Article Text

Download PDFPDF
Disease, diagnosis or syndrome?
  1. J M S Pearce
  1. Correspondence to Dr J M S Pearce, Emeritus Consultant Neurologist, Department of Neurology, Hull Royal Infirmary, 304 Beverley Road Anlaby, East Yorks HU10 7BG, UK; jms.pearce{at}


The advance of medical semantics is, in general, towards causation. As knowledge increases, the common consequence is the re-definition of disease. This starts with symptoms then a disorder of structure or function, abnormalities of images, genetics or biochemistry, the ultimate aim being a specific aetiological mechanism which replaces broader descriptions. But medical terminology of diseases, diagnoses and syndromes is inherently imprecise. Careless nomenclature causes confused dialogue and communication. Symptoms of uncertain cause are commonly lumped together and given a new ‘diagnostic’ label which also may confuse and produce false concepts that stultify further thought and research. Such medicalisation of non-specific aggregations of symptoms should be avoided. The defining characteristics of diseases and diagnoses should be validated and agreed. The pragmatic diagnoses of ‘symptom of unknown cause’ or ‘non-disease’ are preferable to falsely labelling patients with obscure or non-existent diseases.

“I tried to unveil the stillness of existence through a counteracting murmur of words, and, above all, I confused things with their names: that is belief.”

Jean-Paul Sartre (The Words, 1964).

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Diseases and diagnoses imply the converse of health. But defining disease as the opposite of health is fraught with difficulties as definitions of exactly what health is are tricky. For example, the WHO's claim in 1946 that health is “a state of complete physical, mental and social well-being, not merely the absence of disease or infirmity” has been deemed “more realistic for a bovine than a human state of existence”.1

With increasing technology the practice of medicine has changed radically from a patient based, thorough history and physical examination, often to a mechanistic, technological service that may bypass the human contact, which traditional clinical methods necessarily entail. Furthermore, the naming of an illness with attention to verbal accuracy in order to encapsulate its nature has fallen into abeyance at the divergent altars of sophisticated technical description and loosely conceived labels. Both communication and precise nomenclature are hampered.

The non-medical fraternities have been quick to latch onto these new labels. They are vociferous in their unqualified opinions made all the more plausible by the broadly fashioned but inaccurate names they so readily embrace: ‘ME’, for example, displays no good evidence of encephalomyelitis. The separation of illness from normal physiological symptoms—such as fatigue, many transient headaches and dizziness—is commonly overlooked. Symptoms and syndromes of obscure causation, such as benign nocturnal myoclonus, are mistaken for valid diseases. And certain subjective complaints, such as some instances of backache or whiplash, are medicalised and thereby justified as social and occupational disabilities. Imprecise terminology is of major concern because it implies an understanding which stultifies scientific research and is neither conducive to sensible diagnostic processes nor constructive management. Richard Smith, when he was the editor of the British Medical Journal, drew attention “to the increasing tendency to classify people's problems as diseases”.2

The advance of medical semantics is in general, towards causation


The interface of health, disease and illness has puzzled many disciplines: medicine, anthropology, health sociology and, significantly, philosophy and theology. However, it is necessary to use names for sensible communication. The advance of medical semantics is in general, towards causation (figure 1). A common consequence is that re-definition of diseases as disorders of structure or function displaces syndromic terms; the ultimate aim is a specific aetiological mechanism which then replaces broader descriptions. For example, Willis described the “scarcity of the spirits and the spurious palsy” in cases subsequently recognisable as myasthenia gravis, later redefined by electrophysiological and immunological investigation.

Because human diseases affect people of varied attitudes, expectations and cultures, concepts of health depend on context. The basic idea of disease hinges on deviation from health.3,,5 Whether people believe themselves to be ill varies with class, gender and ethnic group. Diverse social, reward and economic factors also change the idea of illness.6

Concepts of disease are inevitably heterogeneous. What is regarded as a disease also changes at different periods of history because of increasing knowledge, changing expectations and improvements in diagnostic technology.7

Disease emerges as a subjective concept, often involving a series of value judgements. Physicians rightly tend to insist on objective manifestations or deviations of measurable values from the norm, the so-called normativism definition. The layman by contrast often believes that disease (‘He's got flu’) or health are empirical facts and nothing more: non-normativism.6 Diseases are frequently mistaken for diagnoses. Both terms demand accurate definition.

Definitions of disease

Originally, from 14th century Middle English, ‘disease’ was a condition of the body, or of some part or organ of the body, in which its functions were disturbed or deranged; … “a departure from the state of health, especially when caused by structural change”. Its present meaning appeared in the 15th century as “a species of disorder or ailment, exhibiting special symptoms or affecting a special organ” (Oxford English Dictionary).

Some highlight a nominative, confining and restrictive approach that excludes all conditions in which a specific cause is not known: “A disease is any disturbance or anomaly in the normal functioning of the body that probably has a specific cause and identifiable symptoms.”8

Mikhail Tombak (Centre for Health Sciences, Moscow) seeks to define disease by its perceived function: “What we call a disease is the defensive reaction of our body's mechanisms designed to maintain us healthy.”9

J G Scadding (figure 2) with scholarly perception gave the following biological, ‘general’ definition: “A disease is the sum of the abnormal phenomena displayed by a group of living organisms in association with a specified common characteristic or set of characteristics by which they differ from the norm for their species in such a way as to place them at a biological disadvantage.”10

Figure 2

J G Scadding (1907–2000) was Professor of Medicine at the University of London and Consultant Physician at the Brompton and Hammersmith Hospitals. He made major contributions in the fields of rational nomenclature and definitions of disease, clinical entities and syndromes.

But Scadding's definition exposed the logical heterogeneity of ‘diseases’,11 which meant abandoning in current nosology any possibility of a ‘unified concept of disease’.12 As a result, he revised it: “In medical discourse, the name of a disease refers to the sum of the abnormal phenomena displayed by a group of living organisms in association with a specified common characteristic or set of characteristics by which they differ from the norm of their species in such a way as to place them at a biological disadvantage.”13

There are therefore no universally accepted criteria for establishing what is or is not a disease. A particular disease can be defined by the common feature which characterises the group on which its description is based. Definitions may therefore be: descriptive or syndromal (eg, sarcoidosis), morbid–anatomical (eg, cancer), pathophysiological (eg, asthma) and aetiological (eg, tuberculosis). As knowledge advances, syndrome tends to be displaced when definition by more objective features (eg, aetiology) takes precedence.8 The practice of making diagnoses is separate to defining diseases or syndromes but is a utilitarian way in which physicians work.

Technological definition?

Boorse stressed the use of descriptive terms in defining health. He sought to exclude subjective judgements and related health and disease to statistical norms.14 In an extension of this notion, Hoffman argues that technology is “constitutive of the concept of disease”. “Technology”, he says, “provides the basic phenomena that define disease and generate and form medical knowledge and action.7 Technology can constitute the defining signs, markers and end points as well as medical taxonomy. It establishes how we act towards disease: thorough technology, diagnosis and treatment are established, the actions that constitute the concept of disease. However, there are many conditions where there are no technological tests (eg, migraine and much of epilepsy). Hence, technology often cannot be constitutive of their definition and diagnosis.” Whiplash syndromes and fibromyalgia have no corresponding technological tests. The advocates of the technological concept see these examples “as borderline cases and classified as syndromes. Non-technological disease entities [sic] are low-status diseases7 15 precisely because they are not technologically testable and treatable”. The essential though subjective human aspects of disease are notably disregarded in this simplified mechanistic view.


Diagnosis (Greek διαγιγνωσκειν to distinguish or discern) is the ascription of a name to an illness, although the word illness itself evades precise definition. Diagnosis implies the distinction of illness or disease from health. Most obviously it points to the correct treatment and, equally importantly, it leads to the prognosis. It has three levels16: first, a category or class of disorder (eg, neuromuscular disorder) to provide a broad frame of reference; second, it particularises the subject (eg, 30-year-old female) to be diagnosed; and finally, it provides a more specific reasoned categorisation (autoimmune myasthenia gravis).

An old but excellent account stated: “Diagnosis consists in bestowing a name upon a certain assemblage of pathological phenomena … which includes a knowledge of the causal factors of the disease; a determination of its character with reference to type and severity; an estimate of the amount and kind of damage, both general and local, which has been sustained by the organism; a forecast of the probable course and duration of the morbidity process; and a cognizance of the personal characteristics of the patient, whether psychic or physical, inherited or acquired (table).”17


Example of types of diagnosis (after Butler17)

There are therefore no universally accepted criteria for establishing what is or is not a disease

Confusion arises when diagnosis is wrongly equated with disease

Diagnosis may be applied to: (a) a described and recognisable combination of symptoms and signs such as vomiting and diarrhoea; (b) phenomena associated with a specified pathology or disorder of function, such as inflammation with altered gastrointestinal motility; and (c) a specific cause or causes, such as bacillary dysentery or cholera (figure 1).

Sir Thomas Clifford Allbutt (figure 3) clarified the issues over 100 years ago: “Clinical diagnosis, however, is not investigation—a distinction some practitioners forget; diagnosis depends not upon all facts, but upon crucial facts. Indeed we may go farther and say that accumulation of facts is not science; science is our conception of the facts: the act of judgment, perhaps of imagination, by which we connect the unknown with the known.”18

Figure 3

Allbbutt's notes on the composition of scientific papers (1904).

Confusion arises when different people use different defining criteria, and when diagnosis is wrongly equated with disease. This leads to problems in discussions, especially of disorders of uncertain or multifactorial cause—for example, asthma, migraine, fibromyalgia, post-traumatic stress disorder (PTSD) and attention deficit disorder.

The fallacies of equating a diagnostic label with a disease are nowhere more problematical than in psychiatry19 where objective phenomena of necessity are sparse and often non-specific. Schizophrenia is a psychosis, a disease of the mind, but recent work has unequivocally shown physical abnormalities in brain imaging and blood flow which indicate that although the aetiology(ies) is unknown, an organic syndrome is present. Similar arguments may apply to endogenous bipolar depression. Many pain syndromes of uncertain cause are complicated by psychological factors which may overshadow the primary organicity of the disorder; migraine and postherpetic neuralgia are examples.

The fashionable but arbitrary criteria which define PTSD is one example of an old disorder disguised as a ‘new syndrome’. Originally it was familiar as ‘shell shock’ of first world war veterans who had experienced or witnessed acute, massive psychological stress of life threatening severity: a comrade blown to pieces stepping on a mine, seeing a friend or relative, or oneself consumed by flames, and in civilian life someone crushed by a falling building or killed by a car. Plainly, such experiences could be profoundly distressing to anyone. The essential criteria of the Diagnostic and statistical manual of mental disorders, fourth edition, text revision (DSM-IV-TR, 2000) are: “A syndrome with onset following a traumatic, usually life-threatening event. The course may resemble panic disorder with the herald attack a real-life threat rather than spontaneous panic. Recurrent images of the original event frequently occur.”20

But PTSD has now been widened to include less severe experiences so that victims of a minor car accident with a back sprain, or minor whiplash or head injury, or uninjured relatives or even witnesses of accidents commonly affect PTSD, often in the context of pending litigation. As the DSM definition suggests, PTSD is no more than an anxiety state dominated by panic attacks, phobias and apprehensions focused on the terror of the original experience. In psychiatric practice it was a valid (albeit imprecisely delineated) clinical syndrome. But, a disservice is done to genuine psychologically ill patients by broadening the elements of symptomatology, relating it to trivial or non-existent trauma, and by imparting a portentous but sleeveless label.

Differential diagnosis is apt to be confused unless the defining characteristics are validated and agreed. We should clearly delineate:

  • A simple collection of symptoms, with precise cause and pathology unknown (eg, essential tremor).

  • Diagnoses determined by gross, microscopic or ultrastructural pathology (eg, acute glomerulonephritis).

  • Diseases named by measurable aberrations of biochemical, immunological or hormonal elements (eg, caeruloplasmin depletion, hypothyroidism).

  • Others marked by aetiological agents: bacterial, viral or by specific disorders of molecular genetics (eg, myotonic dystrophy, Down's syndrome).21

This template ranging from symptoms to exact aetiology is akin to Hughlings Jackson's Spencerian notion of hierarchical levels of brain functions. These levels should be clearly identifiable when we ascribe a category in formulating diagnoses. Those commonly employed resemble the criteria for diagnosis with which disease may be mistaken.

Syndromes are a fertile field for hastily considered ideas and spurious labels

As medical knowledge advances, each of these features tends to displace its antecedent definition as it gets closer to the desirable end point—a specific identifiable or measurable cause.

So, where does a syndrome fit in?

What is a syndrome?

Diagnosis of a syndrome (such as Tourette's syndrome or sarcoidosis) depends on identification of possible combinations of signs and symptoms displayed by a patient. If clearly delineated, syndromes are valuable labels for communication. The longer a syndrome's cause is unknown, the more disagreements may emerge about which features are necessary to sanction a diagnosis. As a result, the constituents of a syndrome can vary over time and in different places. They are the objects of a diagnosis, but since the aetiology is unknown, they do not constitute a disease. Syndromes are a fertile field for hastily considered ideas and spurious labels.

Syndromes have however attracted the colloquial essentialists who with medical sanction promulgate an often imprecise syndromic label to an ill defined set of symptoms. And here lies the danger. When such a label is recognised as a diagnosis and widely used in both medical and lay articles, it acquires both social acceptability22 and often a false notion that the subject discussed, for example PTSD or fibromyalgia, is an actual disease. New clinical diagnoses are often welcomed primarily as opportunities for market growth of drugs and ‘therapists’.23 It is now possible to define most organic diseases with objective criteria, without subjective value judgements. The greatest scientific virtue is the willingness to recognise a theory's refutation. Only if this tenet is observed will advances in knowledge diminish or abolish specious labels posing as definitive diagnoses or diseases.

Essentialist or nominalist?

Issues of nomenclature are, however, more complex. People apply differing philosophical conceits to nomenclature. In the nominalist approach the nature of an illness is decided by reference to a biomedical norm and the names are convenient devices by which the diagnostic process can be stated briefly. Thus meningococcal meningitis combines two diagnostic categories: meningococcal defines aetiology and cause, while meningitis defines the morbid anatomy. Essentialism is based on the Aristotelian notion that natural organisms show an invariant general pattern or essence shared by all members of the group.24 ‘He is sick with a cold’ recognises the general state of being ill and a vague cause which is generally understood in lay terms. Nominalist definitions do not attempt the impossible task of revealing the essence of the condition defined but indicate how words relate to observable phenomena. The philosopher Karl Popper25 pointed out that essentialist definitions, depending on intuitive acceptance for their validity, have no place in science. Thus terminology of disease in medical discourse in effect necessitates nominalist definition.24

Medicalisation of nomenclature

Of many symptom complexes masquerading behind the designation of syndromes are late whiplash, fibromyalgia, chronic temporomandibular syndrome, repetitive strain injury, multiple chemical sensitivities, sick building and Gulf war syndromes. The instinctive quest for a physical cause is evident in their names. One example illustrates the altering concepts and dangers of certain syndromic ‘diagnoses’. For Victorian doctors, fatigue was both a physical and mental affliction. Neurasthenia was a descriptive term coined by George Miller Beard26 for the colloquially known nervous exhaustion; the word neurasthenia has the merits of simplicity and its meaning is self-evident. A celebrated example was the philosopher and psychologist Herbert Spencer. “The retreat into illness was also for Spencer a retreat from social intercourse.”27 The idea of ‘retreat into illness’, beloved of the distinguished psychiatrist, the late Sir Martin Roth, bears its own inferences.

The concept of neurasthenia declined between the 1930s and 1960s but re-emerged in the 1980s as chronic fatigue syndrome.28 This had “adopted the organic inheritance of Beard's ideas of neurasthenia, despite the fact that the question of organicity could not be decisively answered in a single case”.29 Doctors sometimes mistakenly reject fatigue as a symptom of neurosis, for fear of implying it is not genuine. This arises mainly because there is no robust verifiable definition. Fatigue can easily be exaggerated by patients but also by caring professionals whose zealous therapeutic attentions can iatrogenically induce or worsen the symptoms.30 Epidemics of ill founded ‘medicalised’ nomenclature aggravate the confusion and prove irresistible to social scientists.

It appears that understanding an acceptable medical taxonomy is sometimes hindered by inconsistency of definition, methodology and the characteristics of patients.31 But the common enquiry: “is it a disease or is it a syndrome or a diagnostic entity?” in reality is a request for a statement of an acknowledged authority, and is ‘foolishly simplistic’.14 Following Sydenham, doctors define a patient's condition as a ‘disease’ and are then licensed to take various actions. “Each civilisation” wrote Ivan Illich “defines its own diseases. What is sickness in one might be chromosomal abnormality, crime, holiness, or sin in another”.32 Smith commented: “to have your condition labelled as a disease may bring considerable benefit both material (financial) and emotional. However, the diagnosis of a disease may also create problems; you may be denied insurance, a mortgage, and employment”.2


Medical terminology is imprecise. Diseases are defined by differing types of criteria, continually evolving but necessary for any logical discussion and sensible management. But the medicalisation of non-specific aggregations of symptoms describing common human complaints rather than diseases or validated syndromes is to be avoided. In practice, decisions sometimes have to be made about investigations and management of many individuals, which are necessarily based on such value judgements that necessarily include subjective elements on the part of both patient and physician. Woolly thinking and careless nomenclature however, are undesirable ingredients.

A greater danger is in simply lumping symptoms together and giving them a new label. A wholly false concept ensues: that the diagnostic label, often glorified by the term syndrome, represents a new disease. This stultifies further thought and may inhibit research and investigation of the aetiology because doctors commonly regard the syndrome (diagnosis) as a disease.21 They mistakenly accept it as an end point and as an entity. Indeed Smith's survey revealed, “we were not suggesting that the suffering of people with these ‘non-diseases’ is not genuine, … [but] some of these non-diseases already appear in official classifications of disease”.

Better, argued Clifton Meador, to describe a patient in whom a diagnosis could not be made as having a ‘non-disease’ than make “the common error of continuing to label such patients with non-existent diseases”.33

There is much to commend the diagnoses of symptom x, of unknown cause (headaches, backache or diarrhoea) which at least point the way to further observation and scientific exploration.



  • Competing interests None.

  • Provenance and peer review Not commissioned, not externally peer reviewed.

Other content recommended for you