Psychosis and aggression in childhood & adolescence
: investigations in clinically referred, inpatient and general population samples

Student thesis: Doctoral ThesisDoctor of Philosophy


Background: Despite evidence of an increased risk of violence among adults with psychosis, little is known about the relationship between psychosis and aggression in young people. Aim: To compare clinical characteristics and associated features in young people with co-occurring psychosis and aggression to those with psychosis or aggression alone. Hypothesis: Those with both psychosis and aggression will share risk factors and correlates with both ’pure’ groups. Method: Three samples were examined. The first (n=6,770), involved secondary data analysis and used information routinely collected on young people referred to the Maudsley Hospital over a 40 year period; the second study involved new data collection by the author, and focused on young people admitted to inpatient units (n=106); the third (n=2,232) involved secondary data analysis and used data from the Environmental Risk Longitudinal Twin Study, a non-referred community sample. Results: Comparisons of co-occurring cases with those with psychosis or aggression only suggested that co-occurring cases showed symptom profiles and risk factors typical of both ’pure’ conditions; in addition, they had higher rates of callous/unemotional traits and parental antisocial behaviour than either ’pure’ group. Independent predictors of psychosis and co-occurring aggression were low IQ, lower scores on theory of mind tasks, internalising problems, exposure to maltreatment, poor educational attainment and oppositional behaviour. Discussion: Consistent with the proposed hypothesis, young people with psychosis and co-occurring aggression shared risk factors and correlates with both ’pure’ groups and showed some additional distinctive features. Conclusions: It is possible to identify psychosis and co-occurring aggressive behaviour in child and adolescent samples; many of the risk factors for the co-occurring pattern appeared similar to those identified in adult studies. Implication
Date of Award1 May 2013
Original languageEnglish
Awarding Institution
  • King's College London
SupervisorBarbara Maughan (Supervisor) & Tamsin Ford (Supervisor)

Cite this