latent intercepts of occupational class Dependent variable= latent intercept of occupational class Women Men Model 2 Model 3 Model 2 Model 3 Variable European (reference: native German) –0.585*** (0.049) –0.162 –0.577*** (0.044) –0.160 –0.725*** (0.072) –0.136 –0.636*** (0.059) –0.119 Non-European (reference: native German) –0.805*** (0.086) –0.148 –0.672*** (0.078) –0.123 –0.867*** (0.108) –0.110 –0.720*** (0.089) –0.091 Education – 0.250*** (0.007) 0.483 –0.449*** (0.009) 0.631 Age – –1.486*** (0.112) –0.174 ––1.330*** (0.150) –0.113 Coefficients are represented as b (standard error) b*, with bunstandardized coefficient, b* standardized coefficient ***p≤0.001, **p≤0.01, *p≤0.05, (quasi)metric variables mean centered for migrant women compared to native German women. Instead, we notice a health disparity at the outset between European women and native German women. This could be attributed to the inclusion of all migrants rather than just recent ones. In terms of men, both European (b= –0.116, SE= 0.041, b*= –0.044) and nonEuropean (b= –0.153, SE= 0.061, b*= –0.039) migration backgrounds show significant negative effects on the latent intercept of physical health. This suggests that migrant men have lower initial starting points in their physical health trajectories. However, both effects vanish upon controlling for occupational class (p> 0.05) in Model 2 and Model 3. This suggests that the initial health disparities observed in migrant men can be attributed to their occupational class. With the slope parameter of physical health as dependent variable (Table 4), the following observations can be made: Without covariates (Model 1), European migrant women (b= –0.026, SE= 0.009, b*= –0.078) have, on average, a lower latent slope of physical health than native German women. The significance of the effect disappears after controlling for occupational class (Model 2) and further covariates (Model 3). For non-European women there is no significant difference in the latent slope of physical health in any model; i.e., native German women and non-European women experience a similar rate of health deterioration over time. In the analysis of men, there is no latent slope difference in any model for European men compared to native German men. Yet non-European men have, on average, a lower latent slope in physical health than German men (b= –0.034, SE= 0.016, b*= –0.064) in the first model, but when occupational class is controlled for, this effect becomes insignificant (Model 2 and Model 3). The rate of health deterioration is therefore similar for European men and native German men, whereas the health of non-Europeans declines at an accelerated rate. K 44 M. Holz, J. Mayerl In Table 5, the estimates for latent intercept of the EGP class scheme as a dependent variable are shown. In both gender groups, European (women: b= –0.585, SE= 0.049, b*– 0.162; men: b= –0.725, SE= 0.072, b*= –0.136) and non-European migrants (women: b= –0.805, SE= 0.086, b*= –0.148, men: b= –0.867, SE= 0.108, b*= –0.110) have lower levels in the latent intercept of occupational class (Model 1). This effect is stable after controlling for covariates (Model 2). 7 Discussion and Conclusion In this study, we examined health disparities between migrants and native Germans and assessed the role of occupational class in explaining these disparities. We used longitudinal data from a German household panel study spanning from 2002 to 2018 and employed latent growth curve analysis to investigate how migration background influences the trajectory of physical health when occupational class is introduced as a mediator. We intended to describe health trajectories of European migrants, nonEuropean migrants, and native Germans, while accounting for differences between women and men. On a substantial level, this study delivers a tested explanation for postmigration health decline in the context of the HME phenomenon, which has been absent in prior research (Holz 2022). We hypothesized that divergence would emerge as the primary pattern of health disparities between migrants and native Germans over time. Divergence, characterized by widening health gaps, was confirmed in our findings, aligning with existing literature in which it is the predominant pattern in socioeconomic health gaps (Chandola et al. 2007; Leopold and Engelhardt 2011;Prus2007; Sacker et al. 2011;Shaw and Krause 2002). Our study extends this understanding by demonstrating that migrants have a higher propensity to experience the cumulative disadvantage of lower socioeconomic positions. However, this assertion holds true only to some extent. Although such an effect is indeed evident among women, the situation among men is more intricate. Notably, the diverging trend is comparable for both European and non-European women. It is plausible that factors such as discrimination (Scholaske et al. 2019), role overload (Pearson 2008), or workplace harassment (Bader et al. 2018) could contribute to the exacerbated decline in health. Further investigation is warranted to understand how these factors interact with precarious working conditions, particularly among European women, since occupational class does not entirely explain their differing health trajectories. European women are potentially less selected for health due to lower migration costs (Borjas 1987) and might therefore exhibit a greater variance of health levels, increasing the likelihood of observing lower health statuses. Moreover, compared to non-Europeans, Europeans tend to have less favorable dietary habits (Cook et al. 2021; Kleiser et al. 2010), which could also influence these findings. The path for European men appears to follow a trend of continuity (EngelhardtWoelfler and Leopold 2020). Although a health gap persists between European men and native Germans, it remains steady over time, indicating a variation in initial health statuses rather than changes in health over time. This contrast diminishes when occupational status is considered, suggesting that occupational class may account K Can Differing Occupational Class Positions Explain Migrant Health Inequalities? Differences... 45 for this continuity. It is conceivable that lower levels of workplace harassment and fewer caregiving responsibilities (Nutz et al. 2023) alleviate the strain associated with lower occupational positions, thereby sustaining the observed stability. On the contrary, the trajectory for non-European men exhibits divergence, as anticipated. Over time, the challenges encountered by non-European migrants accumulate, resulting in an increasing health gap. This phenomenon appears to be primarily driven by occupational class, which is unsurprising given that non-European migrant men are predominantly employed in the lowest occupational strata. It is plausible that heightened discrimination (Hönekopp et al. 2002)andtrauma (Nesterko et al. 2019) add to the effect of lower occupational status on health deterioration. More research is needed to explain and understand the interplay of work strain and psychosocial stressors of migrants throughout the life course. Another noteworthy observation is that in the univariate analysis of health trajectories, where a simple regression line is applied, the observed patterns differ, potentially leading to alternative conclusions. While the trajectory for women appears quite similar in both Fig. 3and Fig. 4, for men the less elaborated analysis might suggest that European men would eventually converge with German men in terms of physical health. However, a closer examination—considering factors such as measurement errors, growth modeling, and longitudinal and multigroup measurement invariance, and employing a multidimensional approach to health (SF-12)—reveals distinct trajectories, indicating a pattern of continuity or status maintenance. This finding is consistent with previous research, underscoring the importance of accounting for longitudinal and multigroup measurement invariance to avoid significant bias in health trajectory analyses (Holz and Mayerl 2024). Our study benefits from a longitudinal design, spanning from 2002 to 2018, allowing for the assessment of health trajectories over time, which provides a more comprehensive understanding of health disparities among migrants. Further, the use of latent growth curve analysis to model multiple health indicators as latent variables is a robust approach that can better capture the complex interplay of various healthrelated factors over time. Further, the inclusion of occupational class as a growth curve allowed us to control for potential time-variant dynamics. The incorporation of multigroup longitudinal measurement invariance ensures that the observed health disparities are not confounded by measurement bias across different groups or time points, enhancing the validity of the findings. Controlling for different countries of origin allows for a nuanced examination of health disparities among migrants, acknowledging that experiences and health outcomes may vary across different national backgrounds. Nonetheless, our study does reveal certain limitations. Although we managed to account for the time-invariant structure of the EGP variable by modeling it as a latent growth curve, we found only minimal variance over time, which was statistically insignificant. Consequently, we were unable to incorporate the latent slope of EGP into our analysis, restricting its role to covarying with the covariates of interest. It would have been intriguing to investigate how social mobility, specifically in terms of occupational class trajectories, interacts with health trajectories, shedding light on the question of selection—whether healthier individuals tend to secure better jobs K 46 M. Holz, J. Mayerl and consequently experience more gradual declines in health. However, due to our approach, this investigation was not feasible. Additionally, we were unable to fully capture the healthy migrant phenomenon. Despite our initial theoretical considerations incorporating the healthy migrant perspective, logistical constraints, particularly related to cell size issues, prevented us from narrowing the sample exclusively to recent migrants. Given that the paper primarily focuses on postmigration health trajectories, we consider this limitation relatively minor. However, to comprehensively understand health selection prior to migration, the impact of migrating into lower social strata, and the subsequent longterm health implications, it may be necessary to consider the duration of stay as well. Yet obtaining a dataset with extensive observation periods, multidimensional health measures, and distinctions between countries of origin and gender groups presents considerable challenges. While the study identifies the influence of occupational class on health disparities, it does not delve deeply into the exact mechanisms through which this influence operates. It remains unclear whether factors such as those mentioned above add differing health detriments between migrant groups. Further, we did not account for the effect of unemployment and underemployment, which have been demonstrated to be significant factors in explaining health disparities among migrants (EspinozaCastro et al. 2019; Rellstab et al. 2016;Yang2021). It is noteworthy that this study focuses on a historical migrant population, which may not completely reflect the current composition of migrants in Germany. The biggest group within the European migrant sample in this study consists of Eastern Europeans from Poland, Russia, the former Soviet Union, and Romania. The second largest group is Southern Europeans, mainly from Italy and Greece. The non-European group in our sample predominantly comprises Turkish migrants, whose primary migration influx occurred in the past and in the context of guest working contracts and family reunification (Özel and Nauck 1987). To assess the health trajectories and their interaction with occupational class among the more recent migrant groups, several decades must elapse before a comparable analysis with an equal number of observation periods can be carried out. Predicting health trajectories is challenging, but given the similar contexts of former and recent migration patterns, it is reasonable to assume that patterns of health differences will persist. Namely, based on the context of Germany’s migration policies, migrants from both former and recent waves can be grouped regarding the ease of migration, which manifests itself in structural barriers, migration costs, and remigration patterns: Before 2000, Southern European migrants integrated more easily and had higher remigration rates, while Turkish workers faced higher migration costs and integration challenges and showed lower remigration rates (Gundel and Peters 2008; Smolny and Rieber 2016). These patterns persist today, with Eastern European migrants, being European Union citizens, experiencing lower migration costs and easier integration, while non-European migrants such as refugees from Syria, Afghanistan, and Iraq are facing higher migration costs and more barriers. Consequently, health disadvantages are still more pronounced for migrants from non-European countries than from European countries. K Can Differing Occupational Class Positions Explain Migrant Health Inequalities? Differences... 47 Furthermore, recent migration patterns also resemble those of the past among certain refugee groups. Over 80% of Ukrainian refugees in Germany plan to return after the war (Brücker et al. 2023), similar to former Yugoslavian refugees in the 2000s (Bahar et al. 2024). In contrast, only about 38% of Syrian refugees in Germany intend to return (Al Husein and Wagner 2023), resembling Turkish guest workers who stayed longer than initially planned. Thus, the similarities between former and recent migration patterns indicate a continuation of health disparities based on country of origin and migration context. We assume that migration policies that selectively facilitate integration and navigation in the host country will consistently create a group with more advantages and a group with more disadvantages, regardless of the country. Therefore, we believe that the results of this study extend beyond the German case. However, these assumptions have to be tested empirically. This study aimed to uncover the trajectory of migrant health, examining whether health disparities between European migrants, non-European migrants, and native Germans evolve over time. Additionally, we sought to determine whether varying health trajectories could be elucidated by occupational class, serving as a proxy for job exposure. Understanding the complexities underlying these inequalities is challenging, as they are shaped by a combination of factors including the individual’s country of origin and gender. Our findings indicate that divergence is the prevailing pattern in migrant health disparities over time, with the health gap between migrants and native Germans widening progressively. This divergence can be primarily explained by the lower occupational class positions inhabited by migrants. However, for European migrant men, the trajectory more closely resembles a pattern of continuity, resulting in a sustained health disadvantage that does not exacerbate over time. Furthermore, our analysis highlights that women, particularly migrant women, experience multiple health disadvantages compared to both their native counterparts and men in general, leading to a notably pronounced decline in health over time. These results emphasize the critical importance of considering the unique experiences of migrants from different countries and of different genders. Considering these aspects in both policy development and efforts to promote unionization represents essential strategies for addressing labor market discrimination and improving working conditions and thus health, especially for migrant women. Author Contribution M. Holz conceived of the presented idea. M. Holz developed the theory and performed the computations. M. Holz and J. Mayerl verified the analytical methods. J. Mayerl supervised the findings of this work. M. Holz and J. Mayerl discussed the results. M. Holz wrote the manuscript. Funding Open Access funding enabled and organized by Projekt DEAL. Data Availability Statement The data used in this study were made available by the German Socio-Economic Panel Study at the German Institute for Economic Research (DIW), Berlin. The data are not publicly available due to containing information that could compromise the privacy of research participants. The code used during the current study is available from the corresponding author on reasonable request for all interested researchers. Declarations Conflict of interest M. Holz and J. Mayerl declare that they have no competing interests. K 48 M. Holz, J. Mayerl Ethical standards All data analysis procedures were carried out strictly in accordance with the guidelines of the data providers and the standards of good scientific practice. Open Access Dieser Artikel wird unter der Creative Commons Namensnennung 4.0 International Lizenz veröffentlicht, welche die Nutzung, Vervielfältigung, Bearbeitung, Verbreitung und Wiedergabe in jeglichem Medium und Format erlaubt, sofern Sie den/die ursprünglichen Autor(en) und die Quelle ordnungsgemäß nennen, einen Link zur Creative Commons Lizenz beifügen und angeben, ob Änderungen vorgenommen wurden. Die in diesem Artikel enthaltenen Bilder und sonstiges Drittmaterial unterliegen ebenfalls der genannten Creative Commons Lizenz, sofern sich aus der Abbildungslegende nichts anderes ergibt. Sofern das betreffende Material nicht unter der genannten Creative Commons Lizenz steht und die betreffende Handlung nicht nach gesetzlichen Vorschriften erlaubt ist, ist für die oben aufgeführten Weiterverwendungen des Materials die Einwilligung des jeweiligen Rechteinhabers einzuholen. Weitere Details zur Lizenz entnehmen Sie bitte der Lizenzinformation auf http://creativecommons.org/licenses/by/ 4.0/deed.de. References Al Husein, Nawras, and Natascha Wagner. 2023. Determinants of Intended Return Migration among Refugees: A Comparison of Syrian Refugees in Germany and Turkey. International Migration Review 57(4):1771–1805. Allison, Paul D. 2003. Missing data techniques for structural equation modeling. Journal of Abnormal Psychology 112(4):545–557. Arends-Tóth, Judit, and Fons J. R. van de Vijver. 2008. Family Relationships among Immigrants and Majority Members in the Netherlands: The Role of Acculturation. Applied Psychology 57(3):466–487. Assari, Shervin. 2017. Combined Racial and Gender Differences in the Long-Term Predictive Role of Education on Depressive Symptoms and Chronic Medical Conditions. Journal of Racial and Ethnic Health Disparities 4(3):385–396. Aukst-Margeti´ c, Branka, and Branimir Margeti´ c. 2005. Religiosity and Health Outcomes: Review of Literature. Collegium antropologicum 29(1):365–371. Auswärtiges Amt (Federal Foreign Office). 2015. Deutsch als Fremdsprache weltweit. Datenerhebung 2015. Berlin. Bader, Benjamin, Sebastian Stoermer, Anna Katharina Bader and Tassilo Schuster. 2018. Institutional discrimination of women and workplace harassment of female expatriates. Journal of Global Mobility 6(1):40–58. Bahar, Dany, Andreas Hauptmann, Cem Özgüzel and Hillel Rapoport. 2024. Migration and Knowledge Diffusion: The Effect of Returning Refugees on Export Performance in the Former Yugoslavia. Review of Economics and Statistics 106(2):287–304. Bauer, Georg F., Carola A. Huber, Gregor J. Jenny, Frithjof Müller and Oliver Hämmig. 2009. Socioeconomic status, working conditions and self-rated health in Switzerland: explaining the gradient in men and women. International Journal of Public Health 54(1):23–30. Beckett, Megan. 2000. Converging Health Inequalities in Later Life-An Artifact of Mortality Selection? Journal of Health and Social Behavior 41(1):106. Berwing, Stefan. 2019. Migrant self-employment in Germany: on the risks, characteristics and determinants of precarious work. In Self-Employment as Precarious Work, ed. Wieteke Conen and Joop Schippers, 186–214. Edward Elgar Publishing. Bollen, Kenneth A., and Patrick J. Curran. 2006. Latent curve models: a structural equation perspective. Hoboken, NJ: John Wiley & Sons. Borjas, George. 1987. Self-selection and the earnings of immigrants. NBER Working Paper Series No. 2248. National Bureau of Economic Research. Brücker, Herbert, Andreas Ette, Markus M. Grabka, Yuliya Kosyakova, Wenke Niehues, Nina Rother, C. Katharina Spieß, Sabine Zinn, Martin Bujard, Adriana R. Cardozo Silva, Jean Philippe Décieux, Amrei Maddox, Nadja Milewski, Lenore Sauer, Sophia Schmitz, Silvia Schwanhäuser, Manuel Siegert, Hans Steinhauer and Kerstin Tanis. 2023. Ukrainian Refugees in Germany: Evidence From a Large Representative Survey. Comparative Population Studies – Zeitschrift für Bevölkerungswissenschaft 48:395–424. K Can Differing Occupational Class Positions Explain Migrant Health Inequalities? Differences... 49 Brücker, Herbert, Simone Bartsch, Jan Goebel, Simon Kühne, Elisabeth Liebau, Parvati Trübswetter, Ingrid Tucci and Jürgen Schupp. 2014. The new IAB-SOEP Migration Sample: an introduction into the methodology and the contents. SOEP Survey Papers 216:Series C. Brünger, Martin, Sebastian Bernert and Karla Spyra. 2020. Berufstätigkeit als Proxy für Arbeitsbelastungen? Routinedatenanalyse am Beispiel der Rehabilitation. Das Gesundheitswesen 82(S 01):S41–S51. Campos-Serna, Javier, Elena Ronda-Pérez, Lucia Artazcoz, Bente E. Moen and Fernando G. Benavides. 2013. Gender inequalities in occupational health related to the unequal distribution of working and employment conditions: a systematic review. International Journal for Equity in Health 12:57. Chandola, Tarani, Jane Ferrie, Amanda Sacker and Michael Marmot. 2007. Social inequalities in self reported health in early old age: follow-up of prospective cohort study. BMJ (Clinical research ed.) 334(7601):990. Clumeck, Nicolas, Chantal Kempenaers, Isabelle Godin, Michèle Dramaix, Marcel Kornitzer, Paul Linkowski and France Kittel. 2009. Working conditions predict incidence of long-term spells of sick leave due to depression: results from the Belstress I prospective study. Journal of Epidemiology and Community Health 63(4):286–292. Cook, Won Kim, Xinjun Li, Kristina Sundquist, Kenneth S. Kendler, Jan Sundquist and Katherine J. Karriker-Jaffe. 2021. Drinking cultures and socioeconomic risk factors for alcohol and drug use disorders among firstand second-generation immigrants: A longitudinal analysis of Swedish population data. Drug and Alcohol Dependence 226:108804. Corna, Laurie M. 2013. A life course perspective on socioeconomic inequalities in health: a critical review of conceptual frameworks. Advances in Life Course Research 18(2):150–159. Damelang, Andreas, Sabine Ebensperger and Felix Stumpf. 2021. Immigrants’ Labour Market Disadvantages Across Western Europe: the Role of Composition and Context. Journal of International Migration and Integration 22 (4):1525–1550. Darmon, Nicole, and Myriam Khlat. 2001. An overview of the health status of migrants in France, in relation to their dietary practices. Public Health Nutrition 4(2):163–172. Davidov, Eldad, Bart Meuleman, Jan Cieciuch, Peter Schmidt and Jaak Billiet. 2014. Measurement Equivalence in Cross-National Research. Annual Review of Sociology 40(1):55–75. Di Napoli, Anteo, Alessandra Rossi, Gianfranco Alicandro, Martina Ventura, Luisa Frova and Alessio Petrelli. 2021. Salmon bias effect as hypothesis of the lower mortality rates among immigrants in Italy. Scientific Reports 11(1):8033. Dragano, Nico, Morten Wahrendorf, Kathrin Müller and Thorsten Lunau. 2016. Arbeit und gesundheitliche Ungleichheit : Die ungleiche Verteilung von Arbeitsbelastungen in Deutschland und Europa. Bundesgesundheitsblatt, Gesundheitsforschung, Gesundheitsschutz 59(2):217–227. Edberg, Mark, Sean Cleary and Amita Vyas. 2011. A trajectory model for understanding and assessing health disparities in immigrant/refugee communities. Journal of Immigrant and Minority Health 13(3):576–584. Engelhardt-Woelfler, Henriette, and Liliya Leopold. 2020. Gesundheitliche Ungleichheit im Lebenslauf. In Handbuch Gesundheitssoziologie, eds. Peter Kriwy and Monika Jungbauer-Gans, 359–372. Wiesbaden: Springer Fachmedien Wiesbaden. Erikson, Robert, and John H. Goldthorpe. 1993. The constant flux: A study of class mobility in industrial societies. Oxford: Clarendon. Espinoza-Castro, Bernarda, Luis E. Vásquez Rueda, Rossana V. Mendoza Lopez and Katja Radon. 2019. Working Below Skill Level as Risk Factor for Distress Among Latin American Migrants Living in Germany: A Cross-Sectional Study. Journal of Immigrant and Minority Health 21(5):1012–1018. Ferraro, Kenneth F., Tetyana P. Shippee and Markus H. Schafer. 2009. Cumulative inequality theory for research on aging and the life course.InHandbook of theories of aging, 2nd edition, eds.Vern L. Bengston, Daphna Gans, Norella M. Pulney and Merril Silverstein, 413–433. New York: Springer Publishing Company. Fujishiro, Kaori, and Annekatrin Hoppe. 2020. Toward a Life-Course Perspective of Migrant Worker Health and Well-Being. In Health, Safety and Well-being of Migrant Workers: New Hazards, New Workers, eds. Angeli Santos and Francisco Díaz Bretones, 23–42. Cham: Springer. Fuller-Thomson, Esme, Sarah Brennenstuhl, Rachel Cooper and Diana Kuh. 2016. An investigation of the healthy migrant hypothesis: Pre-emigration characteristics of those in the British 1946 birth cohort study. Canadian Journal of Public Health 106(8):e502–8. Geyer, Siegfrid, and Richard Peter. 1999. Occupational health. Occupational status and all-cause mortality. A study with health insurance data from Nordrhein-Westfalen, Germany. European Journal of Public Health 9(2):114–118. K 50 M. Holz, J. Mayerl Gonzalez-Mulé, Erik, and Bethany Cockburn. 2017. Worked to Death: The Relationships of Job Demands and Job Control with Mortality. Personnel Psychology 70(1):73–112. Gundel, Sebastian, and Heiko Peters. 2008. What determines the duration of stay of immigrants in Germany? International Journal of Social Economics 35(11):769–782. Hiesinger, Karolin, and Silke Tophoven. 2019. Job requirement level, work demands, and health: a prospective study among older workers. International Archives of Occupational and Environmental Health 92(8):1139–1149. Hiller, Julia, Kathrina Schatz and Hans Drexler. 2017. Gender influence on health and risk behavior in primary prevention: a systematic review. Journal of Public Health 25(4):339–349. Holz, Manuel. 2022. Health inequalities in Germany: Differences in the ‘Healthy migrant effect’of European, non-European and internal migrants. Journal of Ethnic and Migration Studies 48(11):26202641. Holz, Manuel, and Jochen Mayerl. 2024. Migrant Health Inequalities or Unequal Measurements? Testing for Cross-cultural and Longitudinal Measurement Invariance of Subjective Physical and Mental Health. methods, data, analyses 18.1:20. Hoven, Hanno, Morten Wahrendorf and Johannes Siegrist. 2015. Occupational position, work stress and depressive symptoms: a pathway analysis of longitudinal SHARE data. Journal of Epidemiology and Community Health 69(5):447–452. Hönekopp, Elmar, Gisela Will and Stefan Rühl. 2002. Migrants, minorities and employment in Germany: Exclusion, discrimination and anti-discrimination. (RAXEN Report / European Monitoring Centre on Racism and Xenophobia, National Focal Point Germany). Bamberg: europäisches forum für migrationsstudien (efms) Institut an der Universität Bamberg. Jatrana, Santosh, Ken Richardson and Samba Siva Rao Pasupuleti. 2018. Investigating the Dynamics of Migration and Health in Australia: A Longitudinal Study. European journal of population = Revue europeenne de demographie 34(4):519–565. Johnson, Jeffrey V., Walter Stewart, Ellen M. Hall, Peeter Fredlund and Tores Theorell. 1996. Long-term psychosocial work environment and cardiovascular mortality among Swedish men. American Journal of Public Health 86(3):324–331. Kalter, Frank, and Nadia Granato. 2018. Migration und ethnische Ungleichheit auf dem Arbeitsmarkt. In Arbeitsmarktsoziologie, eds. Martin Abraham and Thomas Hinz, 355–387. Wiesbaden: Springer Fachmedien Wiesbaden. Kleiser, Christina, Gert B. M. Mensink, Hannelore Neuhauser, Liane Schenk and Bärbel-Maria Kurth. 2010. Food intake of young people with a migration background living in Germany. Public Health Nutrition 13(3):324–330. Kline, Rex B. 2023. Principles and practice of structural equation modeling. New York: The Guilford Press. Kogan, Irena. 2011. New Immigrants—Old Disadvantage Patterns? Labour Market Integration of Recent Immigrants into Germany. International Migration 49(1):91–117. Krokstad, Steinar, Kristen Ringdal and Steinar Westin. 2002. Classifying people by social class in population based health surveys. Norsk epidemiologi 12(1):19–25. Kroll, Lars Eric. 2011. Konstruktion und Validierung eines allgemeinen Index für die Arbeitsbelastung in beruflichen Tätigkeiten auf Basis von ISCO-88 und KldB-92. Methoden, Daten, Analysen (mda) 5(1):63–90. Lahelma, Eero, Pekka Martikainen, Ossi Rahkonen, Eva Roos and Peppiina Saastamoinen. 2005. Occupational class inequalities across key domains of health: results from the Helsinki Health Study. European Journal of Public Health 15(5):504–510. Landsbergis, Paul A. 2010. Assessing the contribution of working conditions to socioeconomic disparities in health: a commentary. American Journal of Industrial Medicine 53(2):95–103. Lee, Sunmin, Allison H. O’Neill, Emily S. Ihara and David H. Chae. 2013. Change in self-reported health status among immigrants in the United States: associations with measures of acculturation. PloS one 8 (10):e76494. Leong, Frederick T. L., Donald E. Eggerth and Michael A. Flynn. 2014. A Life Course Perspective on Immigrant Occupational Health and Well Being. In Contemporary Occupational Health Psychology, 97–113. Hoboken, NJ: John Wiley & Sons. Leopold, Liliya, and Henriette Engelhardt. 2011. Bildung und Gesundheitsungleichheit im Alter: Divergenz, Konvergenz oder Kontinuität? Kölner Zeitschrift für Soziologie und Sozialpsychologie 63(2):207–236. Marsh, Herbert W., Bengt Muthén, Tihomir Asparouhov, Oliver Lüdtke, Alexander Robitzsch, Alexandre J. S. Morin and Ulrich Trautwein. 2009. Exploratory Structural Equation Modeling, Integrating CFA K