From the hidden to the obvious: classification of primary and secondary school student suicides using cluster analysis.

Suicide is one of the leading causes of death in children and youth. Using a sample of fatal suicides among school-aged students in Hong Kong, this study aimed to demonstrate how the classification of children and adolescent suicides into distinct subgroups using cluster analysis can alert us to the heterogeneous nature of the student suicide population and increase our understanding of multidimensional underlying causes.  METHODS: Deaths by suicide of Hong Kong primary and secondary school students occurring between 2013-16 were identified. Reports were acquired from the Coroner's Court, Police Force, and Education Bureau in Hong Kong. Information about students' sociodemographic characteristics, suicide circumstances, stressors, and risk factors was extracted and organized for analysis. Based on the indicated stressors (school, family, close relationship, social challenge, finance, risk behaviour, suicide exposure, others) and risk factors (health and mental health, history of self-harm, suicidality, and psychological maladjustment), cluster analysis was conducted to derive distinct profiles of student suicides. A four-cluster solution was found. Patterns of stressors, risk factors, background characteristics and suicide circumstances within each cluster were examined. Four distinct and meaningful profiles of student suicides were characterised as "school distress", "hidden", "family and relationship", and "numerous issues". Findings highlighted the need to approach student suicides in meaningfully differentiated ways. Gathering suicide report data and generating evidence that advances our knowledge of student suicide profiles are important steps towards early identification and intervention. © 2022. The Author(s).

Citation

Anna Wong, Carmen C S Lai, Angie K Y Shum, Paul S F Yip. From the hidden to the obvious: classification of primary and secondary school student suicides using cluster analysis. BMC public health. 2022 Apr 09;22(1):693

Mesh Tags

PMID: 35395743

search → result