Background: The aim of this study was to identify internationally relevant risk factors for the persistence of depression in primary care. None of the previous studies examining primary care patients could examine an equivalently large international sample. The findings from the WHO Collaborative Study on 'Psychological Problems in General Health Care' might be generalised to different cultural environments.
Methods: A primary care sample of depressed subjects (ICD-10) (n=725) from 15 centres from all over the world was reinvestigated for the presence of depression after 12 months. Logistic regression analysis was used to determine the relationship of sociodemographic variables, characteristics of mental illness, and social disability with depression at follow-up.
Results: 33.5% of the subjects met ICD-10 criteria for a current depressive episode at the 12-month follow-up. Logistic regression analysis revealed that years of formal education, unemployment, depression severity, antidepressant use, abdominal pain as the main reason for consulting the general practitioner, and 'Repeated suicidal thoughts' were related to depression after 1 year.
Limitations: We have only poor information about details of the actual course of patients between the two assessments.
Conclusions: Our findings of significant variables are generally comparable to results of previously conducted studies. Sociodemographic and disease-specific variables play an important role in the prediction of depression at the 12-month follow-up. General practitioners must be careful in the consideration of these variables but also of individual patient characteristics.