Longitudinal study of psychiatric symptoms, disability, mortality, and emigration among Bosnian refugees

Context Evidence is emerging that psychiatric disorders are common in popul ations affected by mass violence. Previously, we found associations among d epression, posttraumatic stress disorder (PTSD), and disability in a Bosnia n refugee cohort. Objective To investigate whether previously observed associations continue over, time and are associated with mortality emigration to another region. Design, Setting, and Participants Three-year follow-up study conducted in 1 999 among 534 adult Bosnian refugees originally living in a refugee camp in Croatia. At follow-up, 376 (70.4%) remained living in the region, 39 (7.3% ) were deceased, 114 (21.3%) had emigrated, and 5 (1%) were lost to follow- up. Those still living in the region and the families of the deceased were reinterviewed (77.7% of the original participants). Main Outcome Measures Depression and PTSD diagnoses, based on Diagnostic an d Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) criteria and measured by the Hopkins Symptom Checklist-25 and the Harvard Trauma Que stionnaire, respectively; disability, measured by the Medical Outcomes Stud y Short-Form 20; and cause of death, determined by family interviews with r eview of death certificates, if available. Results In 1999, 45% of the original respondents who met the DSM-IV criteri a for depression, PTSD, or both continued to have these disorders and 16% o f respondents who were asymptomatic in 1996 developed 1 or both disorders. Forty-six percent of those who initially met disability criteria remained d isabled. Log-linear analysis revealed that disability and psychiatric disor der were related at both times. Male sex, isolation from family, and older age were associated with increased mortality after adjusting for demographi c characteristics, trauma history, and health status (for male sex, adjuste d odds ratio [OR], 2.63; 95% confidence interval [CI], 1.17-5.92; living al one, OR, 2.40; 95% Cf, 1.07-5.38; and each 10-year increase in age, OR, 1.9 1; 95% Cl, 1.34-2.71). Depression was associated with higher mortality in u nadjusted analysis but was not after statistical adjustment (unadjusted OR, 3.12; 95% Cl, 1.55-6.26; adjusted OR, 1.85; 95% Cl, 0.82-4.16). Posttrauma tic stress disorder was not associated with mortality or emigration. Spendi ng less than 12 months in the refugee camp (OR, 11.30 95% CI, 6.55-19.50), experiencing 6 or more trauma events (OR, 3.34 95% CI, 1.89-5.91), having h igher education (OR, 1.90; 95% Cl, 1.10-3.29), and not having an observed h andicap (OR, 0.11; 95% CI, 0.02-0.52) were associated with higher likelihoo d of emigration. Depression was not associated with emigration status. Conclusions Former Bosnian refugees who remained living in the region conti nued to exhibit psychiatric disorder and disability 3 years after initial a ssessment. Social isolation, male sex, and older age were associated with m ortality. Healthier, better educated refugees were more likely to emigrate. Further research is necessary to understand the associations among depress ion, emigration status, and mortality over time.