The development and validation of an algorithm to predict future depression onset in unselected youth

• NCJ Number: 304044
• Journal: PSYCHOLOGICAL MEDICINE Volume: 50 Issue: 15 Dated: 2020 Pages: 2548-2556
• Author(s): J. R. Cohen; et al
• Date Published: 2020
• Length: 9 pages
• Grant Number(s): 2018-R2-CX-0022
• Agencies: NIJ-Sponsored

Annotation

The goal of this study was to identify the best screening approach for predicting depression onset in youth.

 

Abstract

Universal depression screening in youth typically focuses on strategies for identifying current distress and impairment; however, these protocols also play a critical role in primary prevention initiatives that depend on correctly estimating future depression risk. In the current study, two multi-wave longitudinal studies (N = 591, AgeM = 11.74; N = 348, AgeM = 12.56) were used as the ‘test’ and ‘validation’ datasets among youth who did not present with a history of clinical depression. Youth and caregivers completed inventories for depressive symptoms, adversity exposure (including maternal depression), social/academic impairment, cognitive vulnerabilities (rumination, dysfunctional attitudes, and negative cognitive style), and emotional predispositions (negative and positive affect) at baseline. Subsequently, multi-informant diagnostic interviews were completed every 6 months for 2 years. The study found that self-reported rumination, social/academic impairment, and negative affect best predicted first depression onsets in youth across both samples. Self- and parent-reported depressive symptoms did not consistently predict depression onset after controlling for other predictors. Youth with high scores on the three inventories were approximately twice as likely to experience a future first depressive episode compared to the sample average. Results suggest that one's likelihood of developing depression could be estimated based on sub=threshold and threshold risk scores. The study concludes that most pediatric depression screening protocols assess current manifestations of depressive symptoms, so screening for prospective first onsets of depressive episodes can be better accomplished via an algorithm that incorporates rumination, negative affect, and impairment. (publisher abstract modified)