Identification of items which predict later development of depression in primary health care.

To improve recognition and treatment of depression in primary care it would be advantageous to have criteria for identification of later depression. Only a few studies were performed on primary care samples to examine risk factors for new depressive episodes. These were mostly cross-sectional and did not include psychiatric symptoms as possible predictors. This is the first one-year-follow-up study investigating prospectively psychiatric symptomatology as risk factors for depressive episodes in a primary care sample. An international primary care sample of initially non-depressed subjects (n = 2,445) was examined for the presence of depression (ICD-10) at follow-up. Initial measures addressed presence of psychiatric symptoms according to the CIDI-Primary Health Care Version and sociodemographic variables. Logistic regression analysis was carried out to determine the relationship with the development of new depressive episodes. After one year, 4.4% of primary care patients met ICD-10 criteria for a depressive episode. Logistic regression analysis revealed that symptoms of depression and physical, mainly pain-related symptoms were associated with depression at follow-up. General practitioners therefore should be careful in the consideration of depressive symptoms but also of somatic complaints which might precede depression or mask depressive symptomatology. The present results might be helpful for future development of prediction scales.