Prospective Prediction of Depression and Anxiety by Integrating Negative Emotionality and Cognitive Vulnerabilities in Children and Adolescents
Negative emotionality (NE) and multiple cognitive vulnerabilities (CVs) (negative inferential style, brooding, self-criticism, dependency, dysfunctional attitudes) independently predict internalizing outcomes. The present study examined whether NE and CVs could be structurally integrated into a common factor reflecting shared variance across risks, and specific factors reflecting unique variance in risks. We evaluated the validity and utility of this integrated model via prospective prediction of future depression and anxiety compared to alternative models (NE and CVs individually, a correlated factor model). Youth from a large community sample (N = 571; M = 13.58 years old; 55% girls, 44% boys) reported on NE and CVs. Depression and anxiety symptoms based on youth report, and disorder onset based on youth and caregiver diagnostic interviews were assessed over a 1½ years follow-up. Results supported a structural model including a general NE-CV dimension and specific dimensions for NE, common cognitive risk, negative inferential style, and brooding; model invariance was obtained from late childhood through late adolescence and for girls and boys. The integrated general NE-CV factor more consistently and strongly predicted future depressive (β = 0.58) and anxiety (β = 0.56) symptoms, and onsets of depression (OR = 1.81) and anxiety (OR = 2.23) relative to NE and CV risk dimensions across alternative models (ps < .01). The general NE-CV dimension represents an important means to efficiently represent transdiagnostic risk for internalizing outcomes among youth.