Self-reported depression among migrant and native populations in Greece in times of crisis

Abstract

Background

Recent evidence indicates to the alarming increase of depression, especially in Southern European countries. According to the literature, depressive symptoms are strongly associated with socioeconomic factors, while the prevalence of depression significantly differs by gender. Several scholars point out that many of the risk factors for self-reported depression increase during an economic crisis. These studies, however, mostly overlook the fact that the impact of these factors may vary between natives and migrants.

Methods

Using data from the MIGHEAL study, we examine the variation of self-reported depression by migrant status and gender in Greece during the crisis. We also investigate the association of socioeconomic status, family-related factors, social trust and perceived discrimination with self-reported depression. Separate multiple linear regression models are fitted across three examined groups (Greeks, Albanians and migrants from other countries) as well as for Greek males and females.

Results

Our findings indicate the mixture of the determinants of self-reported depression by both migrant status and gender. Women report higher rates of self-reported depression than men, as do natives compared to migrants, while getting older is associated with higher levels of depression in both genders and all three citizenship groups. Further, higher socioeconomic position is associated with lower levels of depression, especially in the case of Greek males, while financial strain appears to be a significant determinant of self-reported depression for both native and migrant populations.

Conclusions

Our analysis does not confirm previous evidence that migrants are at an increased risk for depression. Age, gender, perceived discrimination and subjective general health appear to be more decisive determinants of self-reported depression than migrant background. Low socioeconomic status seems to be a risk factor for depression only in the case of Greeks.

Introduction

Depression is considered a leading cause of ill health and disability.^1–3 Many scholars have underlined the adverse effects of economic downturn, as regards the impact of increased levels of unemployment and job insecurity, on general health and depression.^4–7 In particular, various studies have underscored the importance of socio-demographic factors as key determinants of depression; thus women, young persons the unemployed and people with a lower socioeconomic status facing financial strain are more likely to report depressive symptoms.^4,5,8–11 In addition evidence has shown that social trust is a key protective factor that can significantly reduce social stress and the risk of depression^12,13 whereas childhood experiences of financial hardship increase the risk of depression.¹⁴

While scholars do not observe consistent patterns linking mental health problems to specific socioeconomic factors in the European Union, empirical evidence suggests that many of the above mentioned risk factors are exacerbated during periods of economic hardship.¹⁵ The prevalence of depression is, therefore, expected to increase in times of recession. Indeed, recent evidence points out to the alarming increase of depression especially in Southern European countries which were hit by recession, namely Cyprus, Greece and Spain.^12,16,17

Up to now, only a few studies have empirically investigated the impact of the economic crisis on the prevalence of depression in Greece^18–22 indicating increased incidence of serious depression during the economic downturn.^23,24 According to a study carried out by the Institute of Mental Health of the University of Athens, between 2009 and 2011 there was a 20.6% increase in major depressive disorders,¹⁹ whilst the levels of general anxiety disorders remained stable. In the same study, those who were unemployed, married and over 45 years old, more than doubled being at risk of having serious depressive symptoms.¹⁹ Similarly, the most recent nationwide survey on health in Greece conducted in 2014, revealed a sharp increase of depression (80.8% compared to 2009) in the Greek population with women being most severely affected.²⁵

Even though these studies provide useful insights for explaining depression among the native population, they do not pay much attention to the variation of depressive symptoms among native and migrant populations as well as the differentiated association of depression with socio-demographic factors. Evidence from round 3 of the European Social Survey (carried out in 2006–2007) has shown that depression increases with age and is more frequent among women, those without a partner and among migrants originating from outside of Europe.²⁶ As migrants are at increased risk of social exclusion due to lower socioeconomic status and/or perceived discrimination, they are often considered more vulnerable to mental health problems indicated than non-migrants.^27,28

There is a lack of respective evidence for Greece as the country did not participate in ESS rounds 3 and 6 (2006 and 2012). The MIGHEAL (Health Inequalities among Migrant Population) study,²⁹ carried out in 2016, has been designed to fill in this gap providing the most recent data on self-reported depression and socioeconomic factors for both native and migrant populations in Greece. Therefore, the aim of this article is firstly to investigate the variation of self-reported depression by migrant status and gender, and secondly, to identify the key socio-demographic determinants of depression for natives and migrants in Greece^20,30,31 six years after the introduction of fiscal adjustement programmes and the accompanying austerity measures.

Methods

Data

The MIGHEAL survey was carried out according to ESS round 7 specifications, with a representative sample of 1332 individuals (505 migrants and 827 non-migrants). The response rate was 50%.²⁹

The MIGHEAL data presented here correspond to 1327 respondents (49.4% male, 50.6% female) aged over 15 and living in private households in Greece, with sufficient knowledge of the Greek language to be interviewed in Greek. For the purpose of the analysis migrants were defined as those having foreign citizenship (excluding developed countries). Thus, five respondents born in one of the EU member-countries or in developed countries outside the EU were not included in our analysis. Based on the respondent’s citizenship, the sample was divided into three groups: Greeks (64.5%) and two migrant groups, Albanians (23.2%) and citizens of other countries (12.3%). As Albanians have dominated migration to Greece in the last decades, accounting for more than half of the foreign population of Greece in the population censuses of 2001 and 2011, and it was therefore not surprising that they dominated the sample of migrants in the MIGHEAL study and that their number was sufficient to permit their analysis as a separate group.

Dependent variable

Self-reported depression

Self-reported depression was measured using the 8-item scale of the Centre of Epidemiological Studies-Depression (CES-D), which is a shorter version of the CES-D scale originally developed by Radloff (1977) and has been used in ESS7.³² CES-D is not a clinical but a screening tool, as it does not measure all the factors that constitute clinical depression. Nonetheless, it is a valid measure for identifying the population at risk of developing depression.³³

According to this measure, respondents are asked to report the frequency of occurrence, within the week before the interview, of eight symptoms: (i) felt depressed, (ii) felt everything they did was an effort, (iii) had restless sleep, (iv) were happy, (v) felt lonely, (vi) enjoyed life, (vii) felt sad and (viii) could not get going. Four options were given concerning the frequency: ‘never or almost never’, ‘sometimes’, ‘most of the time’, ‘always or almost always’. The items were recoded with a range of 0 through 3. The eight items, of which two positive items were reversed, were summed to form a 24-point scale. To be assigned a sum score, the respondent had to answer all the items.¹⁴ A total of 1302 respondents met this criterion. Higher scores indicate more symptoms. For descriptive analysis, an individual qualifies for high risk of self-reported depression if he/she scores 10 or above on the scale.³⁴

Covariates

Age

Respondent’s age is used as a continuous variable, ranging from 15 to 94 years old.

Socio-economic position

To investigate the association between socioeconomic position and self-reported depression during the financial crisis, we introduced the following variables in the model: labour market status, financial strain and educational level. Respondents identified their labour market status by responding to the question: ‘Which of these descriptions best describes your situation (during the last seven days)’ Those who reported being without employment and not seeking employment, students, retired, housekeepers, sick and disabled, engaged in housework or looking after children, being at community or military service and others, were considered as inactive with respect to the labour market and were collapsed into one group. This yielded the following three labour market categories: being in paid work, unemployed and inactive. Respondents’ perceived financial strain was captured by asking them how they were feeling about their household income at present. The responses were grouped into: ‘Very difficult/difficult on present income’ and ‘living comfortably/coping on present income’. Education level was measured using the International Standard Classification of Education (ISCED) score. We applied ISCED in three broad categories: lower secondary education or less, upper secondary and tertiary.

Experiences of family conflict and economic hardship in childhood

The level of conflict in the family while growing up was measured by asking the respondent ‘how often was there serious conflict between the people living in your household when you were growing up’, providing the following response options: ‘Always’, ‘Often’, ‘Sometimes’, ‘Hardly ever’. The variable was recoded into two groups: “never/hardly/ever/sometimes” and “often/always”. The same pattern was followed for the reported frequency of economic hardship experienced in childhood.

Subjective general health

Self-reported general health was measured by asking respondents to assess their general health status, using the following response options: ‘very good’, ‘good’, ‘fair’, ‘bad’ and ‘very bad’. The original variable was recoded into a binary variable so that high scores represent very bad/bad general health.

Perceived discrimination

Respondents were asked to identify the reasons why they might have suffered discrimination during the year prior to the survey. They could report discrimination on the grounds of nationality, colour or race, religion, language, ethnic group, age, gender, sexuality, disability and other grounds. Discrimination on the grounds of nationality was the most frequently reported reason. However, the above categories were not mutually exclusive and, thus, the total discrimination score was calculated by adding up all the reasons cited by the respondent.

Social trust

Social trust was measured as the mean of the scores given by respondents for each of the following questions regarding trust attitudes and perceptions, using an 11-point scale: ‘Most people can be trusted or you can't be too careful in dealing with people’, ‘Most people try to take advantage of you or try to be fair’ and ‘Most of the time people try to be helpful or mostly looking out for themselves’.

Providing unpaid care and children living at home

Provision of unpaid care was operationalized according to whether or not the respondent was looking after or helping family members, friends, neighbours or others. Similarly, the variable ‘children living at home’ indicates whether or not children were living in the respondent’s household.

Statistical analysis

Data were analysed using linear regression to investigate the association of self-reported depression with the above mentioned factors. In order to identify whether the association of self-reported depression with selected determinants follows the same pattern across native and migrant groups, separate linear regression models were fitted for each of the three groups (Greeks, Albanians and migrants from other countries). Having identified high rates of self-reported depression in Greeks, we run the same regression model by gender to further explore this gender variability in the case of the native population.

To test the robustness of our findings, we adjusted for several covariates related to depression. In the final regression model, all independent variables were included. Reference levels for dummy variables were as follows: ‘male’ for gender, ‘inactive’ for labour market status, ‘tertiary education’ for education level, ‘living comfortably/coping on present income’ for financial strain, ‘never/hardly ever/sometimes’ for family conflict and economic hardship in family while growing up, ‘very good/good/fair’ for self-rated health, ‘no’ for providing unpaid care and ‘yes’ for children living in the respondent’s household.

Results

High risk of depression was reported in all groups, with rates ranging from 26 to 36% (for both genders). Among all citizenship groups, 26.8% of males are at high risk of depression, while the corresponding percentage for females equals to 39.8%. Greeks present higher rates of self-reported depression (36.8%) compared to other groups, with Greek women presenting the highest rates, followed by females from other countries. Astoundingly, more than 43% of Greek women are at high risk of depression (table 1). For both genders, migrants report lower rates of depression than natives.

The regression analysis shows that the determinants of depression vary substantially by ethnic background and gender. Age is the only factor positively and significantly correlated with depression for all groups. Female gender is also strongly associated with depression in the case of both Greeks and Albanians (table 2).

Socioeconomic factors such as low educational level (as opposed to tertiary education) and financial strain are positively associated with self-reported depression only for the native population. In addition, in the case of Greeks, children living at home and social trust are found to be protective factors against depression.

Greeks and migrants from other countries who reported poor health, are at a higher risk of depression, while perceived discrimination seems to be an important determinant for depression in both Greeks and Albanians (P-values <0.045 and 0.001, respectively) (table 2). Childhood experiences are associated in different ways with self-reported depression. Often/always experiencing domestic conflict while growing up is positively associated with depressive symptoms in Greeks, while often/always experiencing economic hardship in childhood is a statistically significant determinant for depression in both migrant groups (Albanians and migrants from other countries) examined.

Regression analysis by gender for the native population illustrates a different mix of factors that are associated with depressive symptoms.

Financial strain, very bad and bad self-reported health and family conflict in childhood are statistically significant factors positively correlated with self-reported depression for both genders. In addition, in the case of Greek males, those being unemployed and having lower secondary or less education are at higher risk of depression. By contrast, the presence of children at home and high levels of social trust reduce the risk of depression among Greek males. Providing unpaid care to family, friends or others is a protective factor against depression for Greek males, but a risk factor for Greek females. Moreover, discrimination is positively correlated with depression in the case of Greek males, with no statistically significant effect for Greek females (table 3).

Discussion

Our results confirm previous evidence that women are more vulnerable to depression than men⁷ and that there is a strong link between depression and socioeconomic position. These findings are consistent with the evidence provided for most Southern and Eastern European countries.^19,35 Several studies reveal that depression among women is approximately twice as prevalent as among men.^10,11,36,37

However, the gender gap in the risk of depression is most pronounced between native women and men and less so between women and men of the two migrant groups. These gender differences in depression can be attributed to a number of gender-specific social risk factors which are associated with the social roles and the social positions of each gender. As Van de Velde et al. (2010) point out, women are faced with an increased risk of depression on account of their conflicting social roles, their lower status levels and their powerlessness and low social esteem.¹⁰ Country-specific gender stratification systems and gendered welfare state regimes also influence the degree to which women are more susceptible to depression than men.¹⁰ Greece is not an exception in this respect. Overtime trends for prevalence rates of depression were found to be twice as high in women as in men³⁸ and are highly correlated with a sharp increase in illicit and legal drug use, especially among adolescents, and a significant increase in suicidal behaviour from the mid-90s onwards.³⁹

Apart from female gender, other determinants such as age, perceived discrimination and bad or very bad self-reported health appear to be more decisive determinants of self-reported depression than is migrant background, as they are significantly associated with self-reported depression for all groups. Recent studies provide evidence as to the increase of health inequalities in times of financial crisis, especially with regard to employment status, age and family type.⁴⁰

Age was ascertained to be positively associated with depression, regardless of citizenship and gender. Contrary to findings from other studies,⁴¹ the elderly population of Greece was found to be more susceptible to experiencing depressive symptoms. This trend could be associated with increased financial distress during the crisis. Favourable socioeconomic status can be a protective factor against the development of late life depression, while economically disadvantaged elders are more likely to experience depressive symptoms.^41–44

Social trust seems to be a protective factor against depression for Greeks. The observed increase in social trust in the early years of the crisis in Greece (2010)^45,46 may have mitigated the mental health effects of the economic crisis by reducing social stress. However, social trust can play a protective role against depression only for individuals in low economic strain and not for those experiencing high financial distress.¹²

Financial strain is positively associated with self-reported depression only in the case of Greeks. Both native males and females who are struggling to make ends meet are more vulnerable to depression, contrary to migrants, who do not seem to be affected by these conditions. Our findings confirm previous evidence as to the long-term effect of the Greek prolonged economic crisis and austerity measures on depressive symptoms among the Greek population.^{20,22,47–49} It appears that the economic crisis can amplify the role of risk factors such as financial strain and health status and weaken the protective impact of other factors such as age.

Education, and specifically, secondary or less, is positively associated with depression only in the case of the native population, while unemployment was found to be a risk factor for depression only for Greek males. Overall, confirming the existing literature,^15,26,30,50 low socioeconomic position seems to increase the risk of depression, compared to high socioeconomic status, especially in the case of Greek males. Confirming previous evidence, our analysis suggests that lower educated and unemployed men are more likely to experience depressive symptoms, especially in the context of crisis.^37,51

Our analysis reflects the mixture of the risk factors for depression in Greek women, including financial strain, bad or very bad health, provision of unpaid care, as well as experiences of domestic conflict or economic hardship in childhood. Our findings are partially consistent with the existing literature that suggests that time spent in primary child care and housework tasks are associated with greater distress for females.⁵² Providing unpaid care, i.e. looking after young children or other members of the family, which remains essentially women’s responsibility in most Southern European countries, does indeed have a negative effect on self-reported depression in Greek females.

Despite the decline of the male breadwinner model and the expansion of women’s employment in Greece in the years preceding the prolonged economic and social crisis, traditional family roles have remained relatively unchanged, with women still bearing the burden of housekeeping and the care of children and the elderly, and thereby juggling multiple roles. Furthermore, the economic crisis in Greece has undermined the capacity of the families to provide their members with financial and moral support as in the past.⁵³ Families with low income are disproportionally burdened by cuts in health and education budgets. These stress-related consequences of economic downturns fuel tensions and stress in the family, triggering depression among its most vulnerable members.³⁵ Apart from gender, the other decisive factors for depressive symptoms appear to be the respondent’s socioeconomic position, namely labour market situation, income and education. Economou et al. (2013) suggest that income affects individuals’ mental health, not only during a period of economic boom but also during recession, for wide strata of the population.¹⁹ This appears contrary to what some scholars believe, that when income loss is suffered by large numbers of people during a recession, the persons affected retain their self-esteem and self-respect, by attributing their loss to events that are universal and beyond their control.⁵⁴

Our analysis does not confirm previous evidence that migrants are at increased risk of depression, especially when they suffer discrimination and social exclusion.²⁶ Male migrants in Greece report similar levels of depressive symptoms as native-born men, whilst women from Albania and other countries report lower rates of depression than do native females. It seems that the native population is less resilient to stressors such as an economic downturn than the migrant population. A possible explanation is that as migrants have already experienced economic hardship in their countries of origin, they are less susceptible to the adverse effects of a sharp deterioration of living standards in the host society and are thus more resilient. Cultural variations in the perception of depression among different ethnic groups^54,55 may also account for this result.

Limitations

Our analysis is limited by the absence of comparable data on depression in Greece for the previous years, thereby preventing us from establishing causal links between depression and the economic crisis.

Funding

This article is part of the project ‘Health Inequalities among Migrant Population (MIGHEAL)’, co-funded by the Financial Mechanism of the European Economic Area (EEA). Programme on Diversity, Inequalities and Social Inclusion, 2009–2014 (Grant number: GR-07: 3807). The Project was led by the National Centre for Social Research, in Athens Greece in collaboration with the Norwegian University for Science and Technology, in Trondheim, Norway and operated by the General Secretariat for Research and Technology, Ministry of Education, Research and Religious Affairs in Greece.

Conflicts of interest: None declared.

References