Poster #16 - Agreement between Adolescent and Parent Reports of Adolescents’ Suicidal Thoughts in the Philadelphia Neurodevelopmental Cohort

• In Event: 3-001 - Poster Session 2 during Breakfast

Abstract

An alarming number of adolescents report having serious thoughts of suicide. Suicidal ideation is a risk factor for a host of negative developmental outcomes, including a future suicide attempt. As such, early identification and treatment of suicidal thoughts are critical for suicide prevention, yet most adolescents experiencing suicidal thoughts do not receive treatment. Parents are “gatekeepers” for youth’s access to mental health services but, unfortunately, parents and adolescents often do not agree on whether the adolescent is experiencing suicidal thoughts. Most concerning is that disagreements stem mainly from parental unawareness of adolescents’ suicidal thoughts, though there is some evidence that parents may report suicidal thoughts that the adolescent denies. Much of this research to date has used clinical or high-risk samples and very little is known about factors that are associated with reporter disagreements about adolescents’ suicidal thoughts.

The present study is the largest to date to examine: (a) adolescent and collateral informant agreement on the presence of suicidal thoughts and (b) factors that are associated with agreement. We utilized publicly available data from the Philadelphia Neurodevelopmental Cohort (PNC), a sample of 9,498 community youth. Analyses focused on adolescents 11-17 years old (n=5,137; mean age=14.53, SD=1.98; 52% female) for whom collateral report of suicidal thoughts was obtained. A structured clinical interview assessed broad domains of psychopathology, treatment history, and lifetime suicidal thoughts. Collateral informants (97% parents) provided demographic information.

Agreement was operationalized in three ways: a) overall agreement on suicidal thoughts, b) collateral unawareness of suicidal thoughts reported by the adolescent, and c) adolescent denial of suicidal thoughts reported by the collateral. We used kappa as an index of overall agreement. We used logistic regression to examine demographic and clinical factors associated with agreement.

Fifteen percent of adolescents reported thoughts of death or dying and 8% reported thoughts of killing themselves. Reporter agreement on thoughts of death or dying was low (κ=.171) and agreement on thoughts of killing themselves was moderate (κ=.466). Sensitivity and specificity estimates indicated that reporter disagreements largely stemmed from collateral unawareness of adolescents’ suicidal thoughts (>75% unaware of thoughts of death/dying; >50% unaware of thoughts of killing themselves (see Table 1). Logistic regression analyses revealed several demographic (e.g., adolescent age) and clinical (e.g., treatment history) factors associated with reporter agreement (see Table 2). Implications of these findings and the potential of publicly available PNC data to address important questions in developmental science will be discussed.

Authors

• Jason Jones, Children's Hospital of Philadelphia
  Presenting Author

• Rhonda C. Boyd, Children's Hospital of Philadelphia

• Monica E. Calkins, University of Pennsylvania

• Annisa Ahmed, Children's Hospital of Philadelphia

• Ran Barzilay, Children's Hospital of Philadelphia

• Tami D. Benton, Children's Hospital of Philadelphia

• Raquel E. Gur, University of Pennsylvania