Background

The province of Kosovo is located in southeastern Europe in a region that is commonly referred to as the Balkans. It consists of over 90% Albanians and 5.3 percent Serbs and other minorities. Ongoing ethnic conflict between its Albanian and Serb populations flared again at the end of the 1990s, with fighting only being ended as a result of NATO intervention in 1999. As a result of the fighting, many of its inhabitants were killed or displaced. Since the end of the war although the everyday situation has changed, the human security of people in Kosovo still remains precarious even today, more than a decade after the war ended. Ongoing tension between the Kosovar Albanians and Serbians is especially acute in the ethnically divided municipality of Mitrovica, as evidenced by the recent incidents there such as the violent clashes after Kosovo's declaration of independence in 2008, and an outbreak of violence in 2010 after the International Court of Justice ruled that Kosovo's declaration of independence was not illegal.

The events of recent years in Kosovo have generated much research concerning the ongoing effects of war in this post-conflict setting. One specific field of research has focused on how conflict and ongoing social tension has impacted on the psychological wellbeing of its inhabitants in terms of their mental health. However, there has been little attempt to go beyond merely describing the prevalence of mental health problems and investigate other factors not directly related to the conflict which might actually be important in terms of mental health outcomes. Previous studies in other settings have indicated that there are a number of factors which may be important in this respect. Specifically, it has been suggested that the possession of social capital might be important in terms of mental health outcomes. Social capital has been conceptualized as the investment in social relations through which individuals gain access to embedded resources. Networks can act as a source of emotional support as well as promote positive psychological states such as a sense of self-worth and belonging. Network relationships may also impact on mental wellbeing in terms of their potential to affect stress by acting as stress-buffering mechanisms.

Objectives

Although a number of previous studies have reported an association between social capital and mental health there is a dearth of information from post-conflict settings on this association. Moreover, there has been little previous research on the relation between ethnicity, social capital and mental health. The aim of this study was therefore to investigate the association between social capital and mental health among the different ethnic groups in Mitrovica, Kosovo.

Method

The survey was conducted by UNDP Kosovo with the aim of exploring public opinions on the overall political and economic situation of Kosovo. Using a three-stage stratified representative sample, 1239 respondents aged 16 and above were recruited into the study in the North and South of the city. The final response rate was 80.1%.

Following the lead of a recent review study which has suggested that there are many demographic and socio-economic factors associated with health outcomes in conflict-affected populations in low- and middle-income countries, data were collected on a number of phenomena. Besides information on the age and sex of respondents, details were also collected on their marital status (single, married, divorced/separated), highest educational level (primary school, secondary school or university), municipality of residence (south or north), and level of household wealth. Principal component analysis (PCA) was used to generate a wealth index based on the household. Information was also obtained on the social capital of the respondents. We asked the interviewees how frequently they had contact with strangers, friends, and relatives with the following questions: 1) "How often do you meet with people that you do not know particularly well?", 2) "How often do you meet a friend as a leisure activity?", 3) "How often do you meet your relatives as a leisure activity?". The answer options were often, sometimes or never. There were two main health outcome variables. The Hospital Anxiety and Depression (HADS) questionnaire was used to identify individuals with depression and anxiety. The scale consists of 14 questions with scores ranging from 0 to 3 for each question, with two subscales for anxiety and depression. Scores were summed for each subscale and a cut-off point of >=8 was used to identify cases of anxiety and depression. The Cronbach α of the HADS scale in this study was 0.67 (anxiety) and 0.76 (depression) for the version in Serbian. The corresponding figures for the version in Albanian were 0.72 and 0.66.

The analyses on the association between social capital and the two health outcomes were adjusted for sex, age, education, wealth, marital status and municipality. Analyses on Serbians were not adjusted for municipality as all resided in the north. All regression analyses were adjusted for the clustering effect within primary sampling units. A p-value<0.05 was considered statistically significant.

Results

The analytical sample consisted of a total of 994 respondents (497 Albanians, 300 Serbs and 197 other minorities) aged 16 years or above. The Northern part of the municipality consisted of 62.0% Serbians, 15.5% Albanians and 22.5% minorities, while the Southern municipality consisted of 82.9% Albanians and 17.1% minorities. In relation to 'Minority' populations, while Bosniaks (49.5%) and Roma (45.0%) constituted the majority share of the minorities in the North, in the South, the Ashkalis (41.4%), Turks (31.0%) and Egyptians (19.5%) were the minority groups which had the largest population distributions.

The mean (SD) age of the respondents was 37(16), 39(17) and 40(15) years among the Albanians, Serbians and minorities respectively. Minorities had a much lower level of education and wealth, and were more likely to be married compared to the other two ethnic groups. The prevalence of depression was high and ranged from 50.7% (Serbian) to 63.8% (minorities) although the differences between ethnic groups were not statistically significant. In regard to anxiety, there was a similar high prevalence ranging from 50.2% (Serbian) to 73.3% (Albanian) although in this instance, inter-ethnic differences were significant. In terms of social capital, characteristics were similar among the three groups for contact with relatives but Albanians had much lower levels of contact with friends and had a different pattern in their frequency of contact with strangers compared to the other two ethnicities.

In regards to depression, never having contact with strangers was associated with a 2.09 (95% CI: 1.09-4.01), 3.48 (95% CI: 2.47-4.91), 13.17 (95% CI: 3.57-48.54) times higher risk of depression compared to those having frequent contact among the Albanians, Serbians and minorities respectively. In addition, less contact with friends was associated with a 1.76 (95% CI: 1.02-3.03) and 2.68 (95% CI: 1.71-4.20) times higher risk of depression among the Albanians and Serbians respectively. Serbians with less contact with relatives were 1.97 (95% CI: 1.42-2.72) times more likely to have depression, but a reverse trend was observed among the Albanians where less contact with relatives was associated with a significant 52% lower risk of depression. Finally, in terms of anxiety, those who never have contact with strangers were 2.54 (95% CI: 1.39-4.63) and 4.67 (95% CI: 1.29-16.96) times more likely to have depression compared to those with frequent contact among the Albanians and minorities but no significant association was observed among the Serbians. Less contact with friends was associated with a 3.27 (95% CI: 1.97-5.43) and 1.82 (95% CI: 1.02-3.24) times higher risk of anxiety among the Albanians and Serbians respectively with no significant relationships observed among the minorities. Frequency of contact with relatives was not significantly associated with anxiety in any of the ethnic groups.

Discussion

Our results suggest that social capital is associated with mental health outcomes in Mitrovica, Kosovo. This finding is especially important in public health terms given the range of mental health challenges that the population faces in contemporary Kosovo. Having lower levels of all three measures of social capital was associated with higher levels of depression and anxiety among all ethnic groups with one main exception: a significant opposite trend was observed among the Albanians where more frequent contact with relatives was associated with higher levels of depression.

The relation we observed between social capital and mental health is even more important given a second main finding of this study - that social capital is not a homogeneous phenomenon but is distributed unevenly between both individuals and groups. As yet, there has been comparatively little research undertaken on the relation between ethnicity and social capital worldwide although our results do mirror those from several previous studies which suggest ethnic differences may exist in the possession of social capital. In the current study these differences may in part, be underpinned by the nature of the traditional extended family structure in this region as well as other cultural differences. However, whatever their source, these differences in access to network structures may be having an important effect in terms of mental wellbeing.

The exceptionally high rates of depression and anxiety recorded in the current study may have several factors underpinning them. Specifically, it has been suggested that since 2009 the security situation in Mitrovica has deteriorated with an increase in outbreaks of inter-community violence. Besides impacting on mental wellbeing this upsurge in ethnic conflict may also be acting to exacerbate a range of everyday stressors (i.e. factors that cause stress which emerge from everyday life) which research in other contexts and Kosovo itself has indicated can affect mental health. It has been suggested for example, that factors such as poverty, unemployment, and a lack of access to medical care may all be linked to worse mental health.

Social capital may be important for mental health outcomes due to the fact that one's location in a network may increase the chance of accessing various forms of support. This is likely to be especially important in Kosovo, where the extended family is traditionally the site where mental health problems are dealt with. Such family networks might not only be important in terms of the treatment of mental ill health but these family social networks and the support resources they can provide might also be important as regards preventing the onset of mental illness. In addition, this study also found that access to 'bridging' social capital i.e. contact with unknown others (strangers) was also beneficial for mental health. Such bridging ties might be useful for accessing new and different forms of information which might result in an improvement in individuals' ability to solve various forms of problem. In a location such as Mitrovica for example, where ethnic conflict has impacted on public health provision this might involve gaining access to essential mental health services.

This study also highlighted the dark side of social capital however, as those Albanians who had more frequent contact with relatives had worse mental health. It is possible that this could be related to the more traditional nature of Albanian family life and family structure which might be particularly burdensome for women for example in terms of having to deal with other family members' problems and stress.

Conclusion

In the presence of extremely high levels of mental ill health across all ethnic groups, our findings have shown that social capital is associated with mental health outcomes. For most groups social capital seems to have a beneficial effect, though the current study has also highlighted the dark side of social capital. Although the causes of mental ill health are numerous and multifaceted and interventions to improve mental health can take many forms, the findings from this study suggest that focusing on the role of social capital and how it can be fostered within and between networks and communities may offer a promising path in the future when it comes to addressing the issue of mental wellbeing in this region.

ソーシャルキャピタルとメンタルヘルスの関連性を東ヨーロッパ、とくに紛争後の地域で調査したものは少ない。本研究は、1999年のコソボ紛争後11年たってもなお、民族間で北と南に分断されている都市ミトロビッツァにおいて、アルバニア人、セルビア人、その他の少数民族のソーシャルキャピタルとメンタルヘルスとの関連を検証する事を目的として行い、下記の結果を得た。

1. 研究デザインはフィールドベースの横断研究とした。研究参加者は、調査地ミトロビッツァで2010年11月に国連開発計画コソボ事務所の調査に任意で参加した16歳以上の市民である。1239名の対象者中、80.1%の994名(アルバニア人497名セルビア人300名、マイノリティ197名)の市民からの回答を得た。メンタルヘルスの測定にはHADSを使用し、ソーシャルキャピタルは親戚、友人、他人とのコンタクトの頻度によって測定した。

2. 今回の調査において鬱傾向を示した者の割合は50.7% (セルビア人)から 63.8% (少数民族)、不安傾向の割合は52.2%(セルビア人)から73.3%(アルバニア人)であり、不安傾向の民族間差は統計学的に有意であった。

アルバニア人、セルビア人、少数民族の中で、他人とのコンタクトが全くない者は、よりコンタクトを取っている者と比べて、2.09倍 [95%信頼区間(CI): 1.09-4.01], 3.48倍 (95% CI: 2.47-4.91)、13.17倍 (95% CI: 3.57-48.54) 高く、鬱傾向に関連していた。

加えて、少ない頻度で友人とコンタクトを取る者はアルバニア人、セルビア人で1.76倍 (95% CI: 1.02-3.03)と 2.68倍 (95% CI: 1.71-4.20) 高く、鬱傾向に関連していた。セルビア人で親族とのコンタクトが少ない者は 1.97倍 (95% CI: 1.42-2.72) 鬱傾向が高いが、アルバニア人については、より親戚とのコンタクトが少ない者ほど鬱傾向が52%低くなっていた。

不安傾向においては、他人とコンタクトを取った事の無い者はアルバニア人で2.54倍 (95% CI: 1.39-4.63)、マイノリティで4.67 倍(95% CI: 1.29-16.96)リスクが高かった。友人とコンタクトをとる頻度が少ない者はアルバニア人で3.27倍(95% CI: 1.97-5.43)、セルビア人で1.82倍 (95% CI: 1.02-3.24) 高く不安傾向を示していた。

3. 本調査地域において鬱傾向と不安傾向は高く、ソーシャルキャピタルは民族間で差異があった。また、本調査はミトロビッツァにおいてソーシャルキャピタルとメンタルヘルスとの間には関連がある事を示した。多くの場合、親戚、友人、他人とのコンタクトが少ない者ほど鬱傾向と不安傾向は高くなっているが、アルバニア人は親族とのコンタクトが少ない者ほど鬱傾向のリスクは低かった。

紛争後、まだ緊迫した状況が続くコソボ・ミロビッツァにおいて、自由にコソボ内を移動できず、十分な医療行為や医療物資が得られない市民が多い中でメンタルヘルスを改善するためには、民族差とソーシャルキャピタルの役割を考慮する事が重要であることが示唆された。

本研究はコソボやその他の旧ユーゴ諸国でソーシャルキャピタルとメンタルヘルスの重要性を意識させる事に大きく貢献すると期待でき、学位の授与に値するものと考えられる。