Research Article - (2016) Volume 13, Issue 3
Henrique Pereira*, Hugo Amaral, Samuel Monteiro, Graca Esgalhado, Rosa Marina Afonso and Manuel Loureiro
Department of Psychology and Education, University of Beira Interior, Estrada do Sineiro, s/n, 6200 Covilha – Portugal
Corresponding Author:
Dr.Henrique Pereira, Department of Psychology and Education, University of Beira Interior, Estrada do Sineiro, s/n, 6200 Covilha – Portugal, E-mail: hpereira@ubi.pt
Background: The Roma population, one of the largest minority groups in Europe, experience discrimination and stigma associated with marginalized social position. Few studies have examined mental illnesses in the Roma, and none have examined the Roma unemployment beneficiaries. The present study estimates mental health among Roma unemployment beneficiaries in comparison to non-Roma beneficiaries in Portugal. Design: A cross-sectional study was carried out. Two clinical psychologists working as research assistants surveyed unemployed individuals looking for jobs at employment agencies. The sample consists of 71 unemployed individuals, beneficiaries of unemployment insurance, and registered in the National Institute of Employment and Professional Training in Portugal. 43.7% were Roma (31) and 56.3% (40) non-Roma. Mean age was 39.93 years old (ranging from 18 to 65 years old; SD = 11.96). Demographics variables were ethnicity, age, gender, marital status, education, and time of unemployment in years. Mental health was assessed using the Brief Symptom Inventory (BSI). Results: Results for the comparison of the two samples indicate significant differences for several dimensions of mental health that are more severe for non-Roma participants, namely, Obsessive-Compulsiveness, Interpersonal Sensitivity, Depression, Anxiety, Paranoid Ideation, Psychoticism, and Global Severity. Conclusions: When compared to participants of specific groups of the general population in the same challenging crisis and conjuncture such as unemployment, the Roma participants appear to have specific resilience processes, or some cultural and ethnic specificities when coping with mental health/ disease factors, as evidenced in the lower distress and lower levels of psychiatric disorders when compared to unemployed non-Roma individuals. As such, the unemployment condition seems to have a much more negative impact on the mental health of non-Roma participants than in the mental health of Roma participants.
Keywords
Roma, Non-Roma, Mental Health, Unemployment, Portugal
Background
There has long been a consensus that compared with the non-Roma population in Europe, Roma have poorer health, and this is closely linked to social determinants of health (European Commission, 2014). Although little is known about the mental health of the Roma people, their life expectancy is lower than the general population, perinatal death rates are higher, and studies on mental health suggest higher rates of mental disorders and suicide (Goward et al. 2006; Scott, 2010).
Roma people have lived and worked in Portugal for over 500 years, and there are an estimated 90.000 to100.000 individuals at the moment (Mendes, Magano, and Candeias 2014). They have a distinct culture and set of beliefs, including identifiable beliefs and attitudes to health. There is a strong emphasis on self-sufficiency, and against seeking help from outsiders, since many of them are mistrustful of ‘the authorities’ e.g. police, local councils, and other statutory bodies including health and social care services. There is also stoicism about poor health and pain, and a belief that these things must simply be tolerated. Another feature is a fatalistic and nihilistic attitude to health ('what will be, will be'), resulting in a view that illness is inevitable, and therefore seeking treatment is pointless (Monteiro et al. 2013).
A high proportion of Roma have very poor levels of literacy, and children have been found to have the lowest educational attainment of any minority ethnic group in Europe (Gyukits et al. 2006). Along with adverse social experiences and distressing perceptions of hostility and racism, Roma people produce coherent cultural beliefs and attitudes that underpin health related behaviors, combining low expectations, normalization of ill health, stoicism, fatalism, ignorance and fear (Van Cleemput et al. 2007).
Research comparing Roma to non-Roma participants has been conducted on several levels of health, emphasizing that socioeconomic status and ethnicity was associated with lower health outcomes, including mental health problems (Kolarcik et al. 2009; Lee et al. 2014), but published research on the mental health needs of the Roma population is sparse mental health indicators being one of the least studied (Hajioff and McKee, 2000).
Discrimination towards the Roma population has been levied on the basis of their ethnicity and culture, and like in most other European countries, this has rendered them one of the most socially excluded and marginalized groups in Portugal on several levels (social integration, education, social participation, health outcomes), creating a very unfavorable socioeconomic situation: only 10% are employed; 50% live on governmental benefits and 40% depend on family businesses (Mendes, Magano, and Candeias 2014).
In fact, unemployment rates in Portugal have increased over the last few years due to economic crisis, and is presently at 12.4% (Eurostat, 2016), affecting both Roma and non- Roma populations, especially on a mental health level. The relationship between unemployment and poor health has been well documented. The unemployed tend to have higher levels of impaired mental health including depression, anxiety, and stress, as well as higher levels of mental health hospital admissions, chronic disease and premature mortality (Dean and Wilson, 2009; O’Campo, Eaton, and Muntaner, 2004).
In Portugal, the National Institute of Employment and Professional Training is responsible for providing support to any employment seeker and although registering within this institution is optional, if one wants to receive social benefits such as professional training or monthly allowances, registration is mandatory.
Research has established ethnicity (Roma or non-Roma) and unemployment status as social determinants of mental health. However, few studies have analyzed mental health disparities among these two groups when unemployed and benefiting from the unemployment insurance, paid by the Government.
Methods
Sample
The sample analyzed for the current study consists of 71 unemployed individuals, beneficiaries of unemployment insurance, and registered in the National Institute of Employment and Professional Training in Portugal. 43.7% were Roma (31) and 56.3% (40) non-Roma. Participants were contacted directly in the waiting rooms of four different job agencies in the region of Beira Interior, mainly a rural area in central Portugal.
Mean age was 39.93 years old (ranging from 18 to 65 years old; SD = 11.96) and in average they had been unemployed for two years (ranging from “never worked” = 0 years to 6 years; SD = 1.27). Also, 45.1% were female and the vast majority had very little education (7% with no studies, 33.8% with some primary education, and 54.9% with up to 6 years of education). Finally, 50.7% said they were single, and 42.3% married.
Measures
Demographics variables: evaluated through direct questions in the survey, after the informed consent letter and before presenting the other scale. Participants were asked for their ethnicity, age, gender, marital status, education, and time of unemployment in years.
The Brief Symptom Inventory (BSI): this is 53-item selfreport measure of psychological distress in the prior week. The BSI (Derogatis, 1993) has been widely used as a psychiatric screening tool in clinical settings and epidemiological studies. The items are rated on a 5-point Likert scale with response options ranging from not at all to extremely. The items yield nine subscales: Somatization, Obsessive-Compulsiveness, Interpersonal Sensitivity, Depression, Anxiety, Hostility, Phobic Anxiety, Paranoid Ideation, and Psychoticism. Also, an overall measure of mental health problems called the Global Severity Index (GSI) can be calculated. The Portuguese version of the BSI (Canavarro, 1999) was used in this study, since this is an efficient indicator of mental health symptoms, and its reliability and validity have been well-established. In this study, internal consistency reliabilities (Cronbach's alpha) for the nine subscales ranged from .82 to .91, findings consistent with available research on the BSI.
Procedure
A cross-sectional study was carried out, collecting information for two months. Two clinical psychologists working as research assistants asked unemployed individuals looking for jobs at the employment agencies to answer to the questionnaires in the same order as they were described above in this article. The research assistants explained each participant the nature of the study and that their participation was completely voluntary and non-related to the services of the institutions where they were contacted.
Statistical analysis
Analyses were done separately by ethnicity (Roma and Non- Roma). First, the distribution of socio-demographic variables was assessed by using chi-square. Non parametric tests were used to compare differences between the two groups, namely the Mann-Whitney U Test.
Results
The socio-demographic characteristics of the samples from both Roma and non-Roma unemployed participants are presented in Table 1. There were significant differences in gender, age, marital status and time of unemployment. The Roma sample had more male, younger, and single participants, as well as less time of unemployment when compared to the non-Roma sample (Table 1).
| Roma % (n) |
Non-Roma % (n) |
p-value | ||
|---|---|---|---|---|
| Gender | Female | 14.1% (10) | 31% (22) | .047* |
| Male | 29.6% (21) | 25.4% (18) | ||
| Age | 18-30 | 21.1% (15) | 8.5% (6) | .004* |
| 31-45 | 14.1% (10) | 19.7% (14) | ||
| 46-65 | 8.5% (6) | 28.2% (20) | ||
| Marital status | Single | 31% (22) | 19.7% (14) | .002* |
| Married | 9.9% (7) | 32.4% (23) | ||
| Civil Union | 2.8% (2) | 0% (0) | ||
| Divorced | 0% (0) | 4.2% (3) | ||
| Education | No education | 4.2 (3%) | 2.8% (2) | .094 |
| Some primary | 14.1% (10) | 7% (5) | ||
| Up to 4 years | 7% (5) | 5.6% (4) | ||
| Up to 6 years | 15.5% (11) | 39.4% (28) | ||
| Up to 9 years | 1.4% (1) | 1.4% (1) | ||
| Secondary | 1.4% (1) | 0 (0%) | ||
| Time of unemployment | Less than 1 year or N/A | 15.9% (11) | 1.4% (1) | .005* |
| 1 year | 7.2% (5) | 17.4% (12) | ||
| 2 years | 10.1% (7) | 26.1% (18) | ||
| 3 years | 5.8% (4) | 2.9% (2) | ||
| 4 years | 4.3% (3) | 5.8% (4) | ||
| 5 years | 0% (0) | 1.4% (1) | ||
| 6 years | 1.4% (1) | 0% (0) |
*p<.05; **p<.001
Table 1: Socio-demographic factors between Roma and non-Roma unemployed (N=71).
Results for the comparison of our two samples indicate significant differences for several dimensions of mental health that are more severe for non-Roma participants, namely, Obsessive-Compulsiveness, Interpersonal Sensitivity, Depression, Anxiety, Paranoid Ideation, Psychoticism, and Global Severity (Table 2). In fact, for the Global Severity Index, non-Roma participants scored above the normal Portuguese population reference levels (.83) (Canavarro, 1999), which was also the case for Roma participants, but well into the severity level (1.40 for clinical Portuguese population) (Table 2).
| BSI -Individuals not emotionally disturbed Mean (SD) | Roma Mean (SD) |
Non-Roma Mean (SD) |
p-value | |
|---|---|---|---|---|
| Somatization | 0.573 (0.916) | 0.89 (0.49) | 1.02 (0.50) | .279 |
| Obsessive-Compulsiveness | 1.290 (0.878) | 1.47 (0.69) | 1.80 (0.51) | .022* |
| Interpersonal Sensitivity | 0.958 (0.727) | 1.15 (0.58) | 1.53 (0.65) | .013* |
| Depression | 0.893 (0.722) | 0.93 (0.56) | 1.35 (0.65) | .005* |
| Anxiety | 0.942 (0.766) | 0.91 (0.56) | 1.51 (1.29) | .019* |
| Hostility | 0.894 (0.784) | 1.27 (0.58) | 1.42 (0.66) | .303 |
| Phobic Anxiety | 0.418 (0.663) | 0.69 (0.54) | 0.99 (0.84) | .089 |
| Paranoid Ideation | 1.063 (0.789) | 1.52 (0.66) | 1.92 (0.69) | .015* |
| Psychoticism | 0.668 (0.614) | 0.85 (0.52) | 1.40 (0.58) | .000** |
| Global Severity Index | 0.480 (1.430) | 1.07 (0.43) | 1.43 (0.47) | .002* |
*p<.05; **p<.001
Table 2: BSI – Comparison between Roma and Non-Roma samples with Portuguese Validation Referential for Individuals not emotionally disturbed.
Discussion
This was the first study of mental health outcomes among Roma and non-Roma unemployed beneficiaries in Portugal, thus revealing the lack of research and scientific knowledge about the mental health of the largest social minority in Europe. Roma ethnic minority, as mentioned before, is commonly perceived as a disadvantaged social group in many countries (Gyukits et al. 2006), as well as in Portugal. There has long been a consensus that compared to the non-Roma population, Roma have poorer (general) health indicators and this could be closely linked to health social determinants, such as unemployment.
While it is widely believed that the health of Roma people is often poorer than the health of the main stream population, these inequalities remain largely un-researched and under-studied from a comprehensive and epidemiological perspective (Zeman, Depken, and Senchina, 2003). Research on the mental health of Roma people suggest high levels of problems including: functional disorders, depression, neurotic difficulties and suicide when compared the general population, but little is known yet about this population mental health (Goward et al. 2006). In this framework, this study main aim was to estimate mental health indicators among Roma unemployment beneficiaries in comparison to non-Roma beneficiaries, in a Portuguese sample.
The analysis of two comparing groups, distributed by participants’ ethnicity (e.g., Lee et al. 2014; Vorvolakos et al. 2012), that shares the same unemployment condition and the fact of being beneficiary of reintegration support measures, allowed to observe some statistically significant differences in terms of socio-demographic variables that reflected, in part, some social differences between the two communities. In terms of the age, Roma sample showed younger participants, with 35.2% of the sample having less than 45 years of age, compared to 28.2% for the group of non-Roma in this age group. Regarding gender, the Roma sample presents more male participants than female, and in the non-Roma sample we found the opposite. This is a demonstration of how gender differences take place when accessing employment and employment services.
In regards to the marital status, the Roma sample does not have any divorced participants and 31% of respondents were married. In the case of non-Roma 32.4% of the participants said that were married and 4.2% divorced. At the formal education level, no statistically significant differences were found, allowing to establish a comparative analysis between participants of two distinctive ethnic groups.
The results for the Brief Symptom Inventory (BSI) among the two samples showed higher scores for all dimensions than the BSI Portuguese reference values for individuals not emotionally disturbed. These results mean that mental health in the two groups is exposed to the same unemployment conjuncture, and members of both communities diverge from the general population pattern, regardless of ethnicity.
Comparison of BSI results in terms of different health dimensions between groups, allowed us to identify no differences in three dimensions (somatization; hostility; and phobic anxiety). Differences were detected in six particular dimensions (obsessive-compulsiveness, interpersonal sensitivity, depression, anxiety, paranoid ideation, psychoticism) and in the Global Severity Index. The mean value obtained in each dimension, was always higher in non-Roma sample than in the Roma sample. These results suggest that there are mental health factors of psychological distress and psychiatric disorders that seem more favorable to the Roma than to the non-Roma individuals.
This last topic could be highlighted as the most important result of this study because it challenges the evidences of other studies, and will help to demystify the well-established trend of Roma population as having (always) more precarious indicators of (general and mental) health. The expected tendency was also verified in this research but only when we compare Roma data to the general population scores. However, this is not the case, when we compare specific groups of Roma and non-Roma in similar specific unemployment conditions. This study warns about the need and importance of developing comparison research between Roma and smaller groups of non-Roma when they are facing the same adverse social circumstances (such as unemployment). This seems a step forward that draws attention to the research limits and biased analysis of merely comparing minority’s data with main stream population data, not considering some alike circumstances.
As other research focused in Roma and unemployment populations this exploratory study also encountered a number of limitations, some of them concerned with the sampling procedure, participant inclusion and its equivalence and population representation. The future studies should be specifically designed to collect a larger and more diverse sample and collect more information about covariates on the living conditions, socioeconomic status and health-related behaviors.
In conclusion, when compared to participants of specific groups of the general population in the same challenging crisis and conjuncture of unemployment, the Roma appear to have specific resilience processes, or some cultural and ethnic specificities when coping with mental health/disease factors, as evidenced in the lower distress and lower levels of psychiatric disorders when compared to unemployed non-Roma. As such, the unemployment condition (with the existence of a financial benefit) seems to have a much more negative impact on the mental health of non-Roma participants than in the mental health of Roma participants, and this kind of situation could hypothetically be seen as not so severe or damaging than other (living) alternatives.
