Biological, Life Course, and Cross-Cultural Studies All point Toward the Value
of Dimensional and Developmental Ratings in the Classification of Psychosis
Rina Dutta1,2, Talya Greene2, Jean Addington3, Kwame
McKenzie4,5, Michael Phillips6,7, and Robin M. Murray2
2Division of Psychological Medicine and Psychiatry, Box No. 63,
Institute of Psychiatry, King’s College London, 5th Floor, Main
Building, De Crespigny Park, London SE5 8AF, UK;
Mental Health Sciences, University College London;5The Centre
Suicide Research and Prevention Center;7Departments of Psy-
chiatry and Epidemiology, Columbia University, USA
The diagnostic criteria for schizophrenia in the fourth edi-
tion of the Diagnostic and Statistical Manual of Mental
Disorders (DSM-IV1) are based on the premise that it is
a discrete illness entity, in particular, distinct from the af-
fective psychoses. This assumption has persisted for more
than a century, even though patients with a diagnosis of
schizophrenia show a wide diversity of symptoms and out-
comes, and no biological or psychological feature has been
found to be pathognomonic of the disorder. However, there
has been sustained, and indeed growing, criticism of the
concept. For example, writing about the diagnosis of
schizophrenia more than a decade ago,2one of Britain’s
most sophisticated nosological experts, Ian Brockington,
enjoined ‘‘It is important to loosen the grip which the con-
cept of ‘schizophrenia’ has on the minds of psychiatrists.
Schizophrenia is an idea whose very essence is equivocal,
hypothesis. Such a blurred concept is ‘not a valid object of
scientific enquiry’.’’3Should Diagnostic and Statistical
Manual of Mental Disorders, Fifth Edition(DSM-V), per-
sist with the neo-Kraepelinian concept of schizophrenia
with all its defects, or should it deconstruct psychosis
into its component dimensions? In this article, we will ad-
dress the question by considering 2 main themes, firstly, the
role of culture and ethnicity in the diagnosis of psychosis,
and secondly, a life course approach to understanding psy-
chosis. We will then discuss whether more progress would
be achievedin DSM-V byabandoningthefamiliar categor-
ical system and instead moving to a dimensional system
which rates both developmental impairment and symptom
factor scores. However, we will begin by briefly reviewing
the recent history of the classification of the psychoses.
Key words: diagnosis/deconstructing psychosis/DSM-IV
The Recent History of the Classification of Psychoses in
For the categorical diagnosis of schizophrenia to be sci-
entifically valid, it should define a syndrome with specific
risk factors, psychopathology, treatment responses, and
outcomes; clear symptom boundaries should separate it
from other conditions such as the affective psychoses.
That such a distinction could be made between ‘‘demen-
tia praecox’’ and ‘‘manic depressive insanity’’ (schizo-
phrenia and affective psychosis) has been fundamental
to psychiatric classificatory systems since Kraepelin’s
original proposal of the dichotomy in the 19th century.
This is despite the fact that in 1920 Kraepelin came to
doubt his own approach and suggested replacing his de-
fining principle with a dimensional-hierarchical model
more appropriate to the heterogeneity of clinical presen-
tations.5Furthermore, in spite of the theoretical distinc-
tion between schizophrenia and mood disorder with
psychotic features, the practicalities of clinical life led
to development of a less than satisfactory intermediate
Attacks on the Concept of Schizophrenia
Szasz, curiously both psychiatrists, who argued that
psychiatric diagnoses such as schizophrenia were arbi-
Then in the 1990s, more academically sophisticated criti-
cism came from British clinical psychologists such as
Richard Bentall and Mary Boyle who argued that a
symptom-based approach was less stigmatizing and
more appropriate from a therapeutic point of view.6,7
However, criticism did not just stem from outside ortho-
dox psychiatry. Phenomenologists such as Brockington,
biological researchers such as Crow, and epidemiologists
such as Van Os have led a growing chorus of dissent from
within the ranks of psychiatrists.
1To whom correspondence should be addressed; tel: þ44 (0)20-
? The Author 2007. Published by Oxford University Press on behalf of the Maryland Psychiatric Research Center. All rights reserved.
For permissions, please email: firstname.lastname@example.org.
Schizophrenia Bulletin vol. 33 no. 4 pp. 868–876, 2007
Advance Access publication on June 11, 2007
The Hope Promised by Operational Definitions
From the late 1960s onward, a number of competing op-
erational diagnostic systems were proposed in an attempt
to improve the reliability of psychiatric diagnosis for
research purposes. These included Feighner’s, Taylor’s,
CATEGO system. These operational definitions were
generally shown to be internally reliable once psychiatrists
were trained in their use. However, the various competing
diagnostic systems were compared with respect to their
reliability, concordance, and prediction of outcome8,9and
found to show wide disparity. For example, the systems
varied by as much as 7-fold in their rates of diagnosing
These criteria which were primarily designed for re-
search purposes were followed by the incorporation of
similar operational rules for clinicians in the third edition
of the Diagnostic and Statistical Manual of Mental Dis-
orders (DSM-III11) published in 1980. Like the Feighner
criteria, the DSM-III definition of schizophrenia was
narrow,requiring6months ofillness beforethe diagnosis
could be made.
In the Camberwell Register study conducted by Castle
and colleagues,12the authors examined the proportion of
patients with a first episode of nonaffective psychosis
who met different criteria. Nearly two-thirds of the 486
cases met the Research Diagnostic Criteria for either
‘‘broad’’ or ‘‘narrow’’ schizophrenia; this is not surprising
given that this is the most liberal system, with no age-at-
onset stipulation and only a 2-week illness duration re-
quirement. However, only 32.6% of 486 cases fulfilled
the criteria for schizophrenia in DSM-III and 32.3% for
criteria were much influenced by the St Louis school from
which the Feighner criteria had emerged. Both the Feigh-
ner and DSM-III criteria had a high degree of predictive
specificity, with one study showing no change in diagnosis
over time using these criteria and an average of 6.5 years
The Continuing Problem of Validity
With training, especially in the use of standardized inter-
views, DSM-III, like the other main competing systems,
produced acceptable interrater reliability. However, reli-
ability does not necessarily mean validity and attempts
to study validity as opposed to reliability were limited.
Robins and Guze14suggested 5 criteria to establish the
validity of psychiatric diagnoses and illustrated their ap-
plicability to schizophrenia, namely, clinical description,
laboratory studies, delimitation from other disorders,
follow-up studies, and family studies. Kendler15devel-
oped this approach by distinguishing between anteced-
ent, concurrent, and predictive validators. However,
although the intention in devising DSM-III was to use
validity’’11including ‘‘the largest reliability study ever
done,’’16the committee chairman Robert Spitzer ac-
knowledged that ‘‘the subjective judgment of the
members of the task force . played a crucial role in
the development of DSM-III, and differences of opinion
could only rarely be resolved by appeal to objective
sis on which psychotic symptoms were required for a di-
agnosis of schizophrenia, in that patients without either
delusions or hallucinations could receive the diagnosis.
In these cases, however, other characteristic psychotic
symptoms were required, namely, gross disorganization
of speech and/or behavior. The diagnostic importance
of Schneiderian symptoms was also reemphasized, as
hallucinations can satisfy a criterion if they involve
one or more voices engaging in running commentary
or ongoing conversation, and delusions can count if
they are bizarre.18
However, to date, the DSM review process has not
used external validators such as quantitative biological
measurements or psychological testing to assist in the
evaluation of diagnostic criteria or to judge whether
changes are improving clinical validity. Furthermore, it
did not prove better than the other systems, and ulti-
mately it was the power and influence of the American
Psychiatric Association rather than any innate scientific
superiority of DSM-IV that determined that it became
most widely accepted throughout the world.
An alternative to choosing between these definitions
was to adopt a polydiagnostic approach, where several
sets of criteria were applied to the same patients.19,20
One tool was the Operation Criteria Checklist for psy-
chotic illness.21This approach uses a suite of computer
programs to generate diagnoses according to 13 different
classification systems. It has been a useful adjunct to re-
search methodology in light of the lack of a clear defini-
tion of the boundaries of schizophrenia and the wide
in everyday clinical practice.
Searching for Subtypes
Another alternative to establishing clear-cut and defen-
sible borders of schizophrenia was to suggest that it
comprised several discrete subtypes and to use external
criteria to try and validate these. The 1980s saw a number
of attempts to account for diagnostic heterogeneity by
probing for subtypes of schizophrenia, for example, pos-
itive, negative, and mixed schizophrenia22; familial and
sporadic schizophrenia23; deficit and nondeficit schizo-
phrenia24; and subtypes with some similarity to tradi-
tional hebephrenic and paranoid forms (‘‘H’’ and ‘‘P’’
subtypes).25Murray and colleagues26later sought to
Biological, Life Course, and Cross-Cultural Studies in the Classification of Psychosis
discriminate developmental from adult onset forms.
Support for their hypothesis came from latent class
analyses, but there remained the problem of intermediate
forms.27,28Furthermore, genetic and environmental
risk factors were seen to operate across diagnostic
DSM-V: A Parochial System for Use in Certain Parts of
North America or an International System?
The reader will have noticed that the above discussion
has been largely confined to proposals and papers ema-
nating from Western countries, particularly the United
States. The nosological paradigms developed to catego-
rize different types of psychotic symptoms are embedded
in specific professional cultures, but unfortunately, no-
sological discussions have rarely involved psychiatrists
working in non-Western countries. This omission would
be of little relevance to those preparing the DSM-V if
it was merely to be used in the United States. However,
the power of the American Psychiatric Association
and American Psychiatry in general has resulted in the
DSM-IV becoming the de facto system adopted by re-
searchers throughout large parts of the world, indeed
in preference to the International Classification of Dis-
eases, 10th Revision. Clearly, if DSM-V seeks to be an in-
ternational system, then it must address issues outside
those of the USA.
Research from Non-Western Countries
Sadly, much ofthe researchon psychoticconditions from
uals with psychotic conditions live—is unknown or
dismissed as methodologically flawed by nosologists
from developed countries. The substantial differences in
the onset, course, and treatment response of psychotic
identified in the international pilot study on schizophre-
nia31have had little effect on the dominant theories of
psychosis which have all been developed in Western
countries and based on data from developed countries.
by western nosologists. It is often assumed that methodo-
logical problems produce the ‘‘aberrant’’ findings, and so
no attempt is made to identify other, more complex,
the fact that experience and understanding of psychotic
symptoms are embedded in a network of local meanings
that vary from nation to nation, within different subcul-
tural groups in a single nation, and over time (as commu-
nities undergo sociocultural changes). Culture influences
an individual’s perception of the world, the content of
Thus, little attention has been paid to
their thoughts, and therefore the form and quality of psy-
chotic symptoms. It helps to determine the interpretation
of symptoms and their subsequent social impact and
guides both help seeking and the response to treatment.
At a group level, culture can be considered important
not only in defining and creating specific sources of stress
and distress but also in providing specific modes of cop-
ing with distress and the social responses to distress
A good example of subcultural differences in the atti-
tudes and help-seeking behavior of patients with schizo-
phrenia and their families comes from China where there
is a significant difference between patients from urban
and rural areas.35In rural areas, mental illness is often
associated with malevolent spirits, and therefore, many
families seek help from witch doctors. One study found
that 73.9% (N = 286 of 387) of rural psychiatry outpa-
tients admitted to previously consulting shamans,36
whereas only 4.9% (N = 21 of 426) of schizophrenia
patients from an urban area in Beijing had done so.
A separate study suggested that while families of rural
patientshadatendencytoblamethe illness on‘‘external’’
factors such as spiritual forces, family members in
urban areas were more likely to employ ‘‘internal’’ causal
explanations. These included blaming the illness on
pressure of studies, failure in love, or inability to adapt
to a new competitive environment; less commonly used
explanatory models involved physiological imbalances
and psychological problems, such as personality quirks,
excessive introversion, or nervousness.37There was also
a higher perceived effect of stigma in urban areas.
Urban patients with a young age of illness onset are
less likely to receive government-sponsored employment
and to find a spouse, and therefore, they are considered
Issues Concerning Ethnicity
An influential study carried out by the World Health
Organization was interpreted by its authors and others
to suggest that the incidence of schizophrenia was un-
varying.39However, subsequent studies have demon-
strated international, intranational, and cross-cultural
differences in rates of psychotic illness.40Furthermore,
differences in the rates of schizophrenia have also been
demonstrated for minority ethnic groups within a coun-
try. Thus, increased rates have been reported for the
diagnosis of schizophrenia in migrant groups in Demark,
France, Sweden, The Netherlands, and the United
Kingdom. A recent meta-analysis of published studies
by Cantor-Graae and Selten41has demonstrated that
different types of migrants have different risks of schizo-
phrenia (table 1).
The Curious Example of African Caribbeans in the United
The group that has been most intensively
R. Dutta et al.
studied is African Caribbeans in the United Kingdom
who show rates of psychosis several times that of the
white British population (eg, incidence rate ratios for
schizophrenia 9.1 and manic psychosis 8.0 in a recent
multicentre study40). Similarly high rates have not been
reported for other immigrant groups, and the rates of
psychosis in the Caribbean are not elevated. The in-
creased risk seems not to be due to being an immigrant
or being African Caribbean but being an immigrant from
the Caribbean living in the United Kingdom.42
The evidence is that there is a significant impact of
living or being born in the United Kingdom, which
puts those African Caribbeans already at genetic risk
of developing schizophrenia at an even greater risk.43
Genetic vulnerability and the social/environmental con-
text appear to be acting together in this cultural group
to markedly increase rates.
Are the higher rates of psychosis in the African
Caribbean UK population due to real increased rates
of schizophrenia or are they due to misdiagnosis? In
one study, a Jamaican psychiatrist was asked to make di-
agnoses on African Caribbean inpatients at a London
teaching hospital. While the UK doctors diagnosed
schizophrenia in 52% of patients and the Jamaican psy-
chiatrist diagnosed schizophrenia in 55% of patients, the
2 only agreed on the diagnosis of schizophrenia in 55% of
patients.44The results were no different whether ICD or
DSM was used. This suggests problems in the reliability
of diagnosing schizophrenia but not of racial bias in ap-
plication of diagnosis.45
The difficulty in categorizing psychiatric illness is
further underlined by differences in the course of schizo-
phrenia between the African Caribbean community and
native whites in the United Kingdom. African Carib-
beans are approximately 40% less likely to suffer from
a continuous illness than British whites,46and it is
suggested that they are less likely to have a history of
obstetric complications or neurological illness pre-
morbidly.47It hasbeen hypothesizedthat the goodsymp-
tomatic prognosis reflects increased rates of illness in less
neurologically and genetically vulnerable people who
have had relatively normal early development but have
been exposed to social stressors that have promoted
psychosis. One possible contributing factor is racial dis-
crimination. Studies show that the darker the skin color,
the more racism an individual is subject to regardless of
mental illness.48One longitudinal study has demon-
strated that those who experience discrimination are at
an increased risk of developing delusional ideation.49
The lesson of these studies is that there may be a different
balance of causes of psychosis, a different spectrum of
Findings from Recent Biological Studies
Evidence that schizophrenia and bipolar disorder are not
from studies showing that antipsychotics are effective in
mood stabilizers was taken as a feature classically dis-
tinguishing it from schizophrenia. Recently, however,
significant reduction in the severity of symptoms was ob-
served in patients with an acute exacerbation of schizo-
phrenia in whom divalproex was added in to olanzapine
or risperidone treatment.51This builds upon earlier
work by Brockington52which showed that lithium and
chlorpromazine were equally effective in schizoaffective
patients and detracts from a notion that there are distinct
psychotic disorders with unique treatment pathways.
Schizophrenia and bipolar disorder occur together in the
in a twin study using blinded diagnostic assessments
and relaxing the normal hierarchical approach whereby
schizophrenia trumps all other diagnoses, Cardno et al53
showed that if one member of a monozygotic twin pair
has schizophrenia, there is about an 8% chance of the
co-twin being diagnosed with schizoaffective disorder
thermore, as discussed elsewhere in this volume, recent
molecular genetic studies, although as yet preliminary,
suggest overlap between risk genes for schizophrenia
and bipolar disorder.54
Brain morphometry studies have shown that schizophre-
nia is associated with distributed gray matter deficits
particularly in the frontotemporal neocortex, medial
temporallobe, insula,thalamus, and cerebellum,whereas
patients with bipolar disorder have no significant areas
of gray matter abnormality. However, both disorders
in regions normally occupied by major longitudinal and
Table 1. Based on Published Meta-analyses of Population-Based
Studies Examining the Association Between Migration and Risk
Risk 95% CI
First-generation migrants2.7 2.3–3.2
Second-generation migrants 4.51.5–13.1
Migrants with ‘‘black’’ skin color 4.83.7–6.2
Migrants with ‘‘white’’ skin color2.3 1.7–3.1
Biological, Life Course, and Cross-Cultural Studies in the Classification of Psychosis
A Developmental Perspective
Thus, pharmacological, genetic, and neuroimaging stud-
ies suggest both similarities and differences between
schizophrenia and bipolar disorder. Some understanding
of the basis of these comes from adopting a life course
perspective on the illnesses. Numerous studies have
shown that preschizophrenic children are characterized
by impairments in cognitive and neuromotor develop-
ment. This was demonstrated very clearly in the Dunedin
study which was also the first to demonstrate that
these are not a feature of those who later develop bipolar
neurocognitive impairment is provided by the Israeli
Draft Board Registry study,57which showed that 68 indi-
viduals hospitalized with bipolar disorder did not differ
from their healthy matched counterparts on any test
of intellectual, language, or behavioral functioning con-
ducted routinely when they were adolescents. A more re-
cent cohort study using national registers to follow all
Swedish children who completed compulsory education
showed that no students with excellent school perfor-
mance developed schizophrenia or schizoaffective disor-
der. By contrast, achieving outstanding grades in certain
school subjectswas asignificant predictoroflaterbipolar
Further evidence that schizophrenia and bipolar disor-
ders are at least partially distinct in etiology comes
from studying complications of pregnancy and delivery.
Obstetric events have been described as being more fre-
quent in schizophrenia.59,60Perinatal hypoxia arising
from birth complications is particularly known to affect
growth of the amygdala and hippocampus, which are
often reported to be smaller in schizophrenia and not in
bipolar disorder.61There is no substantive evidence that
obstetric complications increase the risk of bipolar disor-
der.62Moreover, fetal growth indicators such as birth
weight, birth length, and gestational age have also not
been identified as risk factors for bipolar disorder.63
The similarities and differences between schizophrenia
and bipolar disorder begin to suggest a model (figure 1)
in which given a shared background of genetic predispo-
sition to psychosis, additional specific genetic or early en-
leaving individuals vulnerable to schizophrenia. By con-
trast, in bipolar disorder, developmental impairment is
absent but syndrome-specific genes and environmental
interactions may render individuals susceptible to social
A Dimensional Perspective
Traditionally, first-rank symptoms are given particular
emphasis for making a diagnosis of schizophrenia rather
than bipolar disorder. However, although Cardno and
colleagues64showed that a syndrome characterized by
the presenceof oneor morefirst-rank symptomshas con-
siderable heritability (71%, 95% confidence interval 57–
82, compatible with a genetic contribution to variance in
liability), it remains somewhat lower than that for schizo-
phrenia as defined by established classifications, includ-
ing DSM criteria.
An alternative to considering syndrome-based ap-
proaches to psychopathology is to use identified groups
of correlated symptoms (symptom dimensions) in patient
have extracted usually 4 or 5 different factors or dimen-
sions (eg, depressive, manic, positive, negative, and dis-
organization symptoms), and broadly these have been
Recently, it has been shown that using such symptom
dimensions explains more about disease characteristics
(such as premorbid impairment, the existence of stressors
before disease onset, poor remissions or no recovery be-
tween episodes and exacerbations, response to neurolep-
tics, and deterioration) than diagnoses alone and thus
adds substantial information to diagnostic categories.66
Psychosis as a Dimension Reaching Into the General
Various groups have in recent years pointed out that
minor psychotic symptoms occur in the general popula-
tion67–69and that psychosis is best conceived as a dimen-
sion like hypertension rather than a distinct category.
(Refer to the review of Allardyce, van Os, and Gaebel70
for further discussion of dimensional representations
of psychotic illness.) Further evidence also comes
from studies of those at ultra high risk of developing
There is ample evidence that psychosis is ‘‘brewing’’
long before its manifestation as a diagnosable illness71
Distinctions Between Schizophrenia and Bipolar Disorder (after
Cardno et al53and Murray et al61).
R. Dutta et al.
and that identifiable signs and symptoms preceding the
development of frank psychotic symptoms are evident.72
DSM-IV criteria for schizophrenia include this ‘‘prodro-
mal phase’’ as a construct, but it describes a retrospective
concept because it cannot be defined until there is an
established psychotic illness. DSM-III11identified 9
symptoms considered to be ‘‘prodromal’’ for schizophre-
nia and included them as diagnostic contributors.
However, in a study by the Melbourne group based on
retrospective conceptualization, these 9 symptoms were
itive predictive values between 0.36 and 0.48 but were not
pathognomic of schizophrenic psychosis.73
Indeed, in one study, Yung and colleagues74reported
developed psychosis, diagnoses ranged from schizophre-
nia, through schizoaffective disorder, brief psychotic dis-
‘‘ultra high-risk’’ criteria, it appears as if early signs and
symptoms are predictive of conversion to a spectrum of
psychotic disorders but not of the exact nature of the psy-
chosis that will develop.
It seems that the final diagnosis of a psychotic illness is
merely the endpoint of a risk pathway which in itself is
a slippery slope but not inevitable trajectory into psycho-
sis (figure 3); this view is very compatible with the dimen-
the pathway includes the development of prepsychotic
symptoms, the development of frank but infrequent
psychotic symptoms, the development of persistent psy-
chotic symptoms, and finally social impairment due to
these psychotic symptoms. Moving up or down the path-
way depends on a balance between propsychotic factors
such as individual biological vulnerability, the use of
cannabis, and the social environment and antipsychotic
factors such as individual resilience.
A Scheme Incorporating Developmental and Dimensional
Ratings Offers a Possible Way Forward
schizophrenia within North America, considerably more
in Europe, and psychiatrists from the developing world
regard it as largely ignoring the issues of 3 quarters of
the globe. Difficulties in diagnosing mental illness among
ethnic minority groups highlight the need for a universal
ever, the difference in rates of psychotic illness between
countries and among different ethnic groups within a
country also suggest that viewing culture and ethnicity
as confounding variables in the conceptualization of
mental illness is misguided. Rather, culture and ethnicity
ought to be seen as fundamental elements driving its ex-
pression and interpretation.
By considering psychotic disorders from a life course
perspective, including genetic factors, neurodevelopmen-
tal distinctions, symptomatology, structural neuroimag-
ing, treatment strategies, and groups at ultra high risk
of psychosis, we can see that a scheme which takes
into consideration both developmental and dimensional
characteristics as discussed above appears a possible way
forward. For example, those at ultra high risk of psycho-
sis would be rated at points on dimensions compatible
with the extent and severity of their psychotic symptoms
and affective symptoms. Whether or not they showed
dict the clinical picture of a full-blown psychosis if
and when it developed. Again, as applied to African
Caribbeans with psychosis in the United Kingdom,
Fig. 3. A Risk Pathway to the Diagnosis of Psychosis.
Fig. 2. Schema Incorporating 5 Dimensions (after Van Os et al78) and Explaining the ‘‘Spectrum’’ of Syndromes from Schizophrenia
Through to Bipolar Disorder.
Biological, Life Course, and Cross-Cultural Studies in the Classification of Psychosis
such a model would suggest that this population is more
vulnerable to a largely nondevelopmental illness in which
social etiological factors are particularly important and
which may present with a mixture of schizophrenic and
However, whether diagnoses are based on symptom
dimensions or diagnostic categories, the instruments
for rating symptoms have typically been developed by
selecting a subset of useful items from a large preliminary
pool of items based on the results of a series of studies
involving subjects in Western countries. If the entire pro-
cess was repeated in a non-Western country, it would
almost inevitably result in a very different instrument
with different items and a different factor structure.
For example, studies in China on symptom scales in
schizophrenia75have clearly demonstrated that trans-
lated and back-translated instruments can often achieve
satisfactory test-retest reliability, but substantial revision
is needed in order to achieve internal consistency and
Another problem seen in the use of western diagnostic
instruments in developing countries is the assumption
that a single probe is sufficient to elicit a particular symp-
tom; this is particularly problematic in fully structured
diagnostic instruments that do not allow the interviewer
to revise the question based on the educational and cul-
tural background of the respondent. This single-probe
method may work in developed countries where the ex-
perience and expression of psychological symptoms has
been ‘‘homogenized’’ by frequent media exposure and
other social forces; but for example in China, the huge
sociocultural differences between urban and rural resi-
ture the different methods of experiencing and describing
specific psychological symptoms.76
Thus, if the DSM-V system of classifying psychosis is
to be relevant to patients in the developing world, then
instruments aimed at either making diagnoses or rating
symptoms have to be subject to much more sophisticated
field studies in non-Western countries than hitherto.
Proposal of a Hybrid System
It is clear that the categories of psychosis as used cur-
rently in DSM-IV are not valid in a strictly scientific
sense. Their replacement by a developmental and dimen-
sional approach as outlined above has much to recom-
mend it for DSM-V. However, the current system does
have some utility in terms of the information about eti-
ology, course of illness, outcome, and treatment response
that the different diagnoses convey.77Abandoning it
would be a very dramatic shift, and although we believe
it would be an advance, some information of benefit to
patients and clinicians would be lost.
We consider that at present the best option is to imple-
ment a hybrid of a categorical-dimensional approach in
DSM-V. This would introduce the benefit of increased
explanatory power of clinical characteristics, without
completely dismissing the traditional paradigm of the
Kraepelinian dichotomy. Similarly, including a rating
of developmental impairment would aid understanding
of the longitudinal course of illness evolution, rather
than considering a diagnosis as a cross-sectional perspec-
tive based only on the current clinical picture. Anything
more radical is likely to be premature, with the expecta-
tion of further advances in genetic, neurobiological, en-
vironmental, and psychosocial research in the coming
In parallel with research in individual disciplines, what
is needed is a concerted multicenter effort to look back at
existing epidemiologically based first-onset psychosis
cohorts to investigate how external summary variables,
including measures of cognition, social variables, and
need for care, as well as symptom dimensions, familial
liability scores, and basic structural magnetic resonance
imaging data may sharpen the discriminative potential
of the DSM classification of psychotic disorders. This
should include cohort data from both developing as
well as developed countries.
From our exploration of cultural issues, we suggest
that standardized qualitative and quantitative methods
need to be developed that can be employed in a wide
range of different communities to conduct culturally sen-
sitive assessments of psychotic symptoms. Only then will
it be possible for the nosologist to attempt to identifyuni-
versal ‘‘gold standard’’ criteria (preferably with unique
biological and psychosocial markers) for a discrete set
of psychotic diagnoses.
1. American Psychiatric Association. Diagnostic and Statistical
Manual of Mental Disorders. 4th ed. Washington, DC:
American Psychiatric Association; 1994.
2. Brockington IF. Schizophrenia: yesterday’s concept. Eur
3. Bentall RP, Jackson HF, Pilgrim D. Abandoning the concept
of ‘schizophrenia’: some implications of validity arguments
for psychological research into psychotic phenomena. Br J
Clin Psychol. 1988;27:303–324.
4. van Os J, Verdoux H. Diagnosis and classification of schizo-
phrenia: categories versus dimensions, distributions versus
disease. In: Murray RM, Jones PB, van Os J, Cannon M,
eds. The Epidemiology of Schizophrenia. Cambridge, Mass:
University Press; 2003:364–410.
5. Kraepelin E. The manifestations of insanity [Die Erschei-
nungsformen des Irreseins]. Hist Psychiatry. 1992;3:509–529.
7. Boyle M. Schizophrenia—a scientific delusion?. London:
8. Brockington IF, Kendell RE, Leff JP. Definitions of schizo-
phrenia: concordance and prediction of outcome. Psychol
R. Dutta et al.
9. Kendell RE, Brockington IF, Leff JP. Prognostic implica-
tions of six alternative definitions of schizophrenia. Arch
Gen Psychiatry. 1979;36:25–31.
10. Endicott J. Diagnostic criteria for schizophrenia: reliabilities
and agreement between systems. Arch Gen Psychiatry. 1982;
11. American Psychiatric Association.Diagnostic and Statistical
Manual of Mental Disorders. 3rd ed. Washington, DC: Amer-
ican Psychiatric Association; 1980.
12. Castle DJ, Wessely S, van Os J, Murray RM. Diagnostic
issues and admission policies. In: Psychosis in the Inner
City: The Camberwell First Episode Study. London: Maudsley
13. Helzer JE, Brockington IF, Kendell RE. Predictive validity of
DSM-III and Feighner definitions of schizophrenia: a com-
parison with research diagnostic criteria and CATEGO.
Arch Gen Psychiatry. 1981;38:791–797.
14. Robins E, Guze SB. Establishment of diagnostic validity
in psychiatric illness: its application to schizophrenia. Am
J Psychiatry. 1970;126:983–987.
15. Kendler KS. The nosologic validity of paranoia (simple delu-
16. Spitzer RL, Williams JB, Skodol AE. DSM-III: the major
achievements and an overview. Am J Psychiatry. 1980;
17. Spitzer RL. DSM-III and the politics-science dichotomy syn-
drome: a response to Thomas E. Schacht’s ‘‘DSM-III and the
politics of truth’’. Am Psychol. 1985;40:522–526.
18. Tsuang MT, Stone WS, Faraone SV. Toward reformulating
the diagnosis of schizophrenia. Am J Psychiatry. 2000;157:
19. Berner P, Katschnig H, Lenz G. Poly-diagnostic approach: a
method to clarify incongruences among the classification of
the functional psychoses. Psychiatr J Univ Ott. 1982;7:244–248.
20. Jansson LB, Parnas J. Competing definitions of schizophre-
nia: what can be learned from polydiagnostic studies?
Schizophr Bull. December 8, 2006; doi:10.1093/schbul/sbl065.
21. Farmer AE, Wessely S, Castle D, McGuffin P. Methodolog-
ical issues in using a polydiagnostic approach to define
psychotic illness. Br J Psychiatry. 1992;161:824–830.
22. Andreasen NC, Olsen S. Negative v positive schizophrenia.
Definition and validation. Arch Gen Psychiatry. 1982;39:
23. Murray RM, Lewis SW, Reveley AM. Towards an aetiolog-
ical classification of schizophrenia. Lancet. 1985;1:1023–1026.
Wood F. Strong inference, theory testing, and the neuroanat-
omy of schizophrenia. Arch Gen Psychiatry. 1993;50:825–831.
25. Farmer AE, McGuffin P, Gottesman II. Searching for the
split in schizophrenia: a twin study perspective. Psychiatry
26. Murray RM, O’Callaghan E, Castle DJ, Lewis SW. A neuro-
developmental approach to the classification of schizophre-
nia. Schizophr Bull. 1992;18:319–332.
27. Castle DJ, Sham PC, Wessely S, Murray RM. The subtyping
of schizophrenia in men and women: a latent class analysis.
Psychol Med. 1994;24:41–51.
28. Sham PC, Castle DJ, Wessely S, Farmer AE, Murray RM.
Further exploration of a latent class typology of schizophre-
nia. Schizophr Res. 1996;20:105–115.
29. Done DJ, Crow TJ, Johnstone EC, Sacker A. Childhood
antecedents of schizophrenia and affective illness: social ad-
justment at ages 7 and 11. BMJ. 1994;309:699–703.
30. van Os J, Jones P, Lewis G, Wadsworth M, Murray R. Devel-
opmental precursors of affective illness in a general popula-
tion birth cohort. Arch Gen Psychiatry. 1997;54:625–631.
31. Sartorius N, Gulbinat W, Harrison G, Laska E, Siegel C.
Long-term follow-up of schizophrenia in 16 countries.
A description of the International Study of Schizophrenia
conducted by the World Health Organization. Soc Psychiatry
Psychiatr Epidemiol. 1996;31:249–258.
32. Susser E, Wanderling J. Epidemiology of nonaffective acute
remitting psychosis vs schizophrenia. Sex and sociocultural
setting. Arch Gen Psychiatry. 1994;51:294–301.
33. Alarcon RD, Westermeyer J, Foulks EF, Ruiz P. Clinical
relevance of contemporary cultural psychiatry. J Nerv Ment
34. Kirmayer LJ, Young A. Culture and context in the evolution-
ary concept of mental disorder. J Abnorm Psychol. 1999;108:
35. Phillips MR. The transformation of China’s mental health
services. China J. 1998;39:1–36.
36. Li SX, Phillips MR. Witch doctors and mental illness in
mainland China: a preliminary study. Am J Psychiatry. 1990;
37. Phillips MR, Li Y, Stroup TS, Xin L. Causes of schizophrenia
reported by patients’ family members in China. Br J Psychi-
38. Phillips MR, Pearson V, Li F, Xu M, Yang L. Stigma and
expressed emotion: a study of people with schizophrenia
and their family members in China. Br J Psychiatry. 2002;
39. Jablensky A, Sartorius N, Ernberg G, et al. Schizophrenia:
manifestations, incidence and course in different cultures.
A World Health Organization ten-country study. Psychol
Med Monogr Suppl. 1992;20:1–97.
40. Fearon P, Kirkbride JB, Morgan C, et al. Incidence of schizo-
phrenia and other psychoses in ethnic minority groups: results
from the MRC AESOP study. Psychol Med. 2006;1–10.
41. Cantor-Graae E, Selten JP. Schizophrenia and migration:
a meta-analysis and review. Am J Psychiatry. 2005;162:12–24.
42. Fearon P, Morgan C. Environmental factors in schizophrenia:
the role of migrant studies. Schizophr Bull. 2006;32:405–408.
43. Hutchinson G, Takei N, Fahy TA, et al. Morbid risk of
schizophrenia in first-degree relatives of white and African-
Caribbean patients with psychosis. Br J Psychiatry. 1996;169:
44. Hickling FW, McKenzie K, Mullen R, Murray R. A Jamai-
can psychiatrist evaluates diagnoses at a London psychiatric
hospital. Br J Psychiatry. 1999;175:283–285.
45. Sharpley M, Hutchinson G, McKenzie K, Murray RM.
Understanding the excess of psychosis among the African-
Caribbean population in England. Review of current hypoth-
eses. Br J Psychiatry Suppl. 2001;40:s60–s68.
46. McKenzie K, Samele C, van Horn E, Tattan T, van Os J,
Murray R. Comparison of the outcome and treatment of psy-
chosis in people of Caribbean origin living in the UK and
British Whites: report from the UK700 trial. Br J Psychiatry.
47. McKenzie K, Jones P, Lewis S, et al. Lower prevalence of
pre-morbid neurological illness in African-Caribbean than
White psychotic patients in England. Psychol Med. 2002;32:
48. Klonoff EA, Landrine H. Is skin color a marker for racial
discrimination? Explaining the skin color-hypertension rela-
tionship. J Behav Med. 2000;23:329–338.
Biological, Life Course, and Cross-Cultural Studies in the Classification of Psychosis
49. Janssen I, Hanssen M, Bak M, et al. Discrimination and
delusional ideation. Br J Psychiatry. 2003;182:71–76.
50. Post RM. Comparative pharmacology of bipolar disorder
and schizophrenia. Schizophr Res. 1999;39:153–158.
51. Casey DE, Daniel DG, Wassef AA, Tracy KA, Wozniak P,
Sommerville KW. Effect of divalproex combined with
olanzapine or risperidone in patients with an acute exacerba-
tion of schizophrenia. Neuropsychopharmacology. 2003;28:
52. Brockington IF, Kendell RE, Kellett JM, Curry SH,
Wainwright S. Trials of lithium, chlorpromazine and
amitriptyline in schizoaffective patients. Br J Psychiatry.
53. Cardno AG, Rijsdijk FV, Sham PC, Murray RM, McGuffin
P. A twin study of genetic relationships between psychotic
symptoms. Am J Psychiatry. 2002;159:539–545.
54. Craddock N, O’Donovan MC, Owen MJ. Genes for schizo-
phrenia and bipolar disorder? Implications for psychiatric
nosology. Schizophr Bull. 2006;32:9–16.
55. McDonald C, Bullmore E, Sham P, et al. Regional volume
deviations of brain structure in schizophrenia and psychotic
bipolar disorder: computational morphometry study. Br J
56. Cannon M, Caspi A, Moffitt TE, et al. Evidence for
early-childhood, pan-developmental impairment specific to
schizophreniform disorder: results from a longitudinal birth
cohort. Arch Gen Psychiatry. 2002;59:449–456.
57. Reichenberg A, Weiser M, Rabinowitz J, et al. A population-
based cohort study of premorbid intellectual, language,
and behavioral functioning in patients with schizophrenia,
schizoaffective disorder, and nonpsychotic bipolar disorder.
Am J Psychiatry. 2002;159:2027–2035.
58. Maccabe J, Lambe M, Cnattingius S, et al. Academic achieve-
ment at age 16 has contrasting effects on risk of later bipolar
disorder and schizophrenia. Schizophr Res. 2006;81(suppl):
59. Geddes JR, Verdoux H, Takei N, et al. Schizophrenia and
complications of pregnancy and labor: an individual patient
data meta-analysis. Schizophr Bull. 1999;25:413–423.
60. Cannon M, Jones PB, Murray RM. Obstetric complications
and schizophrenia: historical and meta-analytic review. Am
J Psychiatry. 2002;159:1080–1092.
61. Murray RM, Sham P, Zanelli J, Cannon M, McDonald C. A
developmental model for similarities and dissimilarities be-
62. Scott J, McNeill Y, Cavanagh J, Cannon M, Murray R.
Exposure to obstetric complications and subsequent develop-
ment of bipolar disorder: systematic review. Br J Psychiatry.
63. Ogendahl BK, Agerbo E, Byrne M, Licht RW, Eaton WW,
a Danish national register-based study. Psychol Med. 2006;
64. Cardno AG, Sham PC, Farmer AE, Murray RM, McGuffin
P. Heritability of Schneider’s first-rank symptoms. Br J Psy-
65. van Os J, Gilvarry C, Bale R, et al. A comparison of the
utility of dimensional and categorical representations of
psychosis. UK700 Group. Psychol Med. 1999;29:595–606.
66. Dikeos DG, Wickham H, McDonald C, et al. Distribution of
symptom dimensions across Kraepelinian divisions. Br J Psy-
67. van Os J, Hanssen M, Bijl RV, Ravelli A. Strauss (1969)
revisited: a psychosis continuum in the general population?
Schizophr Res. 2000;45:11–20.
68. Johns LC, Cannon M, Singleton N, et al. Prevalence and cor-
relates of self-reported psychotic symptoms in the British
population. Br J Psychiatry. 2004;185:298–305.
69. Wiles NJ, Zammit S, Bebbington P, Singleton N, Meltzer H,
Lewis G. Self-reported psychotic symptoms in the general
population: results from the longitudinal study of the British
National Psychiatric Morbidity Survey. Br J Psychiatry. 2006;
70. Allardyce J, Gaebel W, Zielasek J, van Os J. Deconstructing
Psychosis Conference February 2006: The Validity of Schizo-
phrenia and Alternative Approaches to the Classification of
Psychosis. Schizophr Bull. June 4, 2007;doi:10.1093/schbul/
71. Poulton R, Caspi A, Moffitt TE, Cannon M, Murray R,
Harrington H. Children’s self-reported psychotic symptoms
and adult schizophreniform disorder: a 15-year longitudinal
study. Arch Gen Psychiatry. 2000;57:1053–1058.
72. McGlashan TH, Zipursky RB, Perkins D, et al. The PRIME
North America randomized double-blind clinical trial of
olanzapine versus placebo in patients at risk of being pro-
dromally symptomatic for psychosis. I. Study rationale and
design. Schizophr Res. 2003;61:7–18.
73. Jackson HJ, McGorry PD, Dudgeon P. Prodromal symptoms
of schizophrenia in first-episode psychosis: prevalence and
specificity. Compr Psychiatry. 1995;36:241–250.
74. Yung AR, Phillips LJ, Yuen HP, et al. Psychosis prediction:
12-month follow up of a high-risk (‘‘prodromal’’) group.
Schizophr Res. 2003;60:21–32.
75. Phillips MR, Xiong W, Wang RW, Gao YH, Wang XQ,
Zhang NP. Reliability and validity of the Chinese versions
of the Scales for Assessment of Positive and Negative Symp-
toms. Acta Psychiatr Scand. 1991;84:364–370.
76. Phillips MR, Shen QJ, Liu XH, et al. Assessing depressive
symptoms in persons who die of suicide in mainland China.
J Affect Disord. 2007;98:73–82.
77. Kendell R, Jablensky A. Distinguishing between the validity
and utility of psychiatric diagnoses. Am J Psychiatry. 2003;
78. van Os J, Gilvarry C, Bale R, et al. Diagnostic value of the
DSM and ICD categories of psychosis: an evidence-based
approach. UK700 Group. Soc Psychiatry Psychiatr Epide-
R. Dutta et al.