Testing the neurodevelopmental, trauma and developmental risk factor models of psychosis using a naturalistic experiment

Background The neurodevelopmental and trauma theories are two widely cited models of psychosis. A third-the developmental risk factor model (DRFM)-recognises the combined role of neurodevelopmental risks and trauma. Our objective was to test these theories using preterm populations as a natural experiment, given the high prevalence of neurodevelopmental deficits and exposure to trauma.MethodsTwo population-based preterm birth cohorts, the Bavarian Longitudinal Study (BLS; N = 399) and EPICure Study (N = 184), were included with term-born controls. Peer victimisation in childhood was assessed by parent and child report and psychotic experiences (PE) were assessed in early adulthood. Different models of psychosis were tested using regression and mediation analyses.ResultsThere was support for the trauma and DRFM in the BLS. Peer victimisation increased the risk of PE for preterm and term-born participants equally [odds ratio = 4.87, 95% confidence interval (CI) 1.96-12.08]. There was an indirect effect where preterm children were more likely to be victimised, which subsequently increased risk of PE [β = 1.12 (s.e. = 0.61), 95% CI 0.11-2.48]. The results were replicated in EPICure.ConclusionsExposure to trauma which is experienced more often by neurodevelopmental risk children rather than neurodevelopmental risk per se increases the risk of PE. The findings are consistent with the trauma model and DRFM. Interventions focused on reducing trauma may reduce the development of PE.