Modelling the co-occurrence of psychosis-like experiences and childhood sexual abuse

Recurrent research evidence indicates that childhood sexual abuse (CSA) is associated with psychosis and psychosis-like experiences (PLEs). Many individuals however who have experienced psychosis have never been sexually abused in childhood and many individuals who have experienced CSA have never experienced psychosis. The current study sought to model the co-occurrence of CSA and PLEs using data from the Adult Psychiatric Morbidity Survey. Latent class analysis was employed to identify distinct classes of individuals in the general population who were characterised by the presence, co-occurrence or absence of PLEs and/or CSA. Multinomial logistic regression analysis was utilised to validate membership of classes characterised by both CSA and PLEs using a series of variables that have been proposed to delineate the co-occurrence of these phenomena. Four hypothesised classes were identified, (1) a CSA-PLE co-occurrence class, (2) a PLE-only class, (3) a CSA-only class and (4) a CSA and PLE free baseline class. CSA-PLE co-occurrence was characterised by neurotic disorder, social isolation, adult sexual molestation and a history of post-traumatic stress disorder (PTSD). PLE occurrence in the absence of CSA was characterised by neurotic disorder, social isolation, a history of PTSD, childhood physical abuse, and uniquely by discrimination and non-sexual trauma post-16 years. The findings indicated that a distinct group of individuals in the population was characterised by the co-occurrence of CSA and PLEs. In the absence of CSA, individuals who experienced PLEs were likely to endure a wide range of other, non-sexual, traumatic and adverse experiences. The CSA-PLE co-occurrence class and its associated psychosocial risk profile was discussed in relation to established trauma-based perspectives of psychosis and PLEs.