Schizophrenia in Camberwell, 1965-1984

This Thesis describes the epidemiology of schizophrenia and related disorders in the defined catchment area of Camberwell, SE London, UK, over the period 1965 to 1984. Cases were ascertained through the comprehensive Camberwell Cumulative Psychiatric Case Register. All first-contact patients with a Register diagnosis of any non-affective non-organic psychotic illness were included in the study. Diagnostic uniformity was ensured by rediagnosis of all cases (n=531) using the computerised OCCPI system, which facilitates rediagnosis according to a wide range of diagnostic criteria. Trends in the incidence of non-affective functional psychoses over the two decades during which the Camberwell Register was operational, are explored. The findings, of a rising rate of illness in Camberwell, are discussed in terms of changes in the demography of the general population over the years, and suggestions offered for discrepancies with other studies of time trends in schizophrenia, particular emphasis being placed on changes in the ethnic composition of Camberwell over this period. A case-control study design is used to explore whether the rising incidence of the illness in the area is due solely or largely to drift into the area of ill individuals, or whether some of the variance can be explained in terms of a pernicious inner-city effect operating during early development (in utero or in early childhood). The findings of an excess of schizophrenia patients actually having been born in the inner city suggests that something about poor households in the inner city might predispose to the illness in later life. This is discussed in the general framework of the neurodevelopmental hypothesis of schizophrenia, which proposes that at least some individuals have a form of illness consequent upon subtle damage to the developing brain. A major focus of the analyses is gender differences in schizophrenia, and late onset schizophrenia. Early-onset males were particularly likely to fulfil stringent diagnostic criteria for the illness, and to show premorbid dysfunction. The results are interpreted in the neurodevelopmental framework, and reference made to differences in male and female brains in their vulnerability to neurodevelopmental illnesses in general. Taking this theme forward, a form of factor analysis called latent class analysis is used to further explore the notion of different subtypes of schizophrenia, one of which is an early-onset severe male-predominant form (theoretically consequent upon neurodevelopmental deviance). The analyses resulted in a "best fit" model of three subtypes, one an early-onset male-predominant type associated with premorbid dysfunction ("neurodevelopmental" type); a later-onset "paranoid" type; and an affect-laden type exclusive to females ("schizoaffective" type). There were associations with a number of variables of potential importance in terms of aetiology, namely an association of the "neurodevelopmental" type with a family history of schizophrenia and obstetric complications; an association of the "paranoid" type with winter birth; and of the "affective" type with a family history of psychiatric disorder other than schizophrenia (predominantly affective disorder). This typology does not adequately account for those patients with a late (over 45 years), or very-late onset of illness (over 60). Phenomenological, premorbid, and other differences between early- and late-onset patients are analysed, and the results discussed in the broader framework of the literature on late-onset non-affective psychoses