Mental health of immigrants from the former Soviet Bloc: a future problem for primary health care in the enlarged European Union? A cross-sectional study

Sixteen years after its dissolution, the Soviet Bloc started to be mentioned again in the media when seven of its former members (Poland, Hungary, Slovakia, the Czech Republic, Estonia, Latvia, and Lithuania) were invited to join the European Union (EU) in 2004, and another two (Bulgaria and Romania) in 2007. It is just as incorrect now to call these countries former members of the Soviet Bloc as to call the United States a former British colony [1]. However, the reasons for making this mistake are easy to explain. The isolation from the West by half a century of war and communist rule, the Iron Curtain, and the forced inclusion in the Soviet Bloc with the compulsory creation of the "Soviet-style regimes" [2] have all had a profound influence on the economy, social fabric, and health of the population in these countries and, although diminishing, the consequences are still visible. The former Soviet Bloc countries are worse off economically and the health rates there are lower than in the Western Europe [3]. Consequently, migration from these countries may cause a change in service needs and/or health patterns in the host societies. To prevent or manage such a change successfully, it is important to learn about the health of this immigrant group. The present study attempted to procure knowledge about the self-reported mental health of immigrants from the former Soviet Bloc currently living in Sweden.

Self-reported mental health (defined here as self-reported psychiatric illness and psychosomatic complaints) was the focus of this study because mental health is fundamentally interconnected with physical and social functioning and health outcomes and is therefore crucial to the overall well-being of individuals, societies, and countries [4]. Moreover, in general, self-reported health is a reliable indicator of the health status of a population, predicting, as it does, mortality, morbidity, and other clinical outcomes [5, 6].

The three groups of immigrant respondents differed slightly with respect to the descriptive characteristics (Table 3). The largest difference was seen in the year of immigration to Sweden, with immigrants from Poland arriving during 1970–1989, immigrants from the Eastern Europe during 1950–1969 and 1980–1989, whereas immigrants from the FSU mostly arrived right after the war and after 1990 (Table 3). It is noteworthy that the year of immigration to Sweden reflected the occurrence of major events in the countries of birth of these immigrants (Table 1). However, it did not fully mirror the arrival data presented in Table 2. Since the data in Table 2 and the data used for the analysis in the present study stemmed from the same source (Statistics Sweden), this slight discrepancy might be explained by the fact that the data in Table 2 were collected during one year (2001), and the data used for the analysis in the present study were pooled from eight years (1994–2001).

These results suggest that there is an association between the outcomes and several of the independent variables, the association which could not be explained by other included variables. The association between the self-reporting of both psychiatric illness and psychosomatic complaints and the country of birth was the most profound: the odds ratios for reporting the outcomes were two times higher among persons born in Poland and other East European countries than among the respondents born in Sweden.

This association has been tested further. The FSU-born respondents were treated as a reference group, Swedish-born respondents were excluded (Table 5, Models 1a and 1b), and migration-related variables were added one-by-one into the analyses (Table 5, Models 2a, 2b and 3a, 3b). Language at home did not influence the odds of the outcomes, but years in Sweden (Model 3b) largely decreased the odds ratio of reporting psychosomatic complaints among immigrants from Poland and other East European countries. Thus it can be regarded as a confounder for psychosomatic complaints, but not for psychiatric illness (Model 3a).

When compared between themselves with both additional variables included simultaneously (Table 5, Models 4a and 4b), the immigrant groups showed the same pattern: both the respondents born in Poland and other East European countries had higher odds ratios than the respondents born in the former Soviet Union, irrespective of the outcome. However, the odds ratios were not significantly increased. This can partly be explained by the small sample size in the group of respondents born in the FSU.

To test the main results of the study, we conducted an analysis of hospital admissions based on data for all hospitalizations in the entire nation. We used the Cox regression to calculate the age- and sex-adjusted hazard ratios (HRs) for first hospital admissions due to mental disorders among immigrants from the former Soviet Bloc 25–84 years old, who arrived to Sweden after 1944 and lived in Sweden during 1994–2001. We found that both women and men born in Poland (HR = 1.43; CI = 1.32–1.56), as well as women born in other East European countries (HR = 1.15, CI = 1.01–1.31) had a higher hazard ratio of being hospitalized due to mental disorders than the Swedish-born women and men. However, persons born in the FSU, regardless of their sex, did not have a higher hazard ratio compared to Swedish-born persons (HR = 1.09, CI = 0.96–1.24).

Country of birth showed to have a profound influence on self-reporting of psychiatric illness and psychosomatic complaints among the immigrants from the countries of the former Soviet Bloc in Sweden. Other variables often mentioned in the literature on the health of immigrants from or populations inhabiting the countries of the former Soviet Bloc (i.e. demographic and socioeconomic variables) influenced the self-reported mental health, but could not explain the association between the country of birth and the outcomes.

This is in conformity with the small number of other studies showing that country of birth has a more profound influence on mental health than age [50] and than socioeconomic status [51, 52] among immigrants from some East European countries and Russia in Sweden.

The influence of the country of birth on the self-reported mental health differed among the three immigrant groups in the present study. Immigrants from Poland and other countries of Eastern Europe had poorer self-reported mental health with, in general, twofold higher odds of reporting psychiatric illness and psychosomatic complaints than persons born in Sweden. This could not be explained by the demographic or socioeconomic variables. However, immigrants from the FSU had odds of reporting psychiatric illness and psychosomatic complaints similar to those of the Swedish-born reference group. This finding was also supported by the analysis of the of hospital admissions due to mental disorders.

The finding of higher odds for poor self-reported mental health among immigrants from Poland and other East European countries in the present study agrees with the studies of self-reports and of hospital admissions from Sweden, Britain, and the US and with other cross-sectional and longitudinal studies, as well as studies on war and on political refugees [39–42]. It is difficult to say whether the results depend on premigration experiences in the countries of birth which determined the health of immigrants from these countries or whether the post-migration experiences in Sweden, particularly difficulties of acculturation in the new country usually experienced by immigrants, defined the result. Probably both. However, the aim of this study was not to clarify that matter and the question remains to be answered.

The finding that immigrants from the FSU had similar odds of poor self-reported mental health to those of the host population does not agree with studies conducted elsewhere (the vast majority of the earlier studies have been made in the US and Israel) [9, 10, 38, 43]. A plausible explanation for this discrepancy is that the mental health of the FSU immigrants in Sweden might differ from that of their countrymen in other countries due to different migration trends and various factors in the host countries. Moreover, many other international studies have focused on specific ethnic groups from the FSU (e.g., only Russians or Russian Jews), geographically defined communities, psychiatric patients, or other selected groups that have higher risks of mental disease [9, 33, 53]. In contrast, the present study included immigrants from various FSU republics living anywhere in Sweden. This was ensured by drawing our sample from the national representative sample of permanent Swedish residents recorded in the Register of the Total Population and interviewed in SALLS between 1994 and 2001.

We can only speculate as to why immigrants from the FSU did not have higher odds of poor self-reported mental health than the Swedish-born host population and had lower odds than immigrants from the other former Soviet Bloc countries (nonsignificant). It is possible that persons born in the FSU are afraid of stigmatization, have a more negative attitude toward mental disorders and their treatments (as a legacy of the Soviet abuse of mental diagnoses and medical services for political purposes in the past) and are inclined not to report poor mental health or seek help for mental disorders. The lower odds for psychosomatic complaints among immigrants from the FSU compared to immigrants from Poland might also be explained by the fact that persons born in the FSU are inclined to hide their private feelings [54], to be ashamed of their physical and psychological problems, and tend not to share them with others [55]. However, Hitch and Rack [41] give another explanation, suggesting that weaker mutual support in Polish communities than in Ukrainian ones, social seclusion from the host population and the fellow nationals might partly explain the higher incidence of first admissions to psychiatric hospitals among Polish immigrants than Ukrainian immigrants.

Another possible explanation of the different odds for poor self-reported mental health among immigrants from the former Soviet Bloc is better mental health among persons born in the FSU than among those born in the other former Soviet Bloc countries. According to the World Health Organization (Health for All Database) in 2002, the prevalence of mental disorders in the FSU (2.8%) was lower than in the former Czechoslovakia (3.5%) and Poland (3.2%).

It is very difficult, however, to compare rates of self-reported mental health since the self-administered screening instruments in psychiatry show different cut-off points in different cultures. For example, psychosomatic complaint is a culturally sensitive variable because representatives of every nation have their specific way of somatizing or expressing their ill-health. Inhabitants in or immigrants from different countries communicate their health, particularly mental health, differently [53, 55–57]. The fact that the psychiatric illness variable was self-reported and not clinically validated was the most important limitation of the study. Self-reported health is a reliable indicator of health [5, 6], preterm mortality [58–61], and health care need [62, 63]. However, it has not been confirmed whether self-reported mental health is also a reliable indicator. It is likely that the prevalence of poor self-reported mental health is undeerestimated in the present study since the respondents are generally probably less willing to report a psychiatric disorder than a somatic one.

Nevertheless, the additional testing of this variable by performing an analysis of the hospital admission data for mental disorders in the same population supported the results of the present study, confirming our assumption that self-reported mental health may be a good indicator of mental health. Moreover, self-reported mental health is associated with mortality as self-reported health is [64, 65] and we therefore believe that self-reported mental health is a valuable indicator in population health monitoring.

Another limitation of the present study is that the immigrant groups focused on probably lacked homogeneity not only between themselves with regard to self-reported mental health and other characteristics (including historical backgrounds, religion, language, and geography), but also within themselves. For example, the three Baltic FSU republics have joined the European Union and differ from the other FSU republics in many other respects. Therefore, it is difficult to say whether the results of the present study are fully applicable to immigrants from these now independent countries when considered separately from the whole FSU; thus this applies to the East European countries. Nevertheless, immigrants with different backgrounds had to be combined in larger categories ("Other East European countries" and "The former Soviet Union") to achieve sufficient statistical power and to conduct the analyses in the present study. Poland was categorized separately due to a large number of Polish immigrants in Sweden. The FSU was categorized separately due to differences in the economy and health of the inhabitants of the FSU compared to other East European countries [14].

The limitations of the cross-sectional design of the present study prevented us from tracing any causal pathways between the migration of people from the former Soviet Bloc and self-reported mental health, and we could only speculate about the determining mechanisms behind the reporting of psychiatric illness and psychosomatic complaints. A lack of data on alcohol consumption was also a limitation for a study on self-reported mental health, although an adjustment of outcomes was made for smoking. The questionnaire was not specifically designed for the present study and we had to use the available data rather than collect our own. Additionally, it was not possible to perform a clinical evaluation to test the reliability of the instruments because the design of the national survey does not include a clinical examination or a clinical instrument.

Finally, the sample size, especially of immigrants from the FSU, was small and possibly affected the significance of the analysis. However, in spite of its small size, it included immigrants from the various former Soviet republics, and the proportion of respondents was representative of all 25–84-year-old FSU immigrants who arrived in Sweden after 1944. The weighted mean population from our sample was similar to the average number of persons of the same age and country of birth living in Sweden registered by Statistics Sweden during 1994–2001 (not shown).

The strengths of the extensive SALLS data balanced the limitations. SALLS is performed annually by Statistics Sweden and is designed to obtain information about the Swedish living conditions of the Swedish population. Statistics Sweden is a governmental statistical bureau with a long tradition of carrying out national surveys and analyzing the collected data. The data collected by Statistics Sweden is used in multiple studies and for the national statistical reports. The reliability of the survey questions is high and the nonresponse is rather low. The nonresponse rate in SALLS during 1994–2001 was about 20%, mostly due to refusal to take part. In the earlier samples it was found that those who refused to participate (2/3 of nonrespondents) had the same mortality rate as the respondents. Regretfully, it was impossible to estimate the particular nonresponse among the immigrants in this study because the sample in SALLS is not chosen according to the country of birth.

A more detailed description of the SALLS can be found elsewhere [66].

According to the findings of the present study, the influence of country of birth on the self-reported mental health was profound and could not be explained by other health determinants, such as age, sex, marital status, social network, SES, or smoking.

The self-reported mental health differed, however, among immigrants from the different former Soviet Bloc countries. Unlike the immigrants from the FSU, immigrants from Poland and other East European countries generally had a twofold higher odds ratio for reporting psychiatric illness and psychosomatic complaints than adults born in Sweden after adjusting for demographic and socioeconomic variables. An adjustment for migration-related variables changed it only to the limited extent.

These findings and prior awareness of the health of immigrants from countries focused on in the present study should be acknowledged, particularly with the continuing expansion of the European Union.