In Europe, previous studies have reported that immigrants generally exhibit a higher prevalence of depressive symptoms than native populations.1–4 In Spain, a country with the third highest proportion of immigrant population in the European Union,5 evidence for this association is mixed,6–9 with studies reporting a higher prevalence,9 lower prevalence,6,7 or no difference in the prevalence of depression between immigrants and native Spanish adults.8 In addition, immigrants are less likely to consume antidepressants.6,8 Thus, it is important to examine the associations of immigration status with depression and related outcomes.

Social, economic, and psychosocial factors influenced the association between immigration status and depression.1,10,11 Factors contributing to these social inequalities include unemployment, lower social support, and previous mental health conditions. Precarious working conditions, for example, have been shown to significantly increase the risk of depression among immigrant workers.12,13 Additionally, immigrants often face unique challenges such as discrimination, which is linked to depressive symptoms.14 Gender inequalities have also been documented among immigrants, with depression being more common among women than men.8 Similarly, social support tends to buffer the stress related to immigration status,15 and thus, may change the relationships of immigration status with depression and antidepressant use.

Spain presents unique challenges regarding the migration processes, as noted in previous studies.16,17 Currently, 17.1% of the population has an immigrant background, mostly from South America or Africa.5 However, our understanding of social inequalities in mental health, based on immigration status, remains limited. Therefore, we aimed to examine whether there is an association of immigration status with depression and antidepressant use in adults who participated in the European Health Interview Survey (EHIS) of Spain in 2014 and 2020, and whether this association differs by sex/gender and social support.

Table 1 displays key characteristics of the study population stratified by immigration status in EHIS 2014 and 2020. Most of the immigrant population originated from Latin American countries. Compared with the Spanish population, regardless of the survey year, immigrants were younger, less likely to be men, less educated, less likely to work in a non-manual job, more likely to have precarious employment or be unemployed, and more likely to rate their health as very good/good. Both self-reported prevalence of depression and antidepressant use were higher in the Spanish population than in the immigrant population.

The overall prevalence estimates for self-reported depression were 9.1% and 7.3% in 2014 and 2020, respectively (data not shown). In both 2014 and 2020, regardless of immigration status, the prevalence of depression was lower among men, those who were married, those with better employment status, and those who rated their health as very good/good (Table 2). Between 2014 and 2020, the prevalence of antidepressant use increased in the immigrant population (62.5% vs. 80.3%), whereas the opposite was true for native Spanish adults (74.4% in 2014 vs. 72.3% in 2020).

Table 3 shows the results of the association between immigration status and depression stratified by survey year. In the unadjusted analyses, the prevalence of depression was 52% (PR: 0.48, 95%CI: 0.36-0.65) and 31% (PR: 0.69, 95%CI: 0.53-0.88) lower among the immigrant population than among native Spanish adults in 2014 and 2020, respectively. In the adjusted analyses, these associations remained the same in 2014 and 2020. However, their strengths were attenuated for both 2014 (PR: 0.53, 95%CI: 0.50-0.79) and 2020 (PR: 0.71, 95%CI: 0.55-0.91).

We observed a similar pattern when assessing antidepressant consumption. For both survey years, immigrants reported a 69% lower use of these medications in 2014 and a 34% in 2020 lower use of this medications, as compared with their native Spanish counterparts. In the adjusted analyses, this association remained statistically significant only in 2014 (PR: 0.36, 95%CI: 0.24-0.55). We found a significant effect measure modification by immigration status and survey year for antidepressant medication use (p interaction: 0.01) but not for depression (p interaction: 0.07). However, we did not find statistically significant interactions for depression or for the use of antidepressants for immigration status with survey year and sex/gender (p interactions: 0.32 depression and 0.12 antidepressant use), or social support (p interactions: 0.28 depression and 0.31 antidepressant use).

Our findings show a lower probability of self-reported prevalence of depression and antidepressant use among immigrants than among native Spanish adults in both EHIS 2014 and 2020. However, the association between immigration status and antidepressants was only significant in 2014 before and after adjustment. Finally, although the association of immigration status with antidepressant use differed with survey year, our findings remained the same regardless of sex/gender or social support.

Our finding of a lower prevalence of depressive symptoms and antidepressant use in the immigrant population in 2014 can be explained, at least partially, by the economic crisis that lasted from 2008 to 2014. This crisis led to the return of immigrant populations to their countries of origin between 2010 and 2015, due to economic difficulties and dramatic job losses.27 Thus, it is possible that immigrants who remained had more favorable conditions in terms of social position and health status. We continue to observe immigrant health advantage in 2020 despite immigrants having more precarious employment after the economic crisis,28,29 being employed in positions considered precarious13 or essential jobs,30 and having worse living conditions than their native Spanish counterparts. Moreover, other potential explanations could be the healthy immigrant effect, where there is a selection effect for healthy adults to migrate to another country, and the low use of the healthcare system observed among immigrants, especially for mental health.23,31,32 The latter is associated with stigma and cultural beliefs among immigrants, leading to avoiding seeking care, and thus, depression may be underestimated for this population.33,34 Despite the low prevalence of depression among immigrant adults, it is worth noting that the prevalence of depression in 2020 was higher among those aged 45 years or older than among their native Spanish peers. This was not the case in 2014, when a higher prevalence of depression among immigrants was observed among those aged 18-34 years. This change could be due because younger immigrants adults in 2014 encountered harsher socioeconomic conditions than their native Spanish counterparts,35,36 whereas older immigrants adults could have experience worst physical and mental health due to cumulative life and working conditions as well as the lock down during COVID-19 than their younger counterparts in 2020.37,38

We did not find a difference in depression or antidepressant use by sex/gender or social support, regardless of migrant status or survey year. Our findings regarding the lack of heterogeneity related to sex/gender are inconsistent with previous research suggesting that immigrant women have worse mental health than Spanish native women. However, this is not the case for immigrant men.8,9 There are several possible explanations for the observed differences. First, men are more likely to minimize depressive symptoms and respond differently to questions regarding their mental health conditions. Second, women are more likely to use the healthcare system,39 and thus, are more likely to be aware of their mental health. Finally, women suffer from poor employment conditions and are responsible for dependent care,8 and their work conditions have worsened over time.40 However, despite their access to healthcare and employment conditions, women, and especially immigrant women, coped better than men with the impact of the economic crisis,41 which could have protected their mental health. Regarding our findings related to social support, evidence suggests that social support plays a significant role in mitigating the negative association between immigration status and depression in Spain.15 However, this relationship may depend on the level of social integration and economic stability experienced by different immigrant groups, such as homeownership rates.42 Further research is needed to understand the specific mechanisms through which sex/gender and social support influence mental health outcomes among immigrants.

An important limitation of this study is the sample size of the immigrant population, which only allows us to analyze immigrants as a single group without accounting for key differences by country of origin or cultural background. Furthermore, we excluded immigrants from countries with HDI >0.80 as they could have similar health status as native Spanish adults. However, we repeated the analyses including immigrants from countries with HDI >0.80, and the results did not change significantly from the results presented in Table 3. The latter finding suggests that such exclusion was unlikely to bias our findings. Second, immigrants experiencing language barriers or precarious conditions (e.g., homelessness) were not represented in these surveys. Third, we did not account for length of stay in Spain for the immigrant population because of small size for some cells and the high correlation of the variable with immigration status. Future studies should consider length of stay, which may provide insights into the health status of the immigrant population. Fourth, we did not consider depression severity. Finally, part of the fieldwork of the 2020 survey was conducted during the state of alarm in Spain due to the COVID-19 pandemic. The latter may have precluded a greater outreach, leading to low participation in the immigrant population. Similarly, COVID-19 may have biased our results, especially for immigrants who bear a greater burden of infection and death due to their essential jobs.43,44 Despite these limitations, the strength of this study is that we provided a nationwide assessment of the evolution of depression prevalence among the immigrant population in Spain using the two most recent waves of the European Health Interview Survey. Moreover, we observed individuals across a wide age range.

Our findings suggest that there is a protective effect for depression among the immigrant population in Spain regardless of sex/gender or social support. Future research should investigate factors contributing to depression and mental health-related outcomes in the immigrant population. The findings of such investigations could provide insights to tailor programs and interventions to address the mental health status of the population, particularly depression, regardless of immigration status, sex/gender, or social support.

We used data from the National Institute of Statistics (Instituto Nacional de Estadistica), Madrid, Spain, available at: https://www.ine.es/dyngs/INEbase/en/operacion.htm?c=Estadistica_C&cid=1254736176784&menu=resultados&secc=1254736195298&idp=1254735573175#tabs-1254736195298

Alberto Lana.

The corresponding author, on behalf of the other authors guarantee the accuracy, transparency and honesty of the data and information contained in the study, that no relevant information has been omitted and that all discrepancies between authors have been adequately resolved and described.

L.N. Borrell conceptualized the study design, conducted the analyses, prepared the first draft of the manuscript and critically reviewed the final manuscript. N. Lanborena, E. Rodríguez-Álvarez contributed to the study conceptualization and critically reviewed the manuscript and the results. J. Díez and S. Yago-González critically reviewed the manuscript and the results. L.N. Borrell, J. Díez, S. Yago-González, N. Lanborena and E. Rodríguez-Álvarez have read and approved the final manuscript.

This work was supported by the Multiannual Agreement reference # CM/BG/2023-002 (L.N. Borrell).

None.