A primary healthcare screening tool to identify depression and anxiety disorders among people with epilepsy in Zambia

Abstract

Among the 50 million people with epilepsy (PWE) worldwide, ~15 to 60% also likely suffer from depression and/or anxiety disorders, and 80% reside in low-income regions where these comorbidities are often underrecognized and undertreated. We developed a 10-item screening tool for the detection of depression and anxiety disorders for use in Zambian primary care clinics where the baseline detection rate of depression and/or anxiety disorders among PWE is ~1%. Consenting adults (n=595) completed the screening tool, and 53.7% screened positive. The screen was validated by a psychiatric clinical officer using DMS-IV criteria. Cronbach's alpha was 0.77 overall and 0.67 and 0.57 for the depression and anxiety components, respectively. Other test characteristics included sensitivity 56.6%, specificity 68.1%, positive predictive value 67.3%, and negative predictive value 57.5%. Interrater reliability (kappa) was 0.85. The psychometric qualities of the tool are inadequate. Development of further, better quality instruments is needed though this will likely require a longer tool which the healthcare workers delivering epilepsy care services have previously deemed nonfeasible for routine use. As we work toward development and acceptability of a more optimal instrument, use of this initial screening tool which healthcare workers are willing to use may increase the identification of comorbid depression and anxiety in the low resource setting described in the study.

Figure 1

ROC Curve for the Screening Tool