Back

Identifying subtypes of youth suicidality based on psychopathology: alterations in genetic, neuroanatomical and environmental features

Wu, X.; Palaniyappan, L.; van Velzen, L.; Yu, G.; Fan, H.; Liu, Y.; Cheng, W.; Zhao, X.-M.; Feng, J.; Sahakian, B.; Robbins, T. W.; Schumann, G.; Schmaal, L.; Zhang, J.

2024-04-15 psychiatry and clinical psychology
10.1101/2024.04.13.24305772 medRxiv
Show abstract

One of the most complex human behaviours that defies singular explanatory models is suicidal behaviour, especially in the youth. A promising approach to make progress with this conundrum is to parse distinct subtypes of this behaviour. Utilizing 1,624 children with suicidal thoughts and behaviors (STBs) and 3,224 healthy controls from the ABCD Study, we clustered children with STB based on thirty-four cognitive and psychopathological measures which capture suicide-related risk-moderating traits. Environmental and genetic risk factors, as well as neuroanatomical characteristics of each subtype, were then compared with controls. We identified five distinct STB subtypes, each revealing unique neuroanatomy, environmental/genetic risks, and persistence patterns. Subtype 1 (Depressive, 9.6%) exhibited the most severe depressive symptoms. Subtype 2 (Externalizing, 20.1%) displayed anatomical and functional alterations in frontoparietal network and increased genetic risk for ADHD. Subtype 3 (Cognitive Deficit, 20.4%) demonstrated lower cognitive performance and widespread white-matter deficits. Subtype 4 (Mild Psychotic, 22.2%) presented higher prodromal psychotic symptoms, often unnoticed by parents. Subtype 5 (High Functioning, 27.6%) showed larger total brain volume, better cognition, and higher socio-economic status, contrasting subtypes 1-4. Only Subtypes 1 and 2 demonstrate persistent STB features at the 2-year follow-up. Our results suggested that youth suicidal behaviour may result from several distinct bio-behavioral pathways that are identifiable through co-occurring psychopathology, and provide insights into the underlying neural mechanisms and corresponding intervention strategies.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.