Survivors of the war in the Northern Kosovo: violence exposure, risk factors and public health effects of an ethnic conflict

Abstract

Abstract Background The aim of this population-based study was to assess the long-lasting effects of ethnic conflict on health and well-being (with a focus on injury and persistent pain) at family and community level. We have also investigated possible risk factors for victimisation during the conflict and factors contributing to healing. Methods We conducted a district-level cross-sectional cluster survey of 1,115 households with a population of 6,845. Interviews were carried out in Mitrovicë district in Northern Kosovo from September to October 2008, using standardised questionnaire to collect lifetime violence exposure, lifestyle factors and health information on individual and household. Results Ethnic Albanians made up 95% of the sample population. Crude mortality and under-five mortality rate was not high in 2008. Over 90% of families had been exposed to at least two categories of violence and human rights violations, and 493 individuals from 341 families reported torture experiences. During the two weeks before the survey, 20% of individuals had suffered physical or mental pain. There were differences in pain complaints according to gender and age, and whether people had been injured within 12 months, had lifetime exposure to violence-related injury, or had been tortured. Patterns of social and political participation in a family could affect the proportion of family members complaining of pain. The proportion of family members with pain complaints was related to a decline in the household income (coef = 9.31, 95% CI = 6.16-12.46, P Conclusions Mitrovicë district is currently characterised by a low level of violence, but the effects of ethnic conflict on health and well-being have not gone. The level of lifetime exposure to violence, the proportion of family members reporting pain and lifetime violence-related injury, and family's financial burden were found to be inter-correlated. The sample confined to one ethnic group in one district limits the generalizability of the findings.