− この都医学研セミナーは終了しました。 −
School of Social and Community Medicine, University of Bristol （Research Fellow）
|会場||東京都医学総合研究所 2階 講堂|
|世話人||西田 淳志 （精神行動医学研究分野 心の健康プロジェクトリーダー）|
Outcomes in psychosis are frequently poor, with approximately 35% relapsing within 3 years and residual symptoms, even after recovery, being common. Co-morbidity with other mental and physical disorders is common at all points of the illness spectrum. Psychosis starts with a prodromal phase, which is often characterised by non-specific symptoms. I would like to describe two areas of my work; Study One seeks to define specific childhood risk factors for adolescent mental health outcomes. Study Two focuses on early detection in primary care services.
Study One : Uses data from a UK birth cohort (ALSPAC) and investigates specific associations between childhood language and social cognition and adolescent psychotic experiences and depressive symptoms. Bi-variate probit analysis models are used in order to unpick specific relations with co-varying outcomes.
Study Two : Uses UK primary care data and a case control design to investigate the development of the psychosis prodrome by investigating consultation patterns in the 5 years before diagnosis of psychosis. Bayesian methods were used to investigate positive predictive values for prodromal symptoms.
Study One : Specific relationships were found between pragmatic aspects of language, rather than intelligibility and fluency, and psychotic experiences, after the association between psychotic experiences and depression was taken into account. Specific associations were also found between locus of control and psychotic experiences.
Study Two : Strong associations were found between all a priori specified prodromal symptoms and a later diagnosis of psychosis, with the exception of poor personal hygiene. The group most at risk of a later diagnosis were young men who repeatedly consulted their primary care clinician for self-harm and suicidal behaviour.
The findings from both studies suggest that there are particular aspects of the way that a child interacts with the social world which may predict later development of psychotic experiences which may be useful for the development of interventions which may reduce the incidence of these phenomena. Also, that it may be possible to develop a prediction model, based on primary care health services use, which could help primary care clinicians decide which of their patients to refer for specialist mental health care.