The term developmental epidemiology was originally confined to the study of the distribution and risks of childhood disorders (Costello and Angold, 1995) but has extended to include the study of early antecedents and risk factors for adult-onset illness and chronic diseases (Buka and Lipsitt, 1994). It is also known as ‘ life-course epidemiology’ (Kuh and Ben-Schlomo, 1997). Developmental epidemiology studies causation in the context of development and investigates causal chain mechanisms and person-environment interaction.

From its first descriptions, schizophrenic psychosis had a longitudinal dimension. Thomas Clouston (1892) recognized a syndrome of ‘developmental insanity’ in which developmental physical abnormalities were associated with early-onset psychotic phenomena, particularly in men. While defining the schizophrenia syndrome more clearly, both Kraepelin (1896) and Bleuler (1911) noted that people who developed the psychotic syndrome were often different from their peers before psychosis began, but these observations were incorporated into the psychodynamic formulations prevalent at the time.

During the 1980s, a ‘neurodevelopmental hypothesis’ of schizophrenia became prominent (Murray & Lewis, 1987; Weinberger, 1987). This hypothesis broadly proposed that interaction between early pathology or insult and normal processes of structural and functional brain development yielded a nervous system prone to psychosis. Jones (1999) has pointed out that the neurodevelopmental hypothesis of schizophrenia ‘ is not really a hypothesis at all, rather a general position or thesis … that has directed research towards early life in terms of causation’.