You are viewing a free preview of this lesson.
Subscribe to unlock all 11 lessons in this course and every other course on LearningBro.
Clinical psychology begins with a deceptively simple question: how do we decide that a pattern of thought, feeling or behaviour is a mental disorder, and how do we sort that disorder into a recognisable category so it can be studied, communicated and treated? Before any explanation, therapy or research can proceed, clinicians must be able to classify disorders (define them as agreed categories) and diagnose individuals (decide whether a particular person meets the criteria for a category). This lesson examines why classification is necessary at all, the two great classification systems used worldwide — the DSM-5 produced by the American Psychiatric Association and the ICD-11 produced by the World Health Organization — the broad categories of disorder they recognise, and the practical process by which a diagnosis is actually reached. Throughout, we adopt the measured, evidence-based register that clinical psychology demands: disorders are treated as conditions to be understood and helped, never as labels to sensationalise.
Key Definition: Classification is the organisation of mental disorders into agreed categories, each defined by a characteristic cluster of symptoms, so that clinicians and researchers share a common language. Diagnosis is the clinical judgement that a specific individual's presentation meets the criteria for a particular category.
This lesson addresses the Edexcel 9PS0 — Paper 2, Topic 5: Clinical Psychology content on the diagnosis and classification of mental disorders: the purpose of classification, the two major diagnostic systems (DSM-5, APA; ICD-11, WHO), the broad categories of mental disorder they contain, and how a diagnosis is made in practice through clinical interview and criteria-matching. It provides the conceptual and procedural groundwork for the whole topic — the reliability and validity of diagnosis, the study of schizophrenia and a second disorder, and the topic's classic study — because deciding what counts as a disorder, and how it is recognised, logically precedes explaining or treating one. In assessment-objective terms, you should be able to describe the purpose of classification, the two systems and the diagnostic process (AO1), apply this to scenarios such as a described patient presentation or a comparison of how the two manuals would handle a case (AO2), and evaluate the use of classification systems — their benefits and their limitations of reliability, validity and cultural neutrality (AO3).
Connects to…
Before examining the systems, it is worth asking what classification is for. Deciding that abnormality exists is only the first step; the four familiar approaches to defining abnormality — statistical infrequency (a characteristic is numerically rare), deviation from social norms (behaviour violates a society's rules), failure to function adequately (the person cannot cope with everyday demands, signalled by features such as personal distress, maladaptiveness and unpredictability), and deviation from ideal mental health (the person departs from criteria of psychological wellbeing) — each capture a real facet of abnormality but none is sufficient alone. Modern diagnostic systems do not rely on any single definition; instead they combine several — a characteristic symptom cluster, personal distress, functional impairment and a minimum duration — into an operationalised category. Classification is the machinery that turns the abstract problem of "what is abnormal?" into a workable clinical decision.
Classification serves several distinct purposes:
| Purpose | What it achieves |
|---|---|
| Common language | Gives clinicians, researchers and services a shared vocabulary, so that "major depressive disorder" means broadly the same thing in London, Lagos and Los Angeles. |
| Communication and continuity of care | Allows a patient's condition to be conveyed reliably between professionals and across time, supporting consistent treatment. |
| Access to treatment and support | Provides the formal basis on which patients receive medication, therapy, benefits and other services; in most systems a recognised diagnosis is the gateway to help. |
| Research and comparison | Enables studies conducted in different places to be pooled and compared, because they are (in principle) studying the same category; without agreed categories, aetiological and treatment research could not accumulate. |
| Prediction and prognosis | A good category should predict the likely course of the disorder and its response to treatment, guiding clinical decisions. |
Exam Tip: The purpose of classification is a frequent AO1 requirement and also supplies a ready-made evaluative counterweight. Whenever you criticise the reliability or validity of diagnosis, you can balance the criticism by noting that abandoning classification would sacrifice these benefits — so the rational response to the flaws is refinement, not abandonment. Balanced judgement of this kind is exactly what lifts an evaluation into the top band.
Two systems dominate clinical practice worldwide. Both are revised periodically as the evidence base develops, and both have converged considerably over time while retaining instructive differences.
| Manual | Full name | Produced by | Scope | Current edition |
|---|---|---|---|---|
| DSM-5 | Diagnostic and Statistical Manual of Mental Disorders | American Psychiatric Association (APA) | Mental disorders only | 5th edition (2013; text-revised as DSM-5-TR, 2022) |
| ICD-11 | International Classification of Diseases | World Health Organization (WHO) | All diseases, with a mental and behavioural disorders chapter | 11th revision (operational from 2022) |
The DSM-5 is the manual most used in the United States and in much research. It is concerned specifically with mental disorders, and for each disorder it provides an operationalised set of diagnostic criteria — a checklist specifying how many symptoms from a defined list must be present, for how long, and with what degree of functional impairment, before the diagnosis applies. A recurring DSM design principle is that a diagnosis normally requires not only the characteristic symptoms but also clinically significant distress or impairment in social, occupational or other important areas of functioning. This built-in distress-or-impairment requirement is important: it means the DSM does not simply pathologise unusual experiences but asks whether they are actually causing the person harm — an attempt to combine a symptom criterion with a functioning criterion. DSM-5 also moved towards a more dimensional and spectrum-based view in several areas (grouping related conditions along a continuum rather than treating them as wholly discrete), while remaining fundamentally a categorical system.
The ICD-11 is the global standard, used for health statistics and clinical coding in most countries and by the majority of the world's clinicians. Because it classifies all diseases, its mental and behavioural disorders form one chapter within a much larger system, which gives it a strongly international and public-health orientation. Compared with the DSM, the ICD has traditionally been written with somewhat more flexible, prototype-style descriptions (a clinical portrait the presentation should broadly match) rather than the DSM's rigid symptom counts, though ICD-11 narrowed this difference by adopting clearer essential-features descriptions. The ICD is freely available and designed for use across very different health systems, including those with limited specialist resources, which shapes its more descriptive style.
The two systems have deliberately moved closer together, so that most disorders are recognised in broadly compatible ways. Instructive differences nonetheless remain: the manuals can specify slightly different symptom thresholds or durations, organise categories differently, and historically differed in how much weight they placed on particular symptoms. These differences are not merely technical, because a person could in principle satisfy one manual's criteria but not the other's — a point that becomes central to the validity discussion in the next lesson.
graph TD
A["Presenting problem<br/>(distress / dysfunction)"] --> B{"Classification system"}
B -->|APA / much US research| C["DSM-5<br/>mental disorders only<br/>operationalised criteria counts"]
B -->|WHO / global standard| D["ICD-11<br/>all diseases<br/>essential-features descriptions"]
C --> E["Category + criteria matched"]
D --> E
E --> F["Diagnosis, communication,<br/>treatment access, research"]
style E fill:#2563eb,color:#fff
style F fill:#059669,color:#fff
Both manuals organise disorders into broad categories or groupings, each containing related conditions defined by a characteristic symptom picture. You are not required to memorise the manuals' full contents, but you should be able to name and briefly describe the principal categories so that you can locate the topic's specific disorders (schizophrenia, and the second disorder — depression or OCD) within the wider structure. Broadly recognised groupings include the following.
| Category (grouping) | Characteristic features | Example conditions |
|---|---|---|
| Schizophrenia and other psychotic disorders | Disturbance of thought, perception and sense of reality; positive and negative symptoms | Schizophrenia, delusional disorder |
| Mood (affective) disorders | Disturbance of mood — persistent low mood, or elevated/alternating mood | Major depressive disorder, bipolar disorder |
| Anxiety and fear-related disorders | Excessive, disproportionate fear or anxiety and avoidance | Phobias, generalised anxiety disorder, panic disorder |
| Obsessive-compulsive and related disorders | Intrusive obsessions and repetitive compulsions | Obsessive-compulsive disorder (OCD) |
| Trauma- and stressor-related disorders | Symptoms arising after exposure to a traumatic or stressful event | Post-traumatic stress disorder (PTSD) |
| Feeding and eating disorders | Disturbed eating and preoccupation with weight/shape | Anorexia nervosa, bulimia nervosa |
| Neurodevelopmental disorders | Onset in the developmental period, affecting personal, social or academic functioning | Autism spectrum disorder, ADHD |
The disorders studied in this topic sit within this structure: schizophrenia falls under psychotic disorders, depression under mood disorders, and OCD within the obsessive-compulsive grouping. A notable feature of the categorical approach is that individual symptoms recur across several categories — low mood appears in both mood disorders and the negative symptoms of schizophrenia, for instance — which foreshadows the symptom overlap problem examined in the next lesson.
Key Definition: A category (or grouping) of mental disorder is a set of related conditions defined by a shared characteristic symptom picture. The categorical model assumes that disorders are qualitatively distinct kinds, in contrast to a dimensional model, which treats them as differing in degree along continuous dimensions.
Classification defines the categories; diagnosis is the act of deciding that a particular individual belongs in one. In practice, reaching a diagnosis is a structured process rather than a single judgement, and it draws on several sources of information.
graph LR
A["Clinical interview<br/>& history"] --> B["Mental state<br/>observation"]
B --> C["Standardised tools<br/>(e.g. SCID, rating scales)"]
C --> D["Match to DSM-5 /<br/>ICD-11 criteria"]
D --> E["Differential diagnosis<br/>rule out alternatives"]
E --> F["Diagnosis"]
style F fill:#059669,color:#fff
The diagnostic process depends heavily on self-report and clinical judgement, and — for most mental disorders — there is currently no objective biological test (no blood marker or brain scan) that can confirm the diagnosis. Diagnosis therefore rests on the accurate description and interpretation of subjective experience, which is exactly why the reliability (will two clinicians agree?) and validity (does the diagnosis correspond to a real, distinct condition?) of the process are the dominant concerns of clinical psychology, taken up in detail in the next lesson. Standardised, structured interviews are the principal safeguard: by fixing the questions and scoring, they reduce the influence of individual clinician idiosyncrasy and are the main reason diagnostic consistency has improved over recent decades.
A useful way to see why standardisation matters is to contrast two styles of interview. In an unstructured interview the clinician follows their own line of questioning, which allows sensitivity to the individual but means two clinicians may gather different information and weight it differently — a recipe for disagreement. In a structured or semi-structured interview the questions, prompts and scoring rules are fixed in advance, so that two clinicians assessing the same person are far more likely to elicit the same information and reach the same decision. This is not a merely administrative point: it is the mechanism by which the field converted diagnosis from a largely idiosyncratic clinical art into a more reproducible procedure, and it is why any evaluation of the reliability of diagnosis must specify which diagnostic method is meant. It is also worth noting that criteria-matching is rarely mechanical: real presentations are messy, symptoms are reported with varying clarity, and clinicians must exercise judgement about whether a described experience genuinely meets a criterion — so even a well-operationalised manual leaves an irreducible element of interpretation, which is one source of the residual inconsistency examined next.
Exam Tip: If asked how a diagnosis is made, do not simply write "the doctor decides." Name the components — clinical interview and history, observation/mental state examination, standardised tools such as structured clinical interviews, criteria-matching against DSM-5/ICD-11, and differential diagnosis — and note that the process leans on self-report because no biological test exists. That final point is the natural bridge into the reliability-and-validity evaluation.
Classification and diagnosis deliver substantial, tangible benefits that any criticism must be weighed against. Standardised systems give clinicians and researchers a shared language, allow findings from different places and times to be pooled and compared, and provide the formal route by which patients access treatment and support. This matters because, without agreed categories, aetiological and treatment research could not accumulate and services would have no consistent basis for allocating help. The implication is that the flaws examined below are reasons to refine and culturally calibrate the systems, not to abandon classification altogether — a balanced position an examiner rewards.
Operationalised, standardised criteria have markedly improved consistency, which is a real strength of the modern systems. By specifying symptom counts, durations and impairment requirements, and by supplying structured interviews such as the SCID, DSM-5 and ICD-11 reduce the reliance on unaided clinical judgement that made earlier diagnosis so variable. The significance is that reliability — the extent to which clinicians agree — is much higher with structured tools than with unstructured interviews. The implication is that the quality of a diagnosis depends heavily on the instrument and training used, not only on the category itself, so "diagnosis is unreliable" is too crude a verdict for contemporary practice.
Because there is no biological test for most disorders, diagnosis remains dependent on self-report and interpretation, which limits its objectivity. Unlike the diagnosis of a physical illness confirmed by a laboratory marker, a mental-disorder diagnosis rests on the description and reading of subjective experience. This matters because it leaves room for inconsistency between clinicians and for the influence of expectation and bias. The implication is that the systems function as carefully standardised judgement aids rather than objective measuring devices, and their outputs should be treated as clinically useful but provisional — a theme developed fully in the reliability-and-validity lesson.
The categorical model can be criticised as imposing sharp boundaries on what may in reality be continuous. DSM-5 and ICD-11 largely treat disorders as discrete categories a person either does or does not have, yet symptoms such as low mood or anxiety are shared across categories and are found, in milder form, in the general population. The significance is that categorical thresholds (why five symptoms and not four?) are partly matters of clinical convention rather than natural dividing lines. The implication is that dimensional models — which DSM-5 partly incorporates through its spectrum framing — may in some respects capture the reality better, though categories remain far easier to use for clear treatment decisions, so the debate is genuinely unresolved.
The DSM and ICD were developed largely in Western contexts, raising a real risk of cultural bias when their criteria are applied universally. Norms of behaviour and emotional expression, and the very forms that distress takes, vary across cultures; experiences that are normative in one culture may be read as symptomatic through the lens of a manual written elsewhere. This matters because it can generate misdiagnosis and disparities in care — an imposed etic, in which one culture's standards are treated as a universal yardstick. The implication is that classification must be applied with explicit attention to cultural context, and that "culture-free" diagnosis is an aspiration rather than an achieved fact; both systems have made efforts (for example, cultural formulation guidance) to address this, with partial success.
Diagnosis carries a risk of stigma and labelling, a socially sensitive limitation of the enterprise. A diagnostic label can shape how others — and the person themselves — interpret subsequent behaviour, and can attract social disadvantage. This matters because the benefits of accessing treatment must be balanced against the potential harms of the label, particularly where a diagnosis is uncertain or contested. The implication is that classification should be used judiciously and communicated carefully, and it is one reason clinicians distinguish describing a person's difficulties from reducing the person to a category — an ethical dimension that a strong evaluation can make explicit.
The two systems' residual differences are themselves informative about the limits of classification. That DSM-5 and ICD-11 can classify the same presentation slightly differently shows that the categories are, in part, agreed conventions rather than discovered natural kinds. This matters because it means the choice of manual can affect whether, and with what, a person is diagnosed. The implication is not that classification is worthless — the two systems agree far more than they differ — but that a diagnosis is best understood as the output of a particular, humanly constructed system, which is precisely why validity (does the category correspond to something real?) is the deeper question the next lesson pursues.
Classification systems are revisable, which is both a scientific strength and a reminder of their constructed nature. Both manuals are periodically updated as evidence accumulates and as society changes, so categories are added, split, merged or removed across editions. On the one hand this responsiveness is a genuine strength — it means the systems are not frozen and can correct past errors, the removal of homosexuality as a disorder in 1973 being the clearest example of a harmful category being discarded. On the other hand, the very fact that the list of recognised disorders shifts over time undercuts any assumption that the categories carve nature at fixed joints; a condition recognised in one edition may be reconceptualised in the next. The implication is that a diagnosis reflects the best current agreed classification rather than a timeless fact, which is a further reason to treat diagnostic labels with appropriate humility while still using them for the practical benefits they confer.
Specimen question modelled on the Edexcel 9PS0 paper format.
Evaluate the use of classification systems (DSM-5 and ICD-11) in the diagnosis of mental disorders. (16 marks)
This 16-mark extended-response question is marked as roughly 6 marks AO1 (accurate, detailed description of the purpose of classification, the two systems and their features, and how a diagnosis is made) and 10 marks AO3 (evaluation — the benefits of classification, the improvement from operationalised/standardised criteria, the dependence on self-report given the absence of a biological test, the categorical-versus-dimensional debate, cultural bias, and stigma). Application (AO2) marks would apply only if a scenario stem were provided. The top band requires sustained, integrated evaluation building to a reasoned judgement rather than a list of isolated points.
Classification means putting mental disorders into categories, and diagnosis is deciding which one a person has. The two main systems are the DSM-5, made by the American Psychiatric Association, and the ICD-11, made by the World Health Organization. The DSM only covers mental disorders and the ICD covers all diseases. To make a diagnosis a clinician interviews the patient and checks their symptoms against the criteria in the manual to see if they match a category like depression or schizophrenia.
One strength is that classification gives everyone a shared language so they know what a disorder means. Another strength is that it lets people get treatment. One weakness is that there is no blood test, so it relies on what the patient says. Another weakness is that the manuals were made in Western countries, so they might not fit other cultures.
Examiner-style commentary: To reach the next band this answer must explain why each point matters rather than naming it — for example, why the lack of a biological test threatens objectivity (diagnosis depends on interpreting subjective self-report, leaving room for inconsistency). The description is broadly accurate but thin: the purpose of classification is barely developed, the diagnostic process omits standardised tools and differential diagnosis, and the categorical-versus-dimensional issue is absent. With the 10:6 weighting towards AO3, the undeveloped evaluation caps the mark.
Classification organises mental disorders into agreed categories defined by symptom clusters, and diagnosis is the judgement that an individual meets a category's criteria. Its purpose is to provide a shared language, support communication and continuity of care, give access to treatment, and enable comparable research. Two systems dominate: the DSM-5 (APA), which covers mental disorders using operationalised criteria (specified symptom counts, durations and a distress/impairment requirement), and the ICD-11 (WHO), the global standard, which covers all diseases and uses more descriptive essential-features. A diagnosis is made through clinical interview and history, a mental state examination, standardised tools such as the SCID, matching to the manual's criteria, and differential diagnosis to rule out alternatives.
A clear strength is that operationalised, standardised criteria and structured interviews have improved consistency, so diagnosis is far more reliable than unstructured clinical judgement. However, because there is no biological test for most disorders, diagnosis still depends on self-report, which limits objectivity. There is also a cultural-bias problem: the manuals were developed in Western contexts, so applying them universally risks an imposed etic that pathologises culturally normal behaviour. A further issue is stigma, since a label can disadvantage the person. These points suggest classification is useful but imperfect.
Examiner-style commentary: The move into the top band is to develop each point through point–evidence–explanation–implication and to weigh the benefits explicitly against the flaws in a reasoned conclusion — for instance, arguing that the flaws justify refinement rather than abandonment. The description and process are accurate and the evaluation points relevant, but the categorical-versus-dimensional debate is missing and the points are stated in parallel rather than built into an integrated judgement.
Classification organises mental disorders into agreed categories, each defined by a characteristic symptom cluster, while diagnosis is the clinical decision that a particular individual meets a category's criteria; the two must be distinguished. Classification exists to provide a shared professional language, to support communication and continuity of care, to give patients formal access to treatment, and to enable research from different settings to be pooled — benefits that frame the whole evaluation. Two systems dominate: the DSM-5 (APA), concerned only with mental disorders and built on operationalised criteria (specified symptom counts, durations and a clinically-significant-distress-or-impairment requirement), and the ICD-11 (WHO), the international standard covering all diseases with a more descriptive, essential-features style. The two have deliberately converged yet still differ enough that the same presentation can occasionally be classified differently. A diagnosis is reached not by a single judgement but through a structured process — clinical interview and history, a mental state examination, standardised tools such as the SCID, criteria-matching against the manual, and differential diagnosis to exclude alternatives — and, crucially, it rests on self-report because for most disorders no biological test exists.
That final fact anchors the evaluation. The strongest defence of classification is its benefit: shared language, comparable research and treatment access are indispensable, so criticisms are reasons to refine rather than abandon the systems. Against this, the absence of a biological test means diagnosis depends on interpreting subjective experience, which limits objectivity and leaves room for inconsistency — though the operationalisation of criteria and the use of structured interviews have substantially improved reliability, so contemporary diagnosis is far more consistent than the unstructured clinical judgement of earlier decades. A deeper problem is validity: the categorical model imposes sharp thresholds on symptoms that are often shared across categories and continuous in the population, so dimensional models (partly reflected in DSM-5's spectrum framing) may capture the reality better, even if categories remain more usable for treatment. Cultural bias compounds this, since manuals developed in Western contexts risk an imposed etic when applied universally, generating misdiagnosis and disparities in care; both systems now offer cultural-formulation guidance, with partial success. Finally, diagnosis carries a socially sensitive risk of stigma, since a label shapes how behaviour is subsequently interpreted. Weighing these together, classification is a genuinely valuable but humanly constructed tool: its reliability has improved markedly through standardisation, but its validity and cultural neutrality remain contested, so a diagnosis is best treated as clinically useful yet provisional — which is precisely why the reliability and validity of diagnosis is the central question clinical psychology goes on to examine.
Examiner-style commentary: This answer is in the top band. It is distinguished by sustained, integrated evaluation: it uses the benefit of classification as an explicit counterweight, chains the "no biological test" point into both reliability and validity, develops the categorical-versus-dimensional and cultural-bias issues through point–evidence–explanation–implication, and reaches a reasoned judgement that diagnosis is useful but provisional. The only refinement would be to note that the two systems' residual differences themselves reveal the categories as agreed conventions. The discriminator is the connectedness of the AO3 reasoning and the quality of the final judgement, not the quantity of description.
A rewarding line of stretch reading is the US National Institute of Mental Health's Research Domain Criteria (RDoC) initiative. RDoC responds directly to the validity limitations of categorical diagnosis by proposing that research should study dimensions of functioning — such as cognition, arousal and social processes — that cut across the traditional DSM/ICD categories, rather than treating a category like "schizophrenia" as a single thing to be explained. Engaging with RDoC gives you a sophisticated answer to the boundary problem: it shows the field is actively debating whether the categorical model should be supplemented, or even partly replaced, by a dimensional one — the very debate the specification asks you to evaluate.
A second productive direction is the concept of reification — the error of treating an abstract category as though it were a concrete, discovered "thing". Because a diagnosis such as "major depressive disorder" names a cluster of criteria agreed by a committee, it is tempting but mistaken to assume it corresponds to a single underlying entity in nature. Reading around the history of how categories have been added to, split, merged or removed from successive editions of the DSM (the removal of homosexuality in 1973 being the best-known example) makes vivid that classification is a revisable human construction responsive to evidence and social change — an insight that reliably separates a sophisticated evaluation from a naive one.
This content is aligned with the Edexcel A-Level Psychology (9PS0) specification.