Nosology

As Coughlan & Cannon point out, studies of the association between childhood trauma and psychosis have varied significantly in their methodology, in terms of defining both trauma and psychosis. This may have a significant impact on the reliability of the associations reported and the implications for clinical practice.

Many of the studies cited are population-based and rely on the identification of psychotic symptoms rather than diagnosed psychotic illness. Reference MatthewsMatthews (2017, this issue) discusses the possible limitations of this approach in his commentary, which concentrates on the associations of childhood trauma with psychotic illness. Although there are similarities in the quality of psychotic symptoms in individuals with and without a diagnosed psychotic illness, there are significant differences in how these symptoms are experienced. It is therefore possible that, owing to the broadly defined nature of ‘psychosis’, many of the studies identified by Coughlan & Cannon are not directly relevant to the population of patients seen by specialist psychiatric services.

Aetiopathological aspects

Coughlan & Cannon reflect on how difficult it is to clarify ‘exact pathways that lead from certain childhood traumas to specific psychopathological outcomes’. They discuss how childhood trauma is associated with a range of psychopathological outcomes (which include, but are not limited to, psychotic illnesses). It may be that the search for specific pathways to a specified outcome will be fruitless: outcomes might reflect the complex interplay of numerous susceptibility factors (including genetics, psychological characteristics and life events) which can manifest in a variety of ways.

This complexity is apparent in how associations may arise. Genetic and biological factors play a major role in the development of psychosis. There is strong evidence of an association between schizophrenia and heightened sensitivity to stress, mediated through hypothalamic–pituitary– adrenal (HPA) axis function. Furthermore, traumatic experiences may activate the HPA axis and contribute to the risk of later psychosis (Reference Read, van Os and MorrisonRead 2005; Reference Schreier, Wolke and ThomasSchreier 2009). Reference Barker, Gumley and SchwannauerBarker et al (2015) have described a biopsychosocial model of how childhood maltreatment may lead to psychotic illness in later life that includes the role of the HPA axis. There is also evidence suggesting that those who develop schizophrenia spectrum disorders may show a range of subtle developmental abnormalities many years before the onset of psychotic symptoms (Reference Fish, Marcus and HansFish 1992; Reference WalkerWalker 1994). It is possible that those who later develop schizophrenia may be more likely to experience certain types of childhood trauma and adversity, such as bullying, by virtue of these developmental abnormalities. Although this clearly does not explain the association between childhood trauma and psychosis in its entirety, it does highlight the complex nature of the association and the fact that it is not automatically unidirectional (Reference Schreier, Wolke and ThomasSchreier 2009). Other possibilities are that young people at higher genetic risk of schizophrenia may be more likely to experience childhood trauma (Reference Wigman, van Winkel and OrmelWigman 2012) or that traumatic experiences may predispose to other factors (such as substance misuse), which may in turn increase the risk of psychosis (Reference Whitfield, Dube and FelittiWhitfield 2005). The pathways between childhood trauma and psychosis are complex and multifaceted, and can interact in a variety of ways which incorporate genetic, biological, psychological and social aspects (Reference FisherFisher 2013; Reference Barker, Gumley and SchwannauerBarker 2015; Reference Morgan and Gayer-AndersonMorgan 2016).

Therapeutic implications for child and adolescent mental health services

In child and adolescent mental health services (CAMHS), many patients will have experienced significant traumatic events. It is extremely tempting to try to associate the nature of these experiences with the nature of psychotic symptoms that may subsequently arise. Psychological theories of psychosis have attempted to address this issue; however, as Coughlan & Cannon describe, the evidence for the links between the nature of trauma and the content of later psychotic symptoms is limited, as are the implications for clinical practice.

In his commentary, Reference MatthewsMatthews (2017, this issue) discusses the limited evidence for the effectiveness of trauma-focused therapies in psychotic illnesses. In CAMHS, it is possible that effective therapeutic intervention for young people with significant trauma may have an effect on the risk of developing a psychotic illness in the future; however, the baseline risk of transition to psychotic illness is relatively low in such patients. It is therefore difficult to argue that the association of childhood trauma with psychosis should alter the delivery of CAMHS.

Finally, the evidence presented does highlight how events occurring in childhood can have significant and enduring effects on health and functioning in later life. This emphasises the neurodevelopmental nature of psychotic illnesses and the need to maintain a developmental perspective in clinical practice.