Depression and suicidal behavior in Uganda : validating the response inventory for stressful life events (RISLE)

Abstract

Background: Depression and suicide ideation are prevalent in the general population but their recognition and detection in primary care is problematic. The present study investigated if the RISLE can be used to detect depressed and or suicidal individuals in the general population.

Methods of study: Members of the general population in two districts of Uganda, Adjumani and Bugiri districts of Uganda, and fresh students at the Faculty of Medicine, Makerere University, participated in the study. Two pilot studies were conducted before the collection of data, namely: at Makerere University among fresh students sampled from all faculties in 2001, and in each of the districts in the course of interviewer and research assistants training in 2002. . The ethical committee of the Faculty of Medicine, and Uganda National Council for Science and Technology granted ethical clearance. The Dean of Students and the health and civic leaderships of the respective study sites granted further permission for the study.

Analysis: Data analysis comprised of general descriptive analysis, correlation and reliability analyses, principal component analysis to determine the main factors that make up the RISLE, discriminant function analysis to determine the main items of the RISLE that together distinguish depressed and or suicidal individuals in the general population who are not depressed and or suicidal, the construction of receiver operating characteristic curves to determine sensitivity and specificity of the RISLE, determination of sensitivity, specificity, predictive values, and likelihood ratios, and the determination of Cohen s kappa values at several cut-off points to determine the level of agreement between the RISLE and clinical interview method as the gold standard in the study

Results: Results revealed that cut-off points varied according to the nature of population studied. Cut-off point for the population in Adjumani district was 10 for any current psychiatric disorder, and 6 for any psychiatric disorder among students. Sensitivity and specificity of the RISLE at cut-off point 10 and 6 were 74.6% and 77.1%, and 88.1% and 60.4% respectively. Positive predictive values for current psychiatric disorder were 82.0% and 75.6% at cut-off points of 10 and 6 respectively. Agreement between the RISLE and clinical interview was 0.508 at cut-off point 10, followed by 0.501 at cut-off points 6 and 12.

Conclusions: The results of the present study show that the RISLE can be used in conjunction with clinical interview method in the detection and confirmation of individuals with current psychiatric illness and suicidal feelings in the general population.