Within- and Between-Person Associations among Internalizing and Externalizing Problems during Middle Childhood

• In Event: Poster Session 03
  In Poster Session: PS 03 Topic - Developmental Psychopathology

Abstract

Childhood mental health problems are largely defined by two broad categories: internalizing and externalizing problems. Approximately 10 to 20% of children and adolescents worldwide experience some type of mental health problem (Kieling et al., 2011; Bor et al., 2014). Internalizing and externalizing problems tend to be highly comorbid in childhood (Lilienfeld, 2003; Wolf, & Ollendick, 2006) and they tend to persist overtime, and predict adolescent and adult mental health problems such as mood disorders, disruptive disorders, depression, social phobia, panic attacks, and substance use (Hofstra et al., 2002; Mason et al., 2004; Mathyssek et al., 2012; Miettinen et al., 2013; Loth et al., 2014). Therefore, it is important to investigate the developmental course of internalizing and externalizing symptoms to promote children’s well-being. The aim of the current study is to disentangle longitudinal relations among internalizing and externalizing problems during mid-childhood.

Majority of the previous studies used traditional Cross Lagged Panel Model (CLPM) to examine bidirectional relations between internalizing and externalizing problems. However, this approach can be problematic because traditional CLPM conflates within-person and between-person level results and might result in biased cross-lagged “causal” effects. Recently, random intercept - CLPMs (RI-CLPM) have been developed as an approach that can disentangle within-person and between person variance (Hamaker et al., 2015). Because RI-CLPM accounts for the time-invariant stability of variables, it gives more accurate regression coefficients than CLPM.

We had three main research questions and/or hypotheses: (1) We expected positive correlation between internalizing and externalizing problems both at between-person and within-person level. (2) We expected externalizing problems to predict internalizing problems in the line with the previous studies (Capaldi, 1991; Van der Giessen, 2013; Young et al., 2014; Quach et al., 2017; Murray et al., 2019). (3) We the role of gender on the associations between internalizing and externalizing problems exploratory due to the mixed results in the literature.

We investigated effect of mother-reported internalizing and externalizing problems on each other in mid-childhood through 3rd to 6th grade (N = 1,061). As Figure 1 illustrates, both random intercepts and within-time fluctuations of internalizing and externalizing problems were positively correlated in the line with our hypothesis. Also, as expected, externalizing problems predicted internalizing problems, which means when children show more externalizing problems in a year than they usually show, they show more internalizing problems next year than they usually show. Although we did not hypothesize for the effect of internalizing problems on externalizing problems, it was also significant. That means when children show more internalizing problems in a year than they usually show, they show more externalizing problems next year than they usually show. Hence, although most of the previous research showed that externalizing problems lead to internalizing problems, we should know that internalizing problems can also trigger externalizing problems during mid-childhood. Multigroup comparison of the model did not show any gender difference, that indicates relations between internalizing and externalizing problems are similar among boys and girls. Future studies may examine the mechanism behind the bidirectional effects between internalizing and externalizing problems.

Authors

• Gizem Keskin, University of British Columbia, Okanagan
  Presenting Author

• Jessica P. Lougheed, University of British Columbia (UBC)
  Non-Presenting Author

• Robert Duncan, Purdue University, West Lafayette
  Non-Presenting Author