Measuring community mental health in developing societies: evaluation of a checklist format in Nepal

Background: There is growing recognition of the importance of mental health problems in developing countries. In large part, however, we have very limited epidemiological data at national and/or community levels about the prevalence of mental illnesses. Aims: The purpose of this paper is to describe the reliability and validity characteristics of an assessment tool that may be useful for conducting community-level surveys (particularly in rural communities of developing countries) to obtain prevalence rates of mental illnesses. Methods: We used a sample of adults residing in a rural village in Nepal to assess disorders with a modified version of the DSM-III-R Checklist. We evaluated construct validity, scale reliability, convergent validity and discriminant validity. Results: There is strong evidence for the construct validity of generalized anxiety and depression in our sample. By contrast, the symptoms associated with mania and schizophrenia were not empirically distinct. Convergent validity is acceptable. As a test of validity characteristics, the pattern of sociodemographic correlations suggests that the specific social origins of disorder in Nepal will require further investigation. Conclusion: The first step in obtaining high quality information on the distribution of mental illness in developing countries is to establish some reliable and valid indicators of disorder. The checklist format for assessing disorder appears to meet this objective and offers the possibility that community-level prevalence studies can be reasonably conducted.