Validating the Center for Epidemiological Studies Depression Scale for Children in Rwanda

Authors: Theresa Betancourt, Pamela Scorza, Sarah Meyers-Ohki,, Christina Mushashi, Yvonne Kayiteshonga, Agnes Binagwaho, Sara Stulac, William R. Beardslee

Date: December 2012

Abstract: 

Objective

We assessed the validity of the Center for Epidemiological Studies Depression Scale for Children (CES-DC) as a screen for depression in Rwandan children and adolescents. Although the CES-DC is widely used for depression screening in high-income countries, its validity in low-income and culturally diverse settings, including sub-Saharan Africa, is unknown.

Method

The CES-DC was selected based on alignment with local expressions of depression-like problems in Rwandan children and adolescents. To examine criterion validity, we compared CES-DC scores to depression diagnoses on a structured diagnostic interview, the Mini International Neuropsychiatric Interview for Children (MINI KID), in a sample of 367 Rwandan children and adolescents aged 10 through 17 years. Caregiver and child or adolescent self-reports endorsing the presence of local depression-like problems agahinda kenshi (persistent sorrow) and kwiheba (severe hopelessness) were also examined for agreement with MINI KID diagnosis.

Results

The CES-DC exhibited good internal reliability (α = .86) and test-retest reliability (r = .85). The area under the receiver operating characteristic curve for the CES-DC was 0.825 when compared to MINI KID diagnoses, indicating a strong ability to distinguish between depressed and nondepressed children and adolescents in Rwanda. A cut point of≥30 corresponded with a sensitivity of 81.9% and a specificity of 71.9% in this referred sample. MINI KID diagnosis was well aligned with local expressions of depression-like problems.

Conclusion

The CES-DC demonstrates good psychometric properties for clinical screening and evaluation in Rwanda, and should be considered for use in this and other low-resource settings. Population samples are needed to determine a generalizable cut point in nonreferred samples.

[button link=”http://www.jaacap.com/article/S0890-8567(12)00689-2/”]Read More[/button]