Common emotional and behavioral symptoms co-occur and are associated with core temperament factors. This study investigated links between temperament and dimensional, latent psychopathology factors, including a general common psychopathology factor (p factor) and specific latent internalizing and externalizing liabilities, as captured by a bifactor model, in two independent samples of youth. Specifically, we tested the hypothesis that temperament factors of negative affectivity (NA), positive affectivity (PA), and effortful control (EC) could serve as both transdiagnostic and specific risks in relation to recent bifactor models of child psychopathology. Sample 1 included 571 youth (average age 13.6, SD =2.37, range 9.3-17.5) with both youth and parent report. Sample 2 included 554 preadolescent children (average age 7.7, SD =1.35, range =5-11 years) with parent report. Structural equation modeling showed that the latent bifactor models fit in both samples. Replicated in both samples, the p factor was associated with lower EC and higher NA (transdiagnostic risks). Several specific risks replicated in both samples after controlling for co-occurring symptoms via the p factor: internalizing was associated with higher NA and lower PA, lower EC related to externalizing problems.

Background

Recurrent depressive episodes during adolescence result in significant impairment and increased risk for subsequent adverse outcomes throughout the life span. Evidence suggests that early pubertal timing predicts the onset of depressive episodes (particularly for girls); however, it is not known if pubertal timing prospectively predicts recurrent depressive episodes in youth.

Methods

At baseline, 603 youth (56% female, at baseline: M_age  = 12.09, SD = 2.35) reported on their pubertal development. Youth and their parents completed a semistructured diagnostic interview to assess depressive episodes at baseline and then evaluated for onset repeatedly every 6 months for a period of 36 months.

Results

Controlling for past history of depression, Cox proportional hazards models examined whether earlier pubertal timing predicted (a) days to first depressive episode from baseline and (b) days to a second (recurrent) depressive episode from the end of the first episode. Early pubertal timing predicted the onset of the first depressive episode after baseline (b = .19, Wald = 5.36, p = .02, HR = 1.21), as well as a recurrent episode during course of study follow-up episode (b = .32, Wald = 6.16, p = .01, HR = 1.38).

Conclusions

Findings reinforce the importance of considering the impact of early pubertal timing on depression risk. Investigation on how pubertal timing interacts with other risk factors to predict depression recurrence is needed.