Distress and impairment in DSM-style diagnosis

Psychiatric classification often requires more than the presence of symptoms. In many DSM categories, symptoms must also cause clinically significant distress or impairment in social, occupational or other important areas of functioning. This requirement is clinically intuitive because it helps distinguish disorder from harmless variation and ties diagnosis to suffering, disability or need for care.

Nevertheless, this creates a conceptual problem. In many DSM categories, distress and impairment do more than indicate clinical significance: they also help determine whether the diagnosis can be made. This is different from much of somatic medicine, where diagnosis, clinical significance and treatment indication are more clearly separated. A person may have hypertension or an asymptomatic tumour without suffering, functional impairment or need for treatment. The diagnosis still stands even if its immediate clinical implications are limited. Psychiatric diagnoses likewise describe patterns of emotion, cognition, perception, behaviour, agency and self-experience. This does not mean that distress or impairment must always be part of the diagnostic criteria. DSM-5-TR partly recognises this distinction, and its general definition of mental disorder does not treat distress or disability as necessary features of disorder, but states that mental disorders are usually associated with significant distress or disability.¹ However, many specific DSM-5-TR criteria do require clinically significant distress or impairment. Thus, although neither distress nor disability is required by the general definition, they can still determine whether a particular diagnosis is made.

Our claim is not that psychiatric classification should be value-free. Psychiatric diagnoses inevitably involve judgements about dysfunction, harm, significance and context.^{Reference Bolton2–Reference Sadler4} Nor do we propose that DSM should become a catalogue of psychological differences: a psychiatric manual still needs criteria in regard to deciding what belongs within psychiatry. The question is where these evaluative judgements should be located. This parallels Ereshefsky’s proposal to distinguish descriptions of states from normative judgements about their significance.^{Reference Ereshefsky5} Applied to psychiatric diagnosis, the point is not that evaluation can be removed: rather, category inclusion, individual classification, clinical significance and decisions about response should be kept separate. Our argument is not that distress or impairment should be removed from all diagnostic criteria but, when they determine whether a person meets the criteria for a particular DSM category, there should be a clear reason for requiring them in that category rather than treating them as a general rule. Adjustment disorder may be one such case, because distress or impairment can help distinguish a diagnosable response to a stressor from an expected response to adversity.

Scope and over-expansion

Dysfunction, harm and distress help determine which conditions belong within psychiatric classification, but they need not be required for every individual diagnosis. A recognisable or statistically distinct pattern is not enough to justify inclusion in a psychiatric manual. The pattern must also have a non-accidental relationship to clinically relevant features such as dysfunction, disability, loss of control, developmental disruption or a characteristic psychiatric or neurodevelopmental course. That relationship can hold at the level of the category, even when a particular individual is not currently distressed or impaired.

Distress or disadvantage arising solely from social disapproval, prejudice or conflict is insufficient to justify inclusion in psychiatric classification. This means that homosexuality, ordinary personality variation, recreational drug use and harmless preferences should not count as psychiatric classes merely because some individuals experience distress or disadvantage. On the other hand, Tourette disorder, autism and hypomania may remain within psychiatric classification without requiring current distress or impairment in every individual. An individual DSM diagnosis means that the person meets the criteria for a psychiatric category whose inclusion in the manual is justified at the category level. A diagnosis alone does not establish the degree of current distress or impairment, treatment need, disability, diminished responsibility or justification for coercion. Assuming that it does so risks reifying the diagnosis.^{Reference Hyman6}

Our proposal is about what an individual DSM diagnosis does and does not imply, without reducing DSM to a catalogue of psychological differences. The familiar phrase ‘clinically significant distress or impairment’ illustrates this issue. Because the criterion is disjunctive, neither subjective suffering nor functional impairment is required on its own: one may be present when the other is absent. The criterion therefore seems to capture clinical significance rather than the phenomenon itself. A DSM diagnosis should not, by itself, be taken to mean that a person is distressed, impaired, in need of treatment or lacking autonomy. It identifies a pattern, whereas its significance and what response is warranted for the individual would require separate judgement.

Precedents and limits

Tourette disorder provides a useful precedent in this context. In DSM-IV, tic disorders required clinically significant distress or impairment. However, DSM-IV-TR removed this requirement. This did not make tics clinically irrelevant, but recognised that a person can meet criteria for a tic disorder without experiencing current distress or impairment – even though tics are associated with both at group level. A child may, for example, meet symptom criteria for Tourette disorder with little distress or impairment, whereas social or academic impairment may arise from comorbid conditions rather than from the tics themselves.^{Reference Roessner, Hoekstra and Rothenberger7} Tourette disorder therefore shows that individual classification does not always need to depend on current distress or impairment. Similar questions arise in hypomania and autism, and with phenomena such as voice-hearing or anosognosia, where distress may be absent. In such cases, requiring distress or impairment for diagnosis can blur the distinction between identifying the phenomenon and judging its significance for the individual.

Distress and impairment have an important protective role. Psychiatric classification concerns experience, agency, behaviour, identity and social functioning, all of which involve value judgements. Requiring distress or impairment can help prevent ordinary social variation from being pathologised and keep diagnosis linked to suffering, disability, risk or need. However, that does not mean they must be required for every individual diagnosis.

The same distinction applies to harm more generally. DSM criteria often specify harm in terms of current distress or impairment in the individual. Our concern is not that DSM is vague about this, but that clearly specified distress or impairment is used to decide whether a diagnosis is made, when they may be better considered in a separate assessment of clinical significance. Wakefield’s harmful dysfunction analysis remains relevant because it understands disorder as involving both dysfunction and harm.^{Reference Wakefield8} It does matter whether harm is a requirement for the broader category or for each individual diagnosis. Either way, harm remains part of deciding what counts as psychiatric disorder, so classification is not purely descriptive. Our point is narrower: current distress or impairment should not automatically be required for individual diagnosis when the reasons for including a category in psychiatric nosology can be separated from the question of how significant that condition is for a particular person.

HiTOP as precedent and caveat

The Hierarchical Taxonomy of Psychopathology (HiTOP) offers another useful example. Its core structural model organises psychopathological phenomena by observed covariation while assessing impairment and disability separately, with dysfunction, harm and clinical significance remaining outside the core model.^{Reference Aftab, Wright, Rodriguez-Seijas, Hankin, Clark and Forbes9} Recent philosophical discussions of HiTOP point to a problem with this separation. If HiTOP does not itself include dysfunction, harm, disability or clinical significance, it becomes less clear what makes its dimensions psychopathological rather than forms of mental variation more generally.^{Reference Aftab, Wright, Rodriguez-Seijas, Hankin, Clark and Forbes9} HiTOP therefore shows both the value and the limits of separating classification from evaluation. DSM-style categories should still be restricted to conditions of medical relevance, rather than arbitrary psychological differences.

Classification, evaluation and response

Psychiatric diagnosis is often used for several purposes simultaneously: to identify a condition, distinguish disorder from variation, indicate clinical significance, justify professional attention and sometimes support decisions about treatment, resources, disability status or coercion. These questions are of course related in practice, but they are not the same. Identifying a psychiatric or neurodevelopmental pattern does not by itself tell us how significant it is for a particular person, or what should be done. Those questions should not be built into the diagnostic criteria by requiring distress or impairment.

We therefore distinguish four questions: category inclusion (why a condition belongs in a psychiatric manual), individual classification (whether a particular person meets criteria for that condition), evaluative significance (what the diagnosis means for that person and in that context) and clinical response (what, if anything, should be done). Distress, impairment, suffering and harm remain central to these questions, and can help determine whether a condition is burdensome, disabling, risky, urgent or in need of care. However, they are not always needed to determine whether a person meets diagnostic criteria. Evaluation therefore remains necessary, but this is a separate process and should not be built into every act of classification.

Clinical implications

First, a psychiatric classification should not automatically be taken to mean that a person is distressed, impaired, defective or in need of treatment. This matters for stigma, including self-stigma, and for how diagnoses are communicated. A diagnosis may identify a psychiatrically relevant pattern without, by itself, determining what that diagnosis means for the individual.

Second, this distinction also fits with neurodiversity. A neurodevelopmental classification need not imply that a person is intrinsically ‘defective’, nor that intervention should aim at normalisation. It may identify a pattern of cognition, perception, communication or development while leaving open whether the relevant difficulty lies in the individual, the environment or the interaction between them. This allows psychiatry to recognise differences without denying disability or support needs.

Third, diagnosis alone cannot justify treatment, disability status or coercion. Decisions about clinical response also require judgements about risk, capacity, proportionality, patient goals, expected benefit, legal criteria, available alternatives and context. A diagnosis may inform these decisions, but it cannot determine them on its own.

The DSM should remain a manual of psychiatric conditions, and not a catalogue of psychological differences. Distress and impairment should not be removed from diagnostic criteria altogether. However, when they determine whether an individual meets the criteria for a particular DSM category there should be a clear reason for including them in that category rather than treating them as a default requirement.

A category can be clinically significant enough to belong in DSM without requiring current distress or impairment in every individual who meets its criteria. Psychiatry therefore needs to distinguish identifying a condition, judging its significance for the individual and deciding what response is appropriate. Distress and impairment remain central to psychiatric assessment, but they need not play the same role at each of these stages.