Person-centred Approaches to Psychopathology in the ABCD Study: Phenotypes and Neurocognitive Correlates

Research output: Contribution to journalArticlepeer-review


Issues with classifying psychopathology using narrow diagnostic categories have prompted calls for the use of dimensional approaches. Yet questions remain about how closely dimensional approaches reflect the way symptoms cluster in individuals, whether known risk factors (e.g. preterm birth) produce distinct symptom phenotypes, and whether profiles reflecting symptom clusters are associated with neurocognitive factors. To identify distinct profiles of psychopathology, latent class analysis was applied to the syndrome scales of the parent-reported Child Behaviour Checklist for 11,381 9- and 10- year-olds from the Adolescent Brain Cognitive Development study. Four classes were identified, reflecting different profiles, to which children were assigned probabilistically; Class 1 (88.6%) reflected optimal functioning; Class 2 (7.1%), predominantly internalising; Class 3 (2.4%), predominantly externalising; and Class 4 (1.9%), universal difficulties. To investigate the presence of a possible preterm behavioural phenotype, the proportion of participants allocated to each class was cross-tabulated with gestational age category. No profile was specific to preterm birth. Finally, to assess the neurocognitive factors associated with class membership, elastic net regressions were conducted revealing a relatively distinct set of neurocognitive factors associated with each class. Findings support the use of large datasets to identify psychopathological profiles, explore phenotypes, and identify associated neurocognitive factors.

Original languageEnglish
Pages (from-to)1195-1212
Number of pages18
JournalResearch on Child and Adolescent Psychopathology
Issue number8
Early online date29 Apr 2023
Publication statusPublished - 1 Aug 2023


Dive into the research topics of 'Person-centred Approaches to Psychopathology in the ABCD Study: Phenotypes and Neurocognitive Correlates'. Together they form a unique fingerprint.

Cite this