Four behavioural syndromes of schizophrenia: a replication in a second inner-London epidemiological sample
Introduction
We have reported extracting four behavioural factors by principal components analysis of Social Behaviour Schedule (SBS) data (Wykes and Sturt, 1986) from a large inner-London epidemiological sample of patients with strictly defined schizophrenia. These factors accounted for 58% of the variance (Harvey et al., 1996). This was the first study to investigate dimensions of social behaviour rather than signs and symptoms of schizophrenia in a large representative sample. Psychiatric services for the chronic severely mentally ill in the United Kingdom are provided almost exclusively by the state-funded National Health Service (NHS), which is free at the point of delivery and service provision is allocated according to the geographical area where the patient resides (catchment area). This allows for such epidemiological studies, which may not be conducted so easily in other countries.
Two Manchester Scale (MS) (Krawiecka et al., 1977) symptom factors accounting for 65% of the variance were also extracted using principal components analysis (Harvey et al., 1996). They were sufficiently similar to two of the `three syndromes' of Liddle and Barnes (1990)that they were named Psychomotor poverty and Reality distortion. The specific pattern of correlations between each social behaviour (SBS) factor and Manchester Scale (MS) symptom factors and items, certain demographic variables and aspects of social functioning suggested that these behavioural factors had sufficient validity to be considered as behavioural syndromes. Reflecting the respective SBS item loadings, these four syndromes were named: Thought disturbance, Social withdrawal, Depressed behaviour and Anti-social behaviour. Similarities between these four syndromes and those identified by others, for example, our Social withdrawal and Wing and Brown's (Wing and Brown, 1970) `social withdrawal' syndrome, our Anti-social behaviour syndrome and the Excitement syndrome extracted by Kay and Sevy (1990), have been highlighted (Harvey et al., 1996). We concluded that the behavioural syndromes offered a different perspective on the components of disability in schizophrenia since, as Smith et al. (1998)have recently observed, even three symptom factors do not capture the structure amongst schizophrenic symptoms. Furthermore, `either more latent variables underlie the inter-correlations among schizophrenic symptoms, or the set of symptoms most often studied is incomplete' (Smith et al., 1998).
A second inner-London epidemiological survey of schizophrenia in South Westminster, which used similar methodology to the first in Camden (Campbell et al., 1990; Harvey, 1996), offered the opportunity to attempt a replication of the four behavioural syndromes in a different but comparable population using identical case-recognition techniques and strict diagnostic criteria. The assessments of psychiatric symptoms and problem behaviours were performed by an appropriately trained psychiatrist (P.D.) who had not been involved in the original Camden study. Furthermore, analysis of the Camden social behaviour data had not been undertaken at the time of data collection in South Westminster. This second study is, therefore, a true attempt at independent replication of the first study.
It was hypothesized that behavioural syndromes similar to those found in Camden would be derived from assessments of patients' problem behaviours and that different behavioural syndromes in the community sub-group would be associated, as in Camden, with different levels of social functioning, such as employment status, type of accommodation, social contacts and leisure activities. If this replication requiring exploratory factor analysis, as in the first study, was successful, it would then allow for the generation of a model of schizophrenic phenomena. Such a model could then be tested by using confirmatory factor analysis for goodness of fit and by the identification of neurophysiological and neuropsychological correlates.
Section snippets
Methods
The South Westminster Schizophrenia Survey took place in a geographically small but densely populated inner-city area of London. Some parts of South Westminster had high levels of social impoverishment, contrasting with very high levels of affluence elsewhere in the area. For example, using the Jarman index as a measure of social deprivation (Jarman et al., 1992), the district was the 27th most socially deprived of 192 districts in England and Wales. Acute admission, day hospital, industrial
Socio-demographic and clinical findings
In South Westminster, 352 patients with a diagnosis of schizophrenia or related psychosis were identified and, after excluding those over 60 years of age, 112 fulfilled the restrictive Feighner diagnostic criteria for probable or definite schizophrenia (Feighner et al., 1972) (see Section 2). No long-stay patients, defined as continuously resident for 2 years or more on a long-stay ward of Horton Hospital, were aged 60 years or under. At the time of assessment, 101 patients were living in
Discussion
Using identical methodology and the same reliable social behaviour rating scale in a second inner-London epidemiological survey of patients with the same strictly defined Feighner diagnostic criteria for schizophrenia, we extracted, by exploratory principal components analysis, four behavioural factors. This is an independent replication of the first study conducted in Camden, London (Harvey et al., 1996). As in the Camden study, the specific pattern of correlations between each social
Conclusion
The SBS-derived behavioural syndromes extracted in the Camden studies, and now by an independent replication in the South Westminster studies, were more robust than the MS symptom-derived syndromes and offer a different perspective on disability. Specifically, they add to the range of phenomena previously studied by factor analytic techniques, and suggest that the exclusion of important phenomena may limit our ability to encapsulate the heterogeneity of schizophrenia. The inclusion of
Acknowledgements
The South Westminster Schizophrenia Survey was supported by a grant from the Locally Organised Research Scheme Fund of The North West Thames Regional Health Authority. Special thanks are due to Dr Patrick McCabe for statistical advice.
References (38)
- et al.
What is the best maintenance dose of neuroleptics in schizophrenia?
Psychiatry Res.
(1980) - et al.
Neuropsychological and olfactory dysfunction in schizophrenia: relationship of frontal syndromes to syndromes of schizophrenia
Schizophr. Res.
(1995) - et al.
Cognitive deficits in obsessive compulsive disorder on tests of frontal–striatal function
Biol. Psychiatry
(1998) - et al.
The structure of schizophrenic symptoms: a meta-analytic confirmatory factor analysis
Schizophr. Res.
(1998) - et al.
Positive and negative symptoms in neuroleptic-free psychotic inpatients
Schizophr. Res.
(1995) - et al.
Parallel organization of functionally segregated circuits linking basal ganglia and cortex
Annu. Rev. Neurosci.
(1986) - American Psychiatric Association, 1987. DSM-III-R: Diagnostic and Statistical Manual of Mental Disorders, 3rd ed., rev....
- et al.
The nature and prevalence of depression in chronic schizophrenic in-patients
Br. J. Psychiatry
(1989) - Blumer, D., Benson, D.F., 1975. Personality changes with frontal and temporal lobe lesions. In: Benson, D.F., Blumer,...
- et al.
Principle and practice of measuring needs in the long-term mentally ill. The MRC needs for care assessment
Psychol. Med.
(1987)