An Examination of Symptom Clusters that Predict Suicidal Ideation in School-Aged Children

• In Event: Poster Session 06
  In Poster Session: PS 06 Topic - Developmental Psychopathology

Abstract

Suicidality in preadolescent children is a crucial public health concern, as suicide is the second leading cause of death among children ages 10-14 in the United States (CDC, 2015). However, suicidal thoughts and behaviors are largely understudied in preadolescent children (Westefeld et al., 2010), and epidemiological research in adult samples suggests that prediction of suicidality may be improved by examining clusters of symptoms rather than restricting prediction based on categorical psychiatric diagnoses (Verona et al., 2004). The present study addresses many gaps in extant research using multi-informant measures to examine how various internalizing symptom clusters predict suicidal ideation in school-aged children. The sample comprised 167 children (56 girls; 68% White/non-Hispanic) ages 8-13 (M=10.34, SD=1.51) recruited for participation in a larger study that over-sampled for children with behavioral and emotional difficulties. Study participation included an in-depth psychoeducational evaluation, during which data for the current study were collected. Children with ADHD, anxiety, mood, learning, and autism spectrum disorders as well as neurotypical children were included. Children self-reported depressive and anxious symptoms and suicidal ideation via standardized questionnaires. Parents reported on children’s symptoms and suicidal ideation via both a semi-structured interview and standardized questionnaires. Data were analyzed to discover which symptom clusters significantly predicted children’s suicidal ideation based on self- and parent-report. Results of two multivariate regressions revealed that, when controlling for all other parent- and child-reported symptom clusters and diagnostic status, only child-reported generalized anxiety symptoms predicted child-reported suicidal ideation (β=.41, p=.007), and only child-reported obsessive/compulsive symptoms predicted parent-reported suicidal ideation (β=.21, p=.036). Exploratory analyses revealed that an internally consistent (α=.71) ‘mini-scale’ comprised of three items from the standardized child anxiety questionnaire predicted child-reported suicidal ideation (β=.22, p=.005). This study emphasizes that certain anxiety symptoms may be particularly relevant to the development and prediction of suicidal ideation in school-aged children, and that these symptoms can be assessed in an efficient manner based on self-report. Implications of these findings are potentially far-reaching, as future research may focus on how the presence of anxiety symptoms leads to comorbid or future suicidality in children. Additionally, these findings are relevant in clinical and school-based practice, as anxiety symptoms should be considered potentially useful when screening for and responding to suicidal thoughts and behaviors in children.

Authors

• Nicole Groves, Florida State Univeristy
  Presenting Author

• Roshni Janakiraman, Florida State University
  Non-Presenting Author

• Elia F Soto
  Non-Presenting Author

• Leah Singh, Florida State University
  Non-Presenting Author

• Carolyn L. Marsh, Florida State University
  Non-Presenting Author

• Caroline E Miller, University of British Columbia (UBC)
  Non-Presenting Author

• Erica Wells, Florida International University
  Non-Presenting Author

• Elizabeth S.M. Chan, Florida State Univeristy
  Non-Presenting Author

• Katherine Schafer, Florida State University
  Non-Presenting Author

• Michelle Jeon, Florida State University
  Non-Presenting Author

• Michael J. Kofler, Florida State University
  Non-Presenting Author

• Thomas Joiner, Florida State University
  Non-Presenting Author