A meta-analysis of the association between the Children's Depression Inventory (CDI) and suicidality. (May 2022)

• Record Type: Journal Article
• Authors: Detullio, David; Millen, Danielle H.; Kennedy, Tom D.
• Is Part Of: Aggression and violent behavior. Volume 64(2022)
• Publication Date: 2022-05
• Subjects: Suicide -- Depression -- Children's Depression Inventory -- Children -- Adolescents; Aggressiveness -- Periodicals; Violence -- Periodicals; Violent offenders -- Periodicals; Electronic journals; 302.54
• DOI: 10.1016/j.avb.2020.101539
• Languages: English
• ISSNs: 1359-1789

Abstract

Abstract: A number of factors are associated with adolescent suicidal behaviors. One such factor is depressive symptoms, and there are multiple measures designed to assess youth depressive symptoms. One of the most commonly used instruments to assess depressive symptoms in youth is the Children's Depression Inventory (CDI). The aim of this meta-analysis was to evaluate the association between the CDI total score and adolescent suicidality (e.g., suicidal ideation, suicide attempts). Research reports were located through searching databases (e.g., PsycINFO) and reviewing reference lists. A total of 56 research reports ( N = 24, 369) were identified as meeting inclusion criteria. Due to non-overlapping subgroups, 66 correlations were included in a random-effects model. The overall correlation between the CDI total score and suicidality was found to be 0.36, 95% confidence interval (CI) [0.32, 0.41], 80% credibility interval (CrI) [0.13, 0.60], I 2 = 94.15. Two coded study characteristics were found to significantly moderate effect sizes: when passive suicidal ideation was incorporated into the measurement of suicidality and sample type (i.e., clinical, school, community, forensic, mixed). Based on the results of this meta-analysis, CDI total scores could be used as a tool for evaluating adolescents' suicide risk. However, a determination of adolescents' suicide risk should never be based on CDI results alone, and the limitations of this meta-analysis should be considered when evaluating the strength of the findings.