Patterns of Suicide Ideation and Behavior in Preteens

Body of Abstract: Research

Aims: Suicide in preadolescent youth is a growing and pressing problem and is currently the 2nd leading cause of death in children aged 10-14 (CDC, 2022). However, little is known about why or how preadolescents come to think about, attempt, or die by suicide, leading to a call to action for researchers to increase suicide-focused research in preadolescents (Ayer et al., 2020). The current study aims to examine the heterogeneous patterns of suicidal thinking and behavior using items of the Columbia -Suicide Severity Rating Scale (C-SSRS) in a community sample of children aged 6-11 years with a family history of suicidal thinking or behaviors. Based on prior research (Bernake et al., 2017), we hypothesize to find a class characterized by a low probability of endorsing any items on the C-SSRS, a class characterized by a high probability of endorsing ideation, but not behavior, and additional classes characterized by high levels of ideation and differing types of suicide-related behavior.

Methods: Participants were 294 children from who participated in either an American Foundation for Suicide Prevention or a National Institute of Health-funded study examining family history of suicidality on the development of child suicidality. IRB approval was obtained for both studies. Children’s age ranged from 6-11 years (Mage = 8.24, SD = 1.64). The majority of youth identified their race as White (50.2%) or Black (32.4%) and identified their ethnicity as Non-Hispanic (86.7%). Child’s self-reported suicidality (i.e., lifetime history of passive ideation, active ideation, active ideation with an identified method, active ideation with an identified method and specific planning, non-suicidal self-injury, previous suicide attempt, previous aborted suicide attempt, previous interrupted suicide attempt) was obtained using items of the C-SSRS. Children’s internalizing and externalizing symptoms were measured using the respective scales of the Child Behavior Checklist (CBCL). A latent class analysis was performed using items on the C-SSRS coded dichotomously (0 = no, 1 = yes) as indicators in the model. Follow-up logistic regressions examined the association between demographic variables (i.e., age, sex, race, ethnicity), and internalizing and externalizing symptoms on latent class assignment.

Results: Approximately 50% of youth endorsed experiencing passive suicidal ideation, 33% endorsed active suicidal ideation, 26% endorsed active suicidal ideation with an identified method, 19% endorsed active suicidal ideation with an identified method and specific plan, 68% endorsed lifetime non-suicidal self-injury, 13% endorsed a prior suicide attempt, 5% endorsed an aborted suicide attempt, and 4% endorsed an interrupted suicide attempt. Latent class analyses with 2-6 models were run to determine the best fitting model for the data. The model with the best fit was determined by examining the frequency of individuals in each latent class, fit information criteria, entropy, and the Lo-Mendell Rubin likelihood ratio test. A 2-class solution provided the best model fit, as it demonstrated only marginally higher Sample-Sized Adjusted Bayesian Information Criteria (1221.641), better ability to discriminate between classes (higher entropy value at .961), a significant Lo-Mendell Rubin likelihood ratio (p <.05), and latent classes with at least 10% of individuals in each class. The first class (32% of sample) was characterized by a high probability of endorsing passive and active ideation and ideation with a method, high probability of endorsing non-suicidal self-injury (.68), and a probability of .56 of endorsing any kind of previous suicide attempt. The second class (68% of sample) was characterized by a low probability of endorsing each item on the C-SSRS (i.e., no ideation or suicide-related behavior). A follow-up logistic regression revealed only sex was significantly associated with class assignment in step 1 (([B (1) = -1.06, p <.05], such that youth who identified as male were more likely to be classified in the “high suicide ideation and NSSI” class. The addition of parent-reported internalizing (p = .331) and externalizing symptoms (p = .323) in step 2 were not significantly associated with latent class assignment; however, the addition of youth’s adverse childhood experiences added significant variance in step 2 ([B (1) = -.244, p <.05], such that a higher number of adverse childhood experiences were associated with being in the “high suicide ideation and NSSI” class.

Conclusions: Findings from the current study reveal preteens do report thoughts of suicide at high frequencies, highlighting the importance of assessing for suicidality in this age group. Contrary to hypotheses, latent class analyses revealed a 2-class solution fit the data best, which may be partially explained by the low frequencies of suicidal behavior in the current sample. Moreover, demographics and parent-reported internalizing and externalizing symptoms were largely not significantly associated with class assignment, though adverse childhood experiences was related to class assignment. While these results may be partially explained by the reliance on parent-report as opposed to child reported psychological symptoms, these results mirror previous research that has found different psychological symptoms associated with child suicide decedents compared to adolescent suicide decedents (Sheftall et al., 2016). Future research should examine whether other relevant psychological symptoms (e.g., impulsivity, emotion regulation) are associated with class assignment. Future research should also continue to examine how these heterogeneous patterns may change as preteens age to examine whether additional patterns of ideation and behavior are observed in older age groups. Overall, these findings further demonstrate the importance of assessing for suicidality in preteens and highlight the need to identify relevant symptoms or other factors that may be associated with suicidality in this age group.

Learning Objectives:

• -Be able to describe frequencies of suicidal thoughts and behaviors in preteens.
• -Be able to list sociodemographic and methodological differences in preteen suicide decedents compared to adolescent suicide decedents.
• -Be able to describe specific ways suicide-related ideation and behavior may co-occur in preteens.