The term ‘diagnosis’ can mean both the name of a particular disease as well as the process of determining or ‘making’ a diagnosis. In medicine, generally, various terms are used to describe a pathological entity. When there is the presence of objective pathology or the presumed understanding of aetiology, the term ‘disease’ is generally used, e.g., pancreatic cancer, strep throat, Alzheimer’s disease. In instances of unknown aetiology or when the disease process is not apparent, the term ‘disorder’ is usually applied with a syndromic characterization, i.e., definition based on symptom presentation, history, and sometimes, associated laboratory findings. Other terms are also used in common parlance, such as ‘illness’ for an individual’s subjective awareness of distress and ‘sickness’ for the inability to perform usual social roles. For the most part, in psychiatry, the term ‘disorder’ is used.

Classification represents the process of placing diagnostic entities into various groupings in a systematic way, based on a set of principles with regard to the similarities and differences among these categories. Depending upon the principles and conceptual framework underlying the categorization process, classifications can be very different.

In some ways, the most important question may be ‘whose needs is the classification primarily intended to address?’ Clinicians want a classification that can categorize as many people that come in for help as possible. They want the classification to facilitate the identification and treatment of patients and provide guidance on prognosis and cause. Researchers want to have groupings that are highly homogeneous in order to test the efficacy of specific treatments and to better understand the aetiology of specific disorders. Educators want a classification system to offer a structure for teaching about psychopathology and differential diagnosis. Public health administrators want to track epidemiology, health utilization, and costs over time. Some argue that psychiatric diagnosis is a reductionistic labelling of individual differences or social deviance and exposes individuals to potential stigma. At a minimum, they would like a psychiatric diagnostic system to be less prone to misuse. Ultimately, most classifications attempt to balance among those competing priorities, not always successfully. In some cases, e.g., the ICD-10, different products are developed for different target groups, i.e., research diagnostic criteria for investigators, a simpler, more aggregated classification for primary care providers, etc.

A range of conceptual issues and their resolution determines the principles and rules governing a system of classification. It is important to note, however, that classification systems do not necessarily apply these rules in a consistent manner. Some of the issues noted below may not be resolved in an absolute manner, but in a way that employs compromises among multiple priorities, e.g. balancing the needs of clinicians, researchers, educators, and public health administrators or having some diagnostic groupings based on a descriptive approach and others on a theory-based approach.

Descriptive v. theory-based: Do the classification principles emanate from a theory regarding the aetiology or mechanisms of psychopathology (e.g. psychodynamics, behavioural, neurobiological) or does the classification attempt to provide a theoretical heuristic framework for describing syndromic entities?

The first international classification of diseases in 1855 was concerned with a nomenclature of causes of death.⁽¹⁾ After many revisions this list was adopted by the World Health Organization (WHO) in 1948 and the so-called Sixth Revision of the International Statistical Classification of Diseases, Injuries and Causes of Death (ICD-6) was produced.⁽²⁾ The sixth edition of the ICD included for the first time a classification of mental disorders, containing 10 categories of psychoses, nine categories of psychoneurosis, and seven categories of character, behaviour, and intelligence. A number of problems with this classification (e.g., many important categories such as the dementias, many personality disorders, and adjustment disorders were not included) rendered it unsatisfactory for use in most countries; only five countries, including the United Kingdom, adopted it officially. The first edition of the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders (DSM-I)⁽³⁾ was published in the United States as an alternative to ICD-6. For the first time in an official classification, glossary definitions of the various disorders were included in addition to the names of the disorders.

Work on ICD-8 began in 1959 with the goal of developing a classification system that would be acceptable to all of its member nations. The resulting system, ICD-8, went into effect in 1968, and in 1974 added a glossary which was largely based on British views about diagnostic concepts.⁽⁴⁾ Coincident with the development of ICD-8, the American Psychiatric Association prepared a second edition of its DSM based on ICD-8, defining each disorder for use in the United States.⁽⁵⁾

The early 1970s saw the introduction of explicit operationalized diagnostic criteria that were developed for research purposes. Although the glossary definitions of disorders in DSM-II and ICD-8 were an improvement over just having a list of diagnostic categories, these brief descriptions were too vague to be useful in identifying diagnostically homogeneous populations for study. Researchers responded to this need by developing their own operationalized criteria. The first set of diagnostic criteria that covered a wide range of disorders was developed by Robins and Guze at Washington University in St. Louis⁽⁶⁾ with the stated purpose of ‘provid[ing] common ground for different research groups so that diagnostic definitions can be emended constructively as further studies are completed’. (p. 57). They were known as the ‘Feighner criteria’ after the first author of the paper that presented them. Criteria sets for 16 disorders were presented and listed those features required for each diagnosis (known as ‘inclusion criteria’) as well as features whose presence would rule out the disorder (known as ‘exclusion criteria’). The Feighner criteria proved to be enormously useful to the research community as illustrated by the large number of times they were cited in other papers (i.e. 1650 citations from 1972 to 1982 as compared to the typical average of 2.1 citations per paper). Several years later, an expanded set of research criteria based on the Feighner criteria was developed to meet the needs of a National Institute of Mental Health-sponsored collaborative project on the psychobiology of depression.⁽⁷⁾ These criteria, known as the Research Diagnostic Criteria (RDC) subsequently became very popular among researchers and were heavily used, especially in research on mood and psychotic disorders.

To develop the mental disorders section for ICD-9, WHO initiated an intensive program to identify problems encountered by psychiatrists in different countries in the use of the mental disorders section of ICD-8 and to formulate recommendations for their solution. A series of eight international seminars were held annually from 1965 to 1972, each of which focussed on a recognized problem in psychiatric diagnosis. The outcome of the seminars formed the basis for the recommendations made for ICD-9,⁽⁸⁾ which was ultimately published in 1978.

As work progressed on the development of ICD-9, the American Psychiatric Association decided to develop a third edition of its diagnostic manual, DSM-III.⁽⁹⁾ This decision was made both because of identified inadequacies of the ICD-9 for research and clinical use, and because the ICD-9 did not include important innovations that had already been demonstrated by researchers to be both technically feasible and useful, like operationalized diagnostic criteria. Under the leadership of Robert L. Spitzer, successive drafts of DSM-III were prepared by 14 advisory committees, with the drafts being distributed among both American and international psychiatrists for comments and review. Many of the DSM-III criteria sets were based on the RDC criteria, with the rest developed based on expert consensus.

The improvement in reliability over DSM-II (which provided only glossary definitions) was demonstrated by a large NIMHsupported field trial in which clinicians were asked to independently evaluate patients using drafts of the DSM-III criteria.⁽¹⁰⁾ The explicit diagnostic criteria provided for each of the disorders in the classification were based on the symptomatic presentation of the disorder rather than on theories about the underlying cause. Even though the DSM-III was a product of the American Psychiatric Association, its adoption of this ‘descriptive approach’ resulted in its widespread acceptance by all mental health professionals in the United States, regardless of their theoretical orientation. For example, clinicians from different orientations might have very different understandings of what causes panic attacks; a cognitively-oriented clinician might attribute a panic attack to the person’s tendency to catastrophize in response to normal physical sensations like increased heart rate; a neurobiologically-oriented clinician might consider panic attacks to be due to overactivity of brain circuitry involved in fight-or-flight responses, and a psychodynamicallyoriented clinician might see panic as a consequence of the breakdown of the defense organization at various levels. Despite these divergent hypotheses, however, all of these clinicians can agree on how a panic attack presents (i.e. a discrete period of apprehension or fear with at least four symptoms such as shortness of breath, palpitations, chest pain, choking, dizziness, etc), thus facilitating communication among them.

DSM-III also introduced the use of a multiaxial system for recording the diagnostic evaluation. The multiaxial system facilitated the use of a biopsychosocial model of evaluation by separating (and thereby calling attention to) developmental and personality disorders (Axis II), physical conditions (Axis III), stressors (Axis IV),
and degree of adaptive functioning (Axis V) from the usually more florid presenting diagnoses (Axis I).

Despite initial opposition among some psychiatrists (most especially those with a psychoanalytic orientation), DSM-III proved to be a great success, becoming the common language of mental health clinicians and researchers for communicating about mental disorders. Although it was intended primarily for use in the United States, it was translated into 13 languages and was widely used by the international research community.

Experience with DSM-III in the few years after its publication in 1980, revealed a number of inconsistencies and lack of clarity in the diagnostic criteria sets. Furthermore, research conducted in the early 1980’s demonstrated errors in some of the assumptions that went into the construction of the DSM-III criteria sets. For example, the DSM-III prohibition against giving an additional diagnosis of Panic Disorder to individuals with both Major Depressive Disorder and panic attacks was shown to be incorrect based on data demonstrating that relatives of individuals with both Major Depressive Disorder and Panic Attacks can have either Major Depressive or Panic Disorder.⁽¹¹⁾ For these reasons, work began on a revision of the DSM-III, which was published as DSM-III-R in 1987.⁽¹²⁾

Initial work began on the development of the psychiatric section of the 10th Revision of the International Classification of Diseases (ICD-10), in 1982 under the chairmanship of Norman Sartorius. After a meeting of WHO representatives and consultants together with representatives of the American Drug and Mental Health Administration in Copenhagen in 1982, several further meetings took place (e.g. in Djakarta and in Geneva in 1984) in which a provisional psychiatric classification was designed. It was decided that the ICD-10 classification of mental disorders would be produced in several versions. The first of these is to be used, as are other parts of the International Classification of Diseases, mainly for statistical purposes, and included a short glossary definition for each category.⁽¹³⁾ This is the version that was officially approved by the World Health Assembly and thus, is the version for which international compatibility is mandated by treaty agreements. The second version, Clinical Descriptions and Diagnostic Guidelines is for the use of the practicing clinician.⁽¹⁴⁾ Each category in this version has a detailed definition specifying the main features of the disorder followed by diagnostic guidelines. The third version, the Diagnostic Criteria for Research is primarily intended for research and contains diagnostic criteria which are stricter in form than those in the clinical diagnostic guidelines from which they were derived.⁽¹⁵⁾ For example, while the guidelines may indicate that a particular disorder ‘usually starts in early childhood’, the diagnostic criteria for research would specify that the diagnosis ‘should not be made if the onset is after the age of 30’, The decision to separate the criteria for research from the clinical guidelines was made because clinicians in their daily work do not observe overly strict rules when making diagnoses, which are of cardinal importance for research.(16) Finally, a version of the mental disorders section was produced for use in primary care settings.⁽¹⁷⁾ It contains a much smaller number of categories (i.e. those that are frequently encountered in every day general practice) as well as treatment guidelines corresponding to these categories.

By 1986, a first draft of the psychiatric chapter, including details of the categories, code numbers, diagnostic guidelines, and precise diagnostic criteria for research had been written, and by June 1987, the clinical diagnostic guidelines were being circulated by WHO’s division of mental health for field trials in 194 different centers in 55 different countries.⁽¹⁸⁾ In 1989, the International Revision Conference, attended by representatives of the health ministries of a majority of WHO member states gave formal approval to the basic categories and text. A draft of the diagnostic criteria for research was produced in 1990 and field trials to evaluate inter-rater agreement, confidence in diagnosis, and ease of use began later in the year.⁽¹⁹⁾ Finally, in 1990, the World Health Assembly formally approved its introduction in member states starting in January 1, 1993.

The American Psychiatric Association started work on the development of DSM-IV in 1988, shortly after the publication of DSM-III-R, spurred on by the need to coordinate its development with the already ongoing development of ICD-10. DSM-IV continued the descriptive atheoretical approach advanced by both DSM-III and DSM-III-R, but this time also incorporated a metaanalytic data-based revision process to guide changes.^(20,21) This was in contrast to both DSM-III and DSM-III-R which by necessity, given the paucity of available empirical data, relied almost exclusively on expert consensus. The DSM-IV workgroups began their deliberations by identifying a series of diagnostic questions to be considered and problems to be addressed and employed a threestage empirical review process to address, these questions. The first stage involved a systematic comprehensive review of the published literature guided by literature searches using rules established at a DSM-IV methods conference. The second stage involved supplementing the literature reviews with a data reanalysis project funded by the MacArthur foundation in which existing data sets collected for other studies were combined and analyzed using meta-analytic methods. These data reanalyses were useful in answering a number of diagnostic questions (e.g. determining the minimum number of panic attacks required in order to justify a diagnosis of panic disorder) but were unfortunately limited by incompatibilities in the data sets and the fact that the data needed to answer specific diagnostic questions often had not been collected. Proposed criteria sets formulated based on the literature reviews and data reanalyses were then tested in 15 NIMH-funded multi-site field trials. The entire empirical review process and the reasons underlying the decisions made by the DSM-IV workgroups have been documented in the four volume DSM-IV Sourcebook.^{(22, 23, 24 and 25)}

In order to increase compatibility between ICD-10 and DSM-IV, a collaborative relationship was established between the DSM-IV workgroups and the developers of ICD-10. Two meetings were convened in which the respective workgroups joined forces with the goal of minimizing the differences between diagnostic definitions in the two systems. Unfortunately, the potential to make the two systems identical was seriously constrained by differences in the timelines between the two revision processes. By the time the DSM-IV workgroups were first convened in 1989, the categories and basic text of the ICD-10 had already been settled by the International Revision Conference.⁽²⁶⁾ Thus, although final DSM-IV and ICD-10 systems were much more similar than were DSM-III and ICD-9, a number of mostly small differences in criteria sets persist. While some of the discrepancies are the result of genuine differences in diagnostic outlook (e.g., the one month duration of ICD-10 schizophrenia vs. 6 month duration of DSM-IV schizophrenia), the overwhelming majority appear not to have any justification.⁽²⁷⁾

One of the most important uses of the DSM-IV has been as an educational tool. This is especially true of the descriptive text that accompanies the criteria sets for the DSM-IV disorders. Given that the interval between DSM-IV and DSM-V was being extended from seven years between DSM-III and DSM-III-R, and between DSM-III-R and DSM-IV to at least 12 years, concerns were raised that the information in the text would become increasingly out-ofdate over time. Therefore, in order to bridge the span between DSM-IV and DSM-V, a revision of the DSM-IV text was undertaken.⁽²⁸⁾ The primary goal of the DSM-IV-TR was to maintain the currency of the DSM-IV text, which reflected the empirical literature up to 1992. Thus, most of the major changes in DSM-IVTR were confined to the descriptive text. Changes were made to a handful of criteria sets in order to correct errors identified in DSM-IV. In addition, some of the diagnostic codes were changed to reflect updates to the ICD-9-CM coding system adopted by the U.S. Government.

A fundamental difference between the ICD-10 and the DSM-IV reflects the different purposes of the two systems, i.e., that ICD-10 is set up as a classification system whereas DSM-IV is a diagnostic nomenclature. The primary goal of the ICD is to facilitate the collection of statistics about those individuals who present themselves to a health care professional. Thus, the ICD has been designed to provide the coder with an unambiguous choice of diagnostic category given a particular case. The main rule for deciding whether to include a diagnostic category in the ICD is its common international usage. Inclusion of a category in the ICD carries with it no implication of diagnostic validity—in fact, a number of categories included in ICD-10 were considered for inclusion in DSM-IV— but were not added because of concerns about their validity (e.g., mixed anxiety depression). In contrast, inclusion of a category in the DSM implies that the category has been officially sanctioned by the American Psychiatric Association as appropriate for clinical and research usage, i.e., the category has both clinical utility and is backed up by an empirical data base. It should be noted, however, that the empirical data base is not equivalent for all of the categories—to minimize disruption, diagnostic categories, that were included in earlier editions of the DSM have been ‘grandfathered’ in. Starting with DSM-IV, new categories were only added if they met these higher standards.

Another important difference between the DSM and ICD approach is the role of impairment in the definition of a disorder. With only a few exceptions (e.g., dementia, phobias), mental disorders in ICD-10 are defined exclusively by the symptomatic presentations—there is no requirement that the symptoms cause any impairment in the individual’s level of functioning. Impairment in functioning caused by the symptoms is indicated in ICD-10 by using an orthogonal scale, the International Classification of Functioning.⁽²⁹⁾ In contrast, most of the DSM-IV criteria sets include a criterion (known as the ‘clinical significance criterion’) requiring that the disturbance causes clinically significant distress or impairment in social, occupational, or other important areas of functioning. According to the introduction of the DSM-IV, this criterion has been included to help establish ‘the threshold for the diagnosis of a disorder in those situations in which the symptomatic presentation by itself (particularly in its milder forms) is not inherently pathological and may be encountered in individuals for whom a diagnosis of ‘mental disorder’ would be inappropriate’. (p. 8). Accordingly, the only diagnoses that do not include this criterion are those whose symptomatic presentations are considered to be inherently indicative of psychopathology (e.g., the psychotic disorders).

The diagnostic implications of this difference can be illustrated in the different ways that specific phobia is defined in DSM-IV and ICD-10. In DSM-IV, a phobia is diagnosed only if ‘the avoidance, anxious anticipation, or distress in the feared situation interferes significantly with the person’s normal routine, occupational (or academic) functioning or social activities or relationships, or there is marked distress about having the phobia’. (p. 449, DSM-IV-TR). ICD-10 has no such requirement; the phobia is diagnosed so long as there is a marked fear or avoidance of a specific object or situation. Thus, an individual residing in New York City who has a snake phobia but who never has any occasion to encounter a snake would not be diagnosed as having a mental disorder in DSM-IV-TR because the phobia does not have any impact on the person’s functioning whereas in ICD-10 such an individual would be diagnosed as having a snake phobia because the person would react with fear or avoidance if he or she had the occasion to be confronted with a snake.

Separating symptoms from the functional impairment that results from them certainly makes conceptual sense. In other areas of medicine, the diagnosis of a disorder is based solely on the presence of pathology and not on the effect that the pathology exerts on a person’s life (e.g., a patient is diagnosed with pneumococcal pneumonia if the patient’s lungs are infected with the pneumococcus bacillus regardless of the impact of the pneumonia on the patient’s level of functioning). The problem with diagnosing mental disorders in this way is that it is not currently possible to define the presence of a mental disorder based on the identification of its underlying pathology. The descriptive symptoms that make up the definitions of mental disorders in the DSM-IV and ICD-10 are not specific to mental disorders but can and do occur in individuals without any mental disorder. Thus, defining disorders exclusively in terms of presenting symptomatology, much of which can occur in normal individuals, can lead to false positive diagnoses. For this reason, in order to avoid false positive diagnoses in the absence of objective evidence of disease, DSM relies on functional impairment or distress to help set the diagnostic threshold between normality and disorder.

The DSM-IV-TR classification begins with disorders usually first diagnosed in infancy, childhood, or adolescence. The inclusion of a separate ‘childhood disorders’ section in DSM-IV-TR is only for convenience—some of these conditions are sometimes diagnosed for the first time in adulthood (e.g., attention-deficit/ hyperactivity disorder) and many disorders included in the rest of DSM-IV-TR can start in childhood (e.g., major depressive disorder, schizophrenia). Thus, a psychiatrist doing a diagnostic assessment of a child or adolescent should not only focus on those disorders listed in this section but also consider disorders from throughout the DSM-IV-TR. Similarly, when evaluating an adult, the psychiatrist should also consider the disorders in this section since many of them persist into adulthood (e.g., stuttering, learning disorders, tic disorders).

While the first set of disorders included in this section (i.e., intellectual disability learning and motor skills disorders, and communication disorders) are not, strictly speaking, regarded as mental disorders they are included in the DSM-IV-TR to facilitate differential diagnosis. Autism and other pervasive developmental disorders are characterized by gross qualitative impairment in social relatedness, in language, and in repertoire of interests and activities and include autistic disorder, Asperger’s disorder, Rett’s disorder, and childhood disintegrative disorder. The Disruptive Behaviour Disorders (i.e. Attention-deficit/hyperactivity disorder, conduct disorder, and oppositional-defiant disorder) are grouped together because they are all characterized (at least in their childhood presentations) by disruptive behavior. The Feeding Disorders of Infancy and Early Childhood include the DSM-IV-TR categories of pica, rumination disorder, and feeding disorder of infancy and early childhood (also known as failure to thrive). Tic disorders, elimination disorders, and other disorders of infancy and early childhood (which include separation anxiety disorder, selective mutism, reactive attachment disorder, and stereotypic movement disorder) round out the childhood section.

In DSM-III-R, delirium, dementia, amnestic disorder, and other cognitive disorders, along with substance-induced mental disorders and mental disorder due to a general medical condition, were included in a section called ‘organic mental disorders’, which contained all disorders that were due to either a general medical condition or substance use. In DSM-IV, the term ‘organic’ was completely eliminated from the classification because of the misleading implication that disorders not included in that section (e.g., schizophrenia, bipolar disorder) did not have an organic component⁽³²⁾. In fact, virtually all mental disorders have both psychological and biological components, and to designate some disorders as ‘organic’ and the remaining disorders in the DSM-IV as ‘non-organic’ reflected a reductionistic mind-body dualism.

As a result of the elimination of the Organic Mental Disorder diagnostic grouping, those disorders originally included in that section had to be redistributed throughout DSM-IV into other diagnostic classes. Delirium, dementia, and amnestic disorder were thus grouped together into a major diagnostic class because of their central roles in the differential diagnosis of cognitive impairment. Although both delirium and dementia are characterized by multiple cognitive impairments, delirium is distinguished by the presence of clouding of consciousness which is manifested by an inability to appropriately maintain or shift attention. Three types of delirium are included in DSM-IV based on causative factors: delirium due to a general medical condition, substance-induced delirium, and delirium due to multiple etiologies.

Dementia is defined by clinically significant memory impairment accompanied by impairment in one or more other areas of cognitive functioning (e.g. language, executive functioning). DSM-IV-TR includes several types of dementia based on aetiology, including dementia of the Alzheimer’s type, vascular dementia, a variety of dementia due to general medical and neurological conditions (e.g., HIV, Parkinson’s disease), substance-induced persisting dementia, and dementia due to multiple etiologies. In contrast to dementia, amnestic disorder is characterized by memory impairment occurring in the absence of other cognitive impairments. Two types are included in DSM-IV: amnestic disorder due to a general medical condition and substance-induced persisting amnestic disorder.

In DSM-IV-TR, most of the mental disorders due to a general medical condition have been distributed alongside their ‘non-organic’ counterparts in the classification (e.g. mood disorder due to a general medical condition and substance-induced mood disorder was included in the mood disorders section). Two specific types of mental disorders due to a general medical condition (i.e. catatonic disorder due to a general medical condition and personality change due to a general medical condition) do not fit into any of the other diagnostic classes and therefore, are included here in this diagnostic class.

In DSM-IV, substance-related disorders include psychiatric disturbances that result from medication side effects and the consequences of toxin exposure, in addition to those that arise due to drug and alcohol abuse. Two types of substance-related disorders are included in DSM-IV: substance use disorders (dependence and abuse), which focus on the maladaptive nature of the pattern of substance use; and substance-induced disorders, which cover psychopathological processes caused by the direct effects of substances (including toxins and medications) on the central nervous system.

Included in this grouping are those disorders in which psychosis is the primary characteristic symptom (i.e. schizophrenia, schizophreniform disorder, schizoaffective disorder, delusional disorder, shared psychotic disorder and brief psychotic disorder). It should be noted that other disorders that may have psychotic features are not included in this grouping (e.g. mood disorders with psychotic features, delirium).

This diagnostic class includes disorders in which the predominant disturbance is in the individual’s mood. Although the term ‘mood’ is generally considered to include emotions such as depression, euphoria, anger, and anxiety, DSM-IV includes in this section only disorders characterized by depressed, elevated, or irritable mood. This diagnostic class is further divided into depressive and bipolar disorders. The term ‘bipolar’ is misleading because the name implies the presence of both ‘down’ and ‘up’ moods. In fact, bipolar disorder is defined by the presence of one or more manic or hypomanic episodes. Thus, patients with multiple manic episodes (i.e. unipolar mania) are considered to be bipolar despite the lack of the second ‘pole’.

The common thread tying together disorders in this section is the fact that the clinical presentation of these disorders is typically characterized by significant anxiety. The rationale for this grouping has been criticized because of evidence suggesting that at least some of the disorders are likely to be etiologically distinct from the others. For example, it has been argued that obsessive-compulsive disorder is most likely part of an obsessive-compulsive spectrum that might include tic disorders, hypochondriasis, body dysmorphic disorder, and perhaps trichotillomania.⁽³³⁾

Somatoform disorders are characterized by their presentation in general medical settings by individuals who do not consider themselves to be suffering from a mental disorder. Individuals with somatoform disorders present with somatic complaints or bodily concerns that are not adequately explained by an underlying general medical condition. Conceptually, the somatoform disorders can be divided into three general types: 1) those in which the individual’s focus is on the physical symptoms themselves (somatization disorder, undifferentiated somatoform disorder, pain disorder, and conversion disorder); 2) those who are preoccupied by the belief that one has a serious physical illness despite medical reassurance (hypochondriasis); and 3) those who are preoccupied by the belief that a part or parts of their body are physically defective (body dysmorphic disorder).

Individuals with a factitious disorder intentionally produce or feign a physical or psychological symptom, motivated by the psychological need to assume the sick role and be taken care of. This is in contrast to malingering (which is not considered to be a mental disorder) in which the person is motivated by secondary gain (e.g. to evade criminal responsibility, to receive disability benefits).