(PS3-38) Modeling the Longitudinal Relationship Between Self-Esteem and Depression: A CT-SEM Approach

Abstract Text: Depression is a major public health concern, and adolescence is a key period for understanding mental health trajectories. While substantial existing research has explored risk factors for depression, methodological limitations often obscure the nuanced relationships between these variables. The present study seeks to address these limitations by employing continuous-time structural equation modeling (CT-SEM) to examine the longitudinal relationship between self-esteem and depressive symptoms across adolescence and into adulthood, offering a more precise analysis of depression risk factors.

Using three waves of data from the National Longitudinal Study of Adolescent to Adult Health (Harris et al., 2009; N = 6233), this study provided a comprehensive examination of the self-esteem and depression relationship. Participants were aged 12-19 in Wave I (1994-1995), 13-20 in Wave II (1995-1996), and 19-26 in Wave III (2001-2002). CT-SEM was used to address key methodological challenges in longitudinal research, including unequal time intervals between waves—a limitation that traditional cross-lagged panel models fail to adequately address.

Results of the continuous time auto- and cross-effects model with unequal intervals reveal a significant asymmetric relationship between self-esteem and depression. Results demonstrated a robust unidirectional effect of self-esteem on depressive symptoms (β = -0.32, 95% CI [-0.41, -0.23]), indicating that higher self-esteem predicted decreased depressive symptoms and, conversely, lower self-esteem predicted increased depressive symptoms. Depression exhibited no statistically significant impact on self-esteem (β = -0.08, 95% CI [-0.45, 0.32]). The depression auto-effect parameter (β = -0.15, 95% CI [-0.52, -0.02]) and self-esteem auto-effect parameter (β = -0.75, 95% CI [-1.34, -0.36]) indicated a tendency for depressive symptoms and self-esteem to return to an equilibrium point over time.

Our findings substantiate the vulnerability model proposed by previous researchers, offering empirical support for the causal role of self-esteem in depression risk using a novel statistical design. These findings support self-esteem enhancement as a promising strategy for reducing adolescent depression risk. Future research directions include exploring additional moderating factors, such as sleep duration, and examining covariates such as gender, race/ethnicity, and socioeconomic status.