Risk factors for depression in the survivors of the 1988 earthquake in Armenia.

Most studies of psychopathology following disasters are concerned with posttraumatic stress disorder (PTSD). The present analyses sought to assess the rate and determinants of depression in adult survivors of the 1988 earthquake in Armenia. Unlike previous studies of earthquakes, the present analyses derive from a well-defined cohort of survivors who underwent diagnostic interviewing to characterize psychiatric morbidity. As part of a cohort study of 32,743 survivors of the 1988 earthquake in Armenia, a stratified population sample of 1,785 persons was interviewed about 2 years following the disaster using a special questionnaire based on the National Institute of Mental Health (NIMH) Disaster Interview Schedule/Disaster Supplement. 52% met the criteria for major depression. Of these, a total of 177 cases of depression with no other psychiatric diagnosis or comorbidity were compared with 583 controls from the same interviewed group who did not fulfill the criteria for any psychiatric disorder. Cases and controls were compared as to data obtained independently at the aftermath of the disaster on a number of exposures and characteristics related to the earthquake. More of the cases involved females (odds ratio [OR] for males 0.7 [95% confidence interval [CI] 0.5-0.9]) and from the city of Gumri, which had some of the worst destruction (OR for residents of Gumri 5.9 [95% CI 4.0-8.8]). Being with someone in the same building at the moment of the earthquake was protective for depression (OR for presence of other people 0.5 [95% CI 0.3-0.6]), and the risk of depression increased with the amount of loss that the family sustained as a result of the earthquake (OR for highest level of loss 2.5 [95% CI 1.3-4.8]). The use of alcohol was protective for depression (OR for those who drink 0.5 [95% CI 0.3-0.8]). In various models of multivariate adjustment and analysis, the increased risk of depression with loss, geographic location, and female gender was maintained. Also, being with someone during the disaster, receiving assistance and support after the earthquake, and alcohol use were protective for depression in these multivariate analyses. Depression is a common sequel to an earthquake. As with our previous study of PTSD, we were able to relate intensity of the disaster and loss to the risk of depression in a general population sample. The role of social support during and after the disaster as a protective mechanism against adverse psychological outcome was highlighted again. Whereas alcohol use in our previous study was not related to PTSD outcome, it is noteworthy that in the present analyses it emerged as a protective factor for depression.