Advances in psychiatric treatment (2014), vol. 20, 280–285 doi: 10.1192/apt.bp.113.

011296

article A comparison of DSM and ICD

classifications of mental disorder
Peter Tyrer

Peter Tyrer is Professor of the disorder without intervention), and a guide to

Community Psychiatry in the Summary
treatment and outcome. Without a classification
Centre for Mental Health in Most disorders in medicine are classified using
the Division of Medicine at system the necessary economical communication
the ICD (initiated in Paris in 1900). Mental and
Imperial College London. His with colleagues to convey information becomes
main interests are in models of
behavioural disorders are classified using
a lengthy description of clinical problems that is
delivering community psychiatric the DSM (DSM-I was published in the USA in
1952), but it was not until DSM-III in 1980 that it self-defeating.
services, the classification and
treatment of common mental became a major player. Its success was largely The difficulties of classification in psychiatry are
illnesses, particularly anxiety influenced by Robert Spitzer, who welded its many and the reasons why we have arguments on
and health anxiety, and the disparate elements, and Melvyn Shabsin, who the subject and why others in medicine do not to
classification and management
facilitated its acceptance. Spitzer pointed out the same degree is summarised in Box 1.
of personality disorders.
Correspondence Professor Peter that most diagnostic conditions in psychiatry
Tyrer, Centre for Mental Health, were poorly defined, showed poor reliability Biological markers
Department of Medicine, Imperial in test–retest situations, and were temporally
The aim of psychiatric classification for many
College London, St Dunstan’s Road, unstable. The consequence was that the beliefs
London W6 8RP, UK. Email: p.tyrer@ of the psychiatrist seemed to matter much more years has been to find an independent set of bio­
imperial.ac.uk
than the characteristics of the patient when it logical or pathophysiological markers that indicate
came to classification. Since DSM-III there has the presence of disease and can therefore confirm
been a split between those who adhere to DSM clinical impressions. This is a perfectly natural
because it is a better research classification aim, as it is following in the footsteps of medicine,
and those who adhere to ICD because it allows where almost every disorder receives respectabil­
more clinical discretion in making diagnoses. ity from having such an independent biological
This article discusses the pros and cons of both basis. Nobody is disputing that every disorder in
systems, and the major criticisms that have been psychiatry has some sort of biological associates,
levelled against them.
but for most of them it is of no value in diagnosis
Learning objectives because our current knowledge is inadequate. The
• Understand the principles and reasoning behind problem is mainly lack of specificity and impaired
classification in medicine and psychiatry. understanding of the fundamental biological
• Be able to describe the recent history of psy­ processes involved in much of mental disorder.
chiatric classification.
• Be able to compare DSM and ICD classifications
of mental disorder. Box 1 Differences between diagnosis of
psychiatric and medical disorders
Declaration of interest
P.T. is Chair of the ICD-11 Working Group for the Psychiatric disorders
Revision of Classification of Personality Disorders • Biology: very few psychiatric disorders have a biological
and has also been a member of the DSM–ICD basis that confirms clinical impressions of disease
Harmonization Coordination Group. • ‘Zone of rarity’ between health and disease rarely
present in psychiatric disorders
• Uncertain threshold of diagnosis
Although diagnosis in psychiatry leads to a lot of
argument – and the extent of this dis­satisfaction is Medical disorders
expressed by an international campaign initiated • Most diagnoses in medicine have a biological basis
by Sami Timimi (2011) on the Critical Psychiatry that can confirm disease (e.g. blood pressure in
Network UK to abolish the ICD and DSM hypertension, blood sugar in diabetes)
classifications altogether – most psychiatrists • ‘Zone of rarity’ between health and disease more often,
regard classification as having some value. A but not always, present
good class­ification provides information about the • Better thresholds for diagnosis (usually because
causes of disorder, possible prevention, clinical independent tests can confirm clinical impressions)
characteristics, natural history (i.e. the course of

280
https://doi.org/10.1192/apt.bp.113.011296 Published online by Cambridge University Press
 Comparison of DSM and ICD classifications of mental disorder

To take one example among many, it has been Take the example illustrated in Fig. 1.
found that many individuals with obsessive– Depression is commonly classified by a score on
compulsive dis­order (OCD) have post-streptococcal a standard inventory or by the number of criteria
auto­immunity autoantibodies related to possible needed to cross the threshold into diagnosis. But
infection in childhood (Dale 2005). In this study, every clinician and researcher knows that the
42% of children with OCD had these auto­ threshold is not a clean one – there are many
antibodies, but no one would suggest that testing people just below the threshold, who have many of
for autoantibodies would be a suitable diagnostic the symptoms of depression but who do not quite
test for OCD, not least as up to 10% of controls also qualify for diagnosis. There is now abundant
had these autoantibodies. research that shows that these people are, not
The same does not apply in medicine for most surprisingly, as unwell as others who are just over
disorders. If you develop lobar pneumonia, X-rays the threshold line (Ayuso-Mateos 2010).
will confirm that you have consolidation in your By contrast, the diagnosis of iron-deficiency
lung and analysis of the sputum will confirm and pernicious anaemia depends on analysis of
that you have a pneumococcal infection. When data from red blood cells that are much more
identified, these abnormalities are indisputable, specific. There is a zone of rarity between those
and the clinical examination of the chest that leads who have macrocytic anaemia (large corpuscles
towards the diagnosis of lobar pneumonia is only characteristic of vitamin B12 deficiency and iron-
a preliminary diagnostic exercise that remains to deficiency corpuscles) and between those who have
be confirmed by the X-ray and sputum analysis. normal blood cells (hence the three groupings in
In psychiatry, we do not have the insight of Fig. 1). If we just relied on clinical examination,
independent tests for most of our disorders, but the separation of these anaemias would be much
this does not mean they do not exist. We do more difficult and might approximate to the data
have independent measures for conditions such for depression. Because we have no equivalent
as Alzheimer’s disease and Down syndrome, biomarker for depression, there is no zone of rarity.
but these are rare compared with more common It is also relevant that for many chronic medical
conditions such as anxiety, depression and conditions, such as osteoarthritis, Parkinson’s dis­
personality disorder. Readers will note from ease and obstructive airways disease, there is no
looking at psychiatric journals that there are zone of rarity either, so although it is an excellent
increasing numbers of neuroanatomical and illustration for disorders such as Huntington’s cho-
imaging studies that find abnormalities in people rea and porphyria, the more common conditions
with different psychiatric disorders, but in none that flesh is heir to do not share this distinction.
of these is the abnormality clearly diagnostic (i.e.
it occurs in every person with the disorder and is Uncertain threshold of diagnosis
absent in those who do not have the disorder). The absence of a zone of rarity indicates that
As a consequence of this lack of independent the threshold for psychiatric diagnosis is usually
evidence, there are many more models of mental arbitrary. When psychiatrists make a decision
disorder than models of physical disorder (Tyrer about a clinical diagnosis they therefore have no
2013). This explains a lot of the controversy that
exists within psychiatry – controversy that is
18 Iron-binding Depression
generally lacking in medicine but still emerges Macrocytes capacity Lymphocytes
16 Two anaemias
when trying to find diagnostic cut-off points for the
treatment of conditions such as hypertension and 14
Proportion of abnormal features

metabolic syndrome. Unfortunately, the presence 12

of many different models still includes an anti- 10
psychiatry one that abhors diagnosis altogether, 8
with only the disease of labelling being allowed to 6
show itself. But this notion has to be put to rest; 4
psychiatry without diagnosis will return us to the 2
Dark Ages and has to be embraced in some form 0
(Craddock 2014). Zone Zone
–2 of rarity of rarity
Pernicious Iron-deficiency Lymphatic
Zone of rarity anaemia anaemia leukaemia
Introduced by Robert Kendell (1968), the ‘zone of Schematic diagram showing the differences based on diagnostic tests between three
fig 1
rarity’ refers to the hiatus between the features of common medical diag­noses, all of which cause anaemia (pernicious anaemia, iron-
a biological disorder with a clear diagnosis and deficiency anaemia and lymphatic leukaemia) and which can be detected by simple blood
other conditions that do not carry this diagnosis. analysis, and the diagnosis of depression, which is based only on common symptoms.

Advances in psychiatric treatment (2014), vol. 20, 280–285 doi: 10.1192/apt.bp.113.011296 281
https://doi.org/10.1192/apt.bp.113.011296 Published online by Cambridge University Press
 Tyrer

guidance in deciding on the cut-off point between because it used operational criteria (derived
disease and wellness. Almost all psychiatrists from psychiatric diagnostic criteria proposed 8
create an artificial boundary between disorder and years earlier – the Feighner criteria; Feighner
normality. Increasingly, it has been recognised 1972). The insight given to Spitzer and colleagues
that a dimensional system of diagnosis is therefore (1975) by this earlier work was that psychiatric
superior to a categorical one, but this is only diagnosis could never be clinically valid or useful
beginning to penetrate into diagnostic systems. if it was unreliable, and the introduction of clear
operational criteria in DSM-III undoubtedly
The DSM and ICD classifications improved reliability.
It is important to remember that the word
Background
‘reliable’ in this context is a statistical concept,
The ICD is the official world classification. The not in the general sense of being trustworthy and
section concerned with psychiatric disorders is consistent. Statistical reliability is the extent to
called ‘Mental and Behavioural Disorders’. This which one assessor will agree with another, and
classification is used to record the diagnoses of all it is a self-evident truth that you will improve
patients seen in psychiatric care across the world agreement between assessors when you give clear
where official statistics are collected. By contrast, definitions of disorder rather than vague ones.
the DSM is the official classification in the USA Reliability includes both interrater reliability
for clinical diagnosis, although its influence (assessments by different people at the same point
now covers the globe, particularly because of its in time) and temporal reliability (assessments
apparent advantages for research and the general carried out at different points in time). Good
belief that it is in some way ‘more accurate’. interrater and temporal reliability were the
driving forces behind much of the work of the St
Important dates Louis group of researchers under Samuel Guze
The history of the official classification of disease in the 1960s and 1970s (Guze 1971, 1975), and
began with two Frenchmen, Achille Guillard, who these were the precursors of the Feighner criteria.
made the first attempt at a worldwide classification These specific diagnostic criteria have dominated
in 1853, and his grandson, Jacques Bertillon, American psychiatric classification ever since.
Chief of Statistics for the City of Paris, who, under By contrast, the ICD classification abhors clear
the auspices of the French government, convoked diagnostic criteria unless they are independently
the first International Conference for the Revision validated, and allows the clinician to make
of the International List of Causes of Death in judgements in the classification of disorder.
Paris on 18 August 1900. This became the ICD Although this almost inevitably leads to less
and has since gone through ten revisions, with agreement in diagnosis, it should not automatically
ICD-10 the latest to be published (World Health be regarded as less reliable (i.e. less accurate).
Organization 1992). The gap between ICD-10 and It could be argued that good clinical judgement
ICD-11 is likely to be 24 years, the longest between deserves to have a place in any classification –
successive revisions since 1900. DSM has been what is the point of training and experience if it is
revised more frequently, with successive revisions not? – and so this is allowed much more space in
in 1987, 1994, 2000 and, more recently, in May the ICD classification than the DSM.
2013 (American Psychiatric Association 1987, It is very important to realise that reliability in
1994, 2000, 2013). itself does not mean a disorder is better described
The DSM was first introduced in 1952 after the or more valid in actually measuring what it
US military decided they needed to have a useful purports to measure. It is always possible to get
classification of mental disorders. It was not widely good agreement by having tight definitions of
used initially, but with the arrival of Robert the condition concerned, but if it is not properly
Spitzer, who coordinated the development of the described its value will be equally limited. One
third revision (DSM-III; American Psychiatric obvious example of reliability in the absence of
Association 1980), it became almost de rigeur to validity would be a survey of members of the Flat
use the classification in research (Decker 2013) as, Earth Society answering the question ‘Is the Earth
by comparison, the ICD equivalents were regarded flat or round?’. All would predict very high levels of
as inadequate. agreement in a survey of this extreme population,
probably a perfect correlation agreement of r  = 1.0
Reliability in favour of flatness, but this remarkable degree
The reason why DSM-III was a much better of reliability would not suddenly make the world
classification than any of its predecessors is less round.

282 Advances in psychiatric treatment (2014), vol. 20, 280–285 doi: 10.1192/apt.bp.113.011296
https://doi.org/10.1192/apt.bp.113.011296 Published online by Cambridge University Press
 Comparison of DSM and ICD classifications of mental disorder

A similar example in psychiatry is the diagnosis

of attention-deficit hyperactivity disorder (ADHD). Box 3 Main differences between ICD and DSM
In DSM-5, ADHD is characterised by a pattern ICD DSM
of behaviour, present in multiple settings (e.g. at • Official world classification • US classification (but used in many other
school, at home) that can result in ‘performance countries)
• Intended for use by all health practitioners
issues’ in social, educational or work settings.
Special attention given to primary care • Used primarily by psychiatrists
Symptoms are divided into two categories of •

inattention and hyperactivity/impulsivity – and low- and middle-income countries • Focused mainly on secondary psychiatric
Major focus on clinical utility (planned care in high-income countries
including failure to pay close attention to details, •

difficulty organising tasks and activities, excessive for ICD-11) with reduction of number of • Tends to increase the number of diagnoses
talking, fidgeting or an inability to remain seated diagnoses with each succeeding revision
in appropriate situations – with a numerical • Provides diagnostic descriptions and • Diagnostic system depends on operational
score of symptoms given to set the threshold for guidance but does not employ operational criteria using a polythetic system for most
diagnosis for the inattention, hyperactivity and criteria conditions (i.e. combination of criteria that
need not all be the same)
impulsivity criteria.
It is perfectly possible to train a set of raters
to achieve high degrees of reliability in making
the diagnosis of ADHD, but this heterogeneous
(Box 3). This is also going to affect practice in the
group of symptoms exhibited by a large number
USA, as although it may not be fully appreciated
of children at some phase of development does not
by practitioners in the UK, every hospital diag­
constitute a valid diagnosis. It also seems to share
nostic record has to use the ICD system, not the
some of the uncertainty of roulette.
DSM one.
In discussing these differences, I must
Differences between DSM and ICD
acknowledge a conflict of interest in that I am
It is important to realise there is a lot of conver­ supporting the development of ICD-11 in various
gence between the two international systems of ways and have the personal view that it should
diagnosis, and that it is also possible to convert the replace the DSM system completely in the course
diag­noses of one system into another. Although of time – or, perhaps put more gently, that the
the main groups of psychiatric disorder are DSM should ultimately merge with ICD. The main
diagnosed similarly (Box 2), there are important argument used by those who favour DSM is that it
differences with individual diagnoses in terms of creates more accurate diagnosis. This may be true,
both definition and name, and these differences are partly because operational criteria are used but
likely to increase with the introduction of ICD-11 also because a much greater amount of resource
and effort goes into making the classification
than with ICD, which received very little funding
Box 2 Main subgroups of psychiatric indeed. As a consequence, ICD tends to follow
disorder in both DSM and ICD palely in the footsteps of its big American brother,
not wanting to be left behind but not quite sure
• Organic, including symptomatic, mental disorders what it is keeping up with.
• Mental and behavioural disorders due to use of
psychoactive substances Validity
• Schizophrenia, schizotypal and delusional disorders As discussed earlier, reliability is no measure of
• Mood [affective] disorders validity, and although operational criteria may
• Neurotic [a term now dropped], stress-related and improve reliability, they can do so at the expense of
somatoform disorders validity. Thus, for example, a patient who satisfies
• Behavioural syndromes associated with physiological all the criteria for major depression using the DSM
disturbances and physical factors system may have developed all the symptoms after
• Disorders of personality and behaviour in adults a major event, not necessarily a traumatic one. The
• Mental retardation
clinician may rightly conclude that the symptoms
are a reaction to the event and not the onset of a
• Disorders of psychological development
depressive disorder requiring specific psychological
• Behavioural and emotional disorders with onset usually or drug treatment. The clinician would therefore
occurring in childhood and adolescence
be thinking of this as an adjustment disorder, even
• Unspecified mental disorders though the operational criteria clearly marked
this out as a major depressive disorder. This is

Advances in psychiatric treatment (2014), vol. 20, 280–285 doi: 10.1192/apt.bp.113.011296 283
https://doi.org/10.1192/apt.bp.113.011296 Published online by Cambridge University Press
 Tyrer

not a textbook issue; in my clinical practice I of Research Domain Criteria (RDoC) that is
MCQ answers
have seen patients misdiagnosed and over­t reated linked more closely to neurobiological correlates
1b 2d 3e 4b 5c
with antidepressants when it was quite clear that (Cuthbert 2013).
a wait-and-see approach would have been much The R DoC f ra mework ma ke s a cle a r
more appropriate. conceptualisation of mental illnesses as brain dis­
orders, but in contrast to neurological disorders
Creation of more diagnoses/pathology where there are identifiable lesions, mental
The other problems associated with the pre­ disorders are regarded as disorders of brain
occupation with reliability are a tendency to circuits. These circuit disorders can be tested by
create more diagnoses rather than fewer, and also electro­physiology, functional neuroimaging and a
to create pathology out of normal variation. Allen range of new methods for quantifying connections
Frances (2013), the former chair of the DSM-IV in vivo. This process is complemented by data
Task Force, has written persuasively about the from genetics and clinical neuroscience to yield
dangers of this, and if clinical judgement is left ‘biosignatures’ that add to clinical symptoms and
aside in discussions over diagnosis, this tendency signs and so improve clinical management.
will only increase. The detection of pathology This may be regarded as fanciful by some,
needs a scholarly overview and in the DSM but when one considers that we had no real
compartmentalised system this is often lacking. understanding of mental functioning in vivo 20
It is a sad fact that most of the diagnoses in both years ago, a great deal of progress has been made
classification systems are seldom used, yet with since then. But it has not yet yielded results of
each succeeding revision of DSM the number of note, and there are stern critics of this approach
diagnoses tends to increase (although there is a that cannot be dismissed lightly (Bracken 2012;
Kleinman 2012).
reduction of three diagnoses between DSM-IV and
DSM-5), and each volume gets fatter.
Conclusions
Who uses the two systems? Both DSM and ICD classification systems in
psychiatry have value and, in particular, the
DSM is mainly used by psychiatrists, although it
high profile of DSM since 1980 has stimulated
is recognising that this is not entirely satisfactory
much more interest in nosology and heightened
and is promoting its use among psychologists
awareness of both the limitations and advantages
and other mental health practitioners. ICD, on
of current classification.
the other hand, has always had in mind ‘the
DSM has been productive in promoting research,
universal mental health practitioner’, whether it
but has handicapped advances in some respects by
be a barefoot doctor in Tanzania or a professor
giving credibility to diagnoses which probably do
of psychiatry in Baltimore. It therefore has to be
not exist (Markon 2013), and has generated much
flexible and simple in the use of language to enable
needless research into issues such as comorbidity
all practitioners – including many with very little
of disorders which share much more than they
formal qualifications in low- and middle-income
differ by.
countries – to be acceptable. One of the tasks of
ICD has been poorly resourced and has not been
ICD is to show to practitioners that it can satisfy
able to generate the same degree of research data
the critiques of diagnosis as spurious labelling and
as DSM, but has steadily improved over the years
can serve practitioners and researchers well, but
and, with better descriptions and definitions, is
it has a long way to go in achieving these desired
likely to be used not only widely, but more seriously
aims. This explains the key focus of ICD-11 –
and accurately.
clinical utility – as a diagnosis is only valuable if
Diagnostic practice remains fluid and it is
it can be used universally.
uncertain which of these three approaches – DSM,
ICD or RDoC – will dominate in the end. None of
Research Domain Criteria these classification systems are going to be able
The DSM system will continue to be used because to overcome the fundamental problems outlined
so much has been invested in it by the American at the beginning of this article until we have
Psychiatric Association and it is a well-resourced independent measures of disease, but for some
and diligent classification, and this remains true mental conditions this will never happen. Without
in spite of the abundant criticism that it has a well-functioning classification system we would
attracted. But it continues to be restrictive and be blind, deaf and stupid in the practice of our
has overreached itself, and the new US-initiated craft and so need to spring to its defence whenever
ambitious plan is to replace it by a new system it is mindlessly assaulted.

284 Advances in psychiatric treatment (2014), vol. 20, 280–285 doi: 10.1192/apt.bp.113.011296
https://doi.org/10.1192/apt.bp.113.011296 Published online by Cambridge University Press
 Comparison of DSM and ICD classifications of mental disorder

References Feighner JP, Robins E, Guze SB, et al (1972) Diagnostic criteria for use in
psychiatric research. Archives of General Psychiatry, 26: 57–63.
American Psychiatric Association (1980) Diagnostic and Statistical
Manual of Mental Disorders (3rd edn) (DSM-III). APA. Frances A (2013) Saving Normal: An Insider’s Revolt Against Out-of-
Control Psychiatric Diagnosis, DSM-5, Big Pharma, and the Medicalization
American Psychiatric Association (1987) Diagnostic and Statistical
of Ordinary Life. William Morrow.
Manual of Mental Disorders (3rd edn revised) (DSM-III-R). APA.
American Psychiatric Association (1994) Diagnostic and Statistical Guze SB, Goodwin DW (1971) Diagnostic consistency in antisocial
Manual of Mental Disorders (4th edn) (DSM-IV). APA. personality. American Journal of Psychiatry, 128: 360–1.

American Psychiatric Association (2000) Diagnostic and Statistical Guze SB (1975) The validity and significance of the clinical diagnosis
Manual of Mental Disorders (4th edn, text revision) (DSM-IV-TR). APA. of hysteria (Briquet’s syndrome). American Journal of Psychiatry, 132:
138–41.
American Psychiatric Association (2013) Diagnostic and Statistical
Manual of Mental Disorders (5th edn) (DSM-5). APA. Kendell RE (1968) The Classification of Depressive Illnesses (Institute
Ayuso-Mateos JL, Nuevo R, Verdes E, et al (2010) From depressive of Psychiatry, Maudsley Monograph, No. 18). Oxford University Press.
symptoms to depressive disorders: the relevance of thresholds. British Kleinman A (2012) Rebalancing academic psychiatry: why it needs to
Journal of Psychiatry, 196: 365–71. happen – and soon. British Journal of Psychiatry, 201: 421–2.
Bracken P, Thomas P, Timimi S, et al (2012) Psychiatry beyond the current Markon KE (2013) Epistemological pluralism and scientific development:
paradigm. British Journal of Psychiatry, 201: 430–4. an argument against authoritative nosologies. Journal of Personality
Craddock N, Mynors-Wallis L (2014) Psychiatric diagnosis: impersonal, Disorders, 27: 554–79.
imperfect and important. British Journal of Psychiatry, 204: 93–5.
Spitzer RL, Endicott J, Robins E (1975) Clinical criteria for psychiatric
Cuthbert BN, Insel TR (2013) Toward the future of psychiatric diagnosis: diagnosis and DSM-III. American Journal of Psychiatry, 132: 1187–92.
the seven pillars of RDoC. BMC Medicine, 11: 126.
Timimi S (2011) Campaign to Abolish Psychiatric Diagnostic Systems such
Dale RC, Heyman I, Giovannoni G, et al (2005) Incidence of anti-brain as ICD and DSM (CAPSID). Critical Psychiatry Network UK.
antibodies in children with obsessive-compulsive disorder. British Journal
of Psychiatry, 187: 314–9. Tyrer P (2013) Models for Mental Disorder (5th edn). Wiley-Blackwell.
Decker H (2013) The Making of DSM-III: A Diagnostic Manual’s Conquest World Health Organization (1992) ICD-10: Classification of Mental and
of American Psychiatry. Oxford University Press. Behavioural Disorders. WHO.

MCQs d usually lacking biomarkers 5 The main differences between ICD and
Select the single best option for each question stem e having many more diagnoses. DSM diagnoses in psychiatry are:
a ICD is more comprehensive than DSM
1 The DSM classification: 3 ICD-11: b DSM is more accurate than ICD
a is preferred by clinicians across the world a is likely to be very similar to DSM-5 c ICD is the official international classification in
b shows better agreement between assessors b will have more diagnoses than ICD-10 psychiatry
than ICD c will not be used in North America d DSM is the only classification used in the USA
c is the most valid classification available d will be the last revision of the ICD e ICD is used by more psychiatrists than DSM.
d is revised every 5 years e is focused on clinical utility.
e is the standard classification for international
data. 4 Compared with DSM diagnoses, ICD
diagnoses:
2 DSM and ICD classifications differ from a are interchangeable
other disease classifications in: b allow more clinical judgement in making
a being more widely used diagnoses
b having clearer definitions c take longer to complete
c being more appropriate for low- and middle- d are more stable over time
income countries e are preferred by researchers.

Advances in psychiatric treatment (2014), vol. 20, 280–285 doi: 10.1192/apt.bp.113.011296 285
https://doi.org/10.1192/apt.bp.113.011296 Published online by Cambridge University Press